TW201211508A - Navigation devices and methods carried out thereon - Google Patents

Abstract

This invention concerns a method of determining a route using map data comprising a plurality of navigable paths, the map data divided into a plurality of regions. The method comprises using at least one processing apparatus to: receive an origin and a destination on the map data and a travel time, determine a route from the origin to the destination using the map data and minimum cost data that identifies minimum cost paths between regions of the map data. The minimum cost data identifies more than one minimum cost path between a pair of the regions if different minimum cost paths exist between the pair of regions at different times and determining a route comprises identifying from the minimum cost paths for the pair of regions comprising the origin and destination. the minimum cost path having a lowest cost at the travel time.

Description

201211508 VI. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates to a computerized method and related system for determining a route using map data. In particular (but not exclusively), the method can be used in navigation devices to facilitate route planning. [Prior Art] A route planning algorithm (such as the Dijkstra method, the A* method, or the M〇ore/Pape method) is known. However, these methods can be a bit slower. An example of how this route planning can be increased is shown in US Pat. No. 6,636, the disclosure of which is incorporated herein by reference. Us 6 The novice teaches pre-processing map data by dividing the map into zones and determining the tree of the shortest path (minimum cost path) to each zone of the map. The data processed by this 婉T is stored in the memory, and the preprocessed data is calculated during the calculation of the route.于内特二: 'The processing required to be used as a line in the area where the starting point is located in an area and the destination is located in the other part of the route" ^ This side reduces the determined way of finding the way, Because it reduces the number of paths that are explored when using the bite to get out of the line... The way to make sure that the line is understood is that "the path with the smallest distance is meant, but the factor considered in the formula. ^ Path ^ Cost will be as specified in the cost and fuel use. It can be considered very fast when the shortest path between the zones is kept static. This method is extremely successful. 150832.doc 201211508 However The change in traffic conditions over time can be changed to the path of the take-up path between the zones, such that the pre-processed tree no longer identifies the true shortest path between the zones. This can result in determining a for a given criterion A route that is not an optimal route. [Invention] - According to a first aspect of the present invention, there is provided a method of determining a route using map data including a plurality of navigable paths, the map data being divided into plural numbers Zone, the method comprising using at least one processing device to: receive the map > a starting point of a material and a destination and a journey time, using the map data and identifying a minimum cost path between the regions of the map data The cost data determines a route from the starting point to the destination, characterized in that, in the case where there are different minimum cost paths between a pair of zones in the zones at different times, the minimum cost data identifies the pair of zones One or more least cost paths between, and determining that a route includes the least cost path having a lowest cost at the journey time q from the least cost paths of the pair of zones including the origin and the destination. Where different minimum cost paths exist between a pair of zones in a zone at different times, the identification is increased at the time of the journey relative to the use of data that does not identify more than one least cost path between the zones The possibility of a minimum cost path between zones. However, using the minimum cost data enables the determination to be made from scratch The method of small cost path determines the route more quickly. For example, predictable changes in traffic conditions (such as the peak time of the rule) can change the minimum cost path between zones. Pre-calculated between zones at different times The minimum cost path may allow for periodic changes to be considered from 150832.doc 201211508. Determining the route may include identifying a minimum cost path between a pair of zones based on the minimum cost profile independent of time, and for one or both derived from the journey time Performing a cost analysis on the identified minimum cost paths for a plurality of relevant times to determine a minimum cost path at the journey time. This may allow the minimum cost data to be of a much smaller size because of the information about the least cost path There is no need to cross-reference the time when the paths are the least cost path. 士士: The route planner of the navigation device can determine which of the path subsets identified by the least cost data is the least cost path at the relevant time. This configuration does transfer a small extra s burden to the route planner, but in return' substantial savings in the size of the minimum cost data. In the alternative configuration, the minimum cost profile identifies, for each least cost path, a reference time for the path to the minimum cost path, and confirms that the route includes selecting based on the minimum cost. The minimum cost path for the reference time derived from the -Hour journey time or at multiple relevant times. This configuration has the advantage of 'because the two-way route will have to be found to find the least cost route.相关 The relevant time can be between the nodes connected by the least cost path. The nodes are the navigable paths on the map, ', 曰, Α, and/or the 戥忐-r end of the navigation path. Therefore, via the minimum cost road based on the route: ... 《 "Not all needles: Inter I will be for different time (depending on the arrival of the road 2, for example, if the cost is predicted - hour The second timeline will identify/select the minimum cost path for the hour of the hour based on the node's routing along the route. Minimum cost information and route determination _ knife selection Ordering 'because only the part of the route can be selected for the minimum cost data. 疋The journey time can be from the departure time of the starting point, to or to the time of the specific location (such as the node) along the route. To achieve the desired level of accuracy, the resolution of the minimum cost data should be of the order of seconds, preferably with a tenths of a reading, and optimally, of the order of percent to the second. It has been found to be coarser. At the time of rating ^ ^ The minimum cost data of the abundance leads to the accuracy level of the unreachable text. In addition, it can be processed within the dry circumference of Λ A f- PR Η ^ ^ ^ heart or one hundredth of a second. For the processing of the object-seconds, the simplification meter needs to avoid a finer level than one-hundredth of a second, because this can require a large number of bits of the car to store the data. It should be understood that the resolution can be one hundredth of a second. Stored as a 32-bit string, and higher resolution may require Μ 1 string 〇〇 " path selection algorithm can be the base (4) this formula assigns cost to different path' path selection algorithm. Map data can include identification The speed profile of the expected speed on the road at different times. The complex data of the map data can be divided into a plurality of navigable segments, and the velocity profile data can be identified - the velocity profile of each V navigation segment The cost of the path may be determined based, at least in part, on the speed profile of the navigation I that constitutes a path. In an embodiment, the method includes determining a speed for the correlation time based on the speed profile data, wherein the route is considered The determined speed of each of these is used to calculate the cost. 150832.doc 201211508 In an alternate embodiment, the method includes multiple profiles based on the speed profile of the temple navigation segment (not a single; the course of the route. Change the value of the taxi \i value) to determine the cost profile. The cost formula receiving profile (ie, function) outputs an overview of the cost of the route at different times. Entering and determining changes in the cost of the route over time. The method may include determining a plurality of routes for different journey times; and, for example, the plural: how different journey times affect The route and / or use,,, L - the predicted time of the route between the land used for comparison. For example, the starting point and the target period of the main road may not be the minimum cost between the areas: diameter, :::: : into a minimum cost path. This can result in different routes between the starting point and the destination for different travels::: and several routes between the travels can allow the user to make a decision based on the determined time of the return. The comparison of the route of the 疋 就 就 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 旅 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定 确定Determining the optimal journey time between the starting point and the destination for the defined criteria, and in response, the cost profile of the several routes (per-route targeted - different journeys and showing the results to the user. The result display can be The display of the journey time of the best route. Therefore, the method may include determining which of the plurality of routes for the defined criteria based on the cost profile (eg, 'screening the results by comparing the values of the cost profiles') In this way, the processing device determines the best route for the user 150832.doc 201211508 and does not need to be analyzed by the user. Alternatively, the display of the results can be a display of the cost profile (such as a 'map). In this way, the user can select a route based on an evaluation of the defined criteria and selecting the time that best suits him/her. For example, at the starting point and the destination The fastest time to travel between the grounds may be at night, but the user may choose the next best route because it is at a more convenient journey time (sometime during the day). By showing the multiple routes For all routes or to Ο Ο less - the results of the above routes, the user can make an informed decision on the travel time and how this affects the route. In one implementation, the profile request includes the time period, and the search for the stock line is limited to (4). For example, the user m journey day (four) should be between two % (for example, between 6 am and 9 am), or on a certain day (for example, in the week -), on a certain date ( For example, July i), a week or = month. Or, the processing device can automatically search for the journey time to fall within the window of the journey time specified by the person (for example, the user journey time coffee: route) In the case where it is found that the route with different journey times has a lower cost than the route of the determined journey time, the user is provided with 4 alternative travel time. In the embodiment, the method includes self-contained travel time via the user interface. Make a display of the image of the route determined by the cough. 2 = The user can change the journey time while checking the route 2 of the shirt. For example, in response to the journey, the text can be changed. The specified journey time determines the new route, and the image of the new route updates the display. Alternatively, it can be (for example: line = 150832.doc 201211508 most) cost _ set the minimum cost route for other time. It can be "instant" (for example, In the number of milliseconds ^ update of the display, for example, less than 1000 亳 seconds and 轸 '' seconds. Can be used to identify between the zones ^ less than 500 * 采 丨 丨 h , , , , , , , , , , The quickest recalculation time for the lowest cost and poor material makes the entire camp renewed, the old route fades out and the new route is lighter and the daylight is refreshed. It should be understood that the new route may be the same as or different from the route, depending on whether there is a change in the least cost path. In an embodiment, the method can include causing the display of the travel time 4' and responding to the maker interacting with the slider to determine a new route and update the image with the new route (4). The method can include determining and displaying a new route such that the controller has an imperceptible delay between when the display of the new route is changed (4). For example, the delay between the user moving the slider and displaying the new route is less than ... and preferably less than 0.5 seconds. According to a second aspect of the present invention, a data carrier storing instructions is provided. The instructions, when executed by a processing device, cause the processing device to perform the method according to the first aspect of the present invention. According to a third aspect of the present invention, a data carrier is provided, which stores map data including a plurality of navigable paths, the map data is divided into a plurality of regions, and the map data is discriminated at different times. The minimum cost of the minimum cost path between the zones. This batting carrier is advantageous 'because it can be used for the determination of the route according to the first aspect of the invention 150832.doc 10·201211508. According to a fourth aspect of the present invention, a computer device including a memory and a processing device is provided, the memory storing: including a plurality of navigable paths, data, the map data is divided into a plurality of regions, and the map is identified The minimum cost data of the least cost path between the zones of the data; the location: the device is configured to receive a point on the map data and a destination and a journey time, and use the map data and the minimum cost Data to determine a route from the starting point to the destination point, wherein the minimum cost data identifies the pair if, at different times, Pa1 has different minimum cost paths between the pairs in the zones More than one least cost path for the zone, and the determination-route includes identifying a minimum cost path for the route between the zone containing the origin and the destination based on the minimum cost profile when considering the journey time. According to a fifth aspect of the present invention, there is provided a method of determining a route using map data, minimum cost data, and a speed profile, the map material comprising a plurality of navigable segments in which Q represents a segment of the mappable path of the map data; The minimum cost data identifies a minimum cost path between the zones of the map data, wherein different minimum costs exist between a pair of zones in the zones at different times; in the case of a route, the minimum cost profile identifies between the pair of zones One of the following: The speed profile of the 5th private navigation segment identifies how the speeds on the navigable segments change over time, the method comprising using at least one processing device. Searching for one or more routes from the origin to the destination on the map data, the search comprising determining whether one or more of the set of navigable segments connected to a node is minimized to 150832. Doc • 11 · 201211508 This greedy material is identified as part of the minimum cost path that contains the starting point and one of the areas of the destination' and the navigable segments in the group Where one or more navigable slaves are identified as part of a least cost path, only one or more navigable slaves identified as part of a least cost path are explored from the set and according to the The velocity profiles of the explored navigable segments determine a cost profile for one of the one or more routes along with the past. The HA cost profile allows a user and/or navigation device to determine how changing the travel time will change the cost of the route. In this way, the journey time can be selected based on the profile. According to a sixth aspect of the invention, there is provided a data stored with instructions =' such instructions, when executed by a processing device, cause the processing device to perform the method according to the fourth aspect of the invention. According to a seventh aspect of the present invention, there is provided a computer device comprising a memory and a processing device, the memory storing map data, the map data comprising a segment of a navigable path indicating the map data, and the like The overview of the speed of the navigable section, the overview of the catching degree of Cicada, Hangfeng and Hai, and how the speed can change with time; the process is set to search for the map. One or two routes from the starting point of the city to the destination: the search includes the navigation segment n At WL· Μ ^ 确定 that determines the connection to the node, and the cost is identified as one of the zones containing the occupied destination. ^ ^ ^ -Γ ^#, and in the case of the , or , , , , or part of the navigable cost path in the segment, the part of the path that takes the small cost The one or more are the smallest guide slaves, and determine the cost of one of the one or more routes over time based on the speed profiles of the navigable segments of the detected 150832.doc -12·201211508 Overview. According to an eighth aspect of the present invention, a navigation apparatus is provided, comprising: a bit system, a display, a user interface, and a computer device according to a fourth sorrow of the present invention, wherein the processor is configured to Deriving the starting point and destination on the user's face and displaying the determined result on the display. The processor can be based on the input on the user interface, «as indicated by - internal clock Determining the journey time by the current time, or determining a plurality of routes having different journey times. According to the ninth aspect of the present invention, 'providing a method for determining and using a map data using a map material' a plurality of navigable segments of the mapable data path and a speed profile of the navigable segments, the method comprising determining a route along the navigable segments in response to input at the user interface Determining the expected speed of one of the navigable segments along the route based on the associated speed profiles, i showing the determined route and the guidance for the route An indication of the expected speed on at least some of the navigable segments of the segment. According to a tenth aspect of the present invention, a data stored with instructions is provided that the instructions are executed by the processor According to a tenth aspect of the present invention, there is provided a computer device comprising: a display, a user interface, a memory and a processing device. The memory is stored Map data 'The map material contains a plurality of navigable segments representing the navigable path of the map data and the speed of the navigable segments is 150832.doc -13- 201211508, the speed profiles are represented at different time points In the case of predicting the speed of travel, the device is configured to determine a route along the navigable segments in response to input at the user interface, determining the edge based on the associated velocity profiles The expected speed of one of the navigable segments of the route, and causing the display to display the determined route and the pre-position on at least some of the navigable segments of the route Speed-indication. The processing device can cause the display to display a journey time, which changes the journey time by interacting with the user interface, wherein the return time is changed to the time of the journey, 兮卢. ...... The processing benefit determines the update of the navigable segment of the route for the new journey time ^ ^ v 4t pre-J speed, and updates the route to not indicate the expected speed of the update. The journey time can be displayed by the display First, the ((7)# means no, the user can move 哕, and the tempo to adjust the journey time. The second shoulder/month expected speed indication can be copable~ ^ ^ The horse has the color code of the flight segment. For example +, if the expected speed is higher than the first - cow case and the δ system is within 20%), the flight slave speed limit) can color the navigable segment to the general public "I, green, color is brother The color (for example, the top period speed is lower than the second threshold segment is colored to the second color (for example, red, " the limit value can be the same value. s product limit and second pro in this way 'user can determine the Qingsai point can (as appropriate) change 路 + road', where on the spring, D father more journey time on this Kui [implementation Mode] The effect of the U-plug point. The following is a detailed description of the ginger lost by way of example only. References to Embodiments of the Invention With reference to the accompanying drawings, the following reference numerals will be used to identify like parts. Throughout the description of the various embodiments, reference is made to the flowchart of Figure 22 in which the reference numerals are in the 19 lamp series. Position system (GPS), a global positioning system navigation system, which can be used to determine infinite speed, time, and (in some cases, Figure 1 schematically shows the global position (GPS) as a continuous position of a number of users based on satellite_radio ,

B) Direction information. The past, known as NAVSTAR2Gps, incorporates a number of satellites that operate around the Earth in extremely precise ways. Based on these precise orbital GPS hygiene, the location can be relayed to the receiving unit as a Gps data. However, it should be understood that a global positioning system such as gl〇SNass, European Galileo positioning system, c〇MpASS positioning system or deleted (India regional navigation satellite system) may be used, when equipped to receive GPS The device of the data begins to scan the radio frequency to discover the GPS satellite signal 'implementation of the GPS system. After receiving a radio signal from a GPS satellite, the device determines the exact location of the satellite via one of a plurality of different conventional methods. In most cases, the device will continue to scan to find the signal until it has acquired at least three different satellite signals (note that other triangulation techniques can be used to determine the position by only two letters, although this is very example). In the case of performing geometric triangulation, the receiver uses three known locations to determine its own two-dimensional position relative to the satellites. This greening can be done in a known manner. In addition, obtaining a fourth satellite signal allows the receiving device to know the geometric calculation to calculate its three-dimensional position. Unlimited number of users = Instantly update location and speed data. 15〇832.d〇c -15- 201211508 As shown in FIG. 1, the GPS system 1 includes a plurality of satellites 102 that operate around the Earth 1 〇4. The GPS receiver 1 〇 6 receives GPS data as a spread spectrum GPS satellite data signal 1 〇 8 from a plurality of satellites of the plurality of satellites 102. The spread spectrum data signals 1 〇 8 are continuously transmitted from each satellite 102, and each of the transmitted spread spectrum data signals 108 includes a data stream including information identifying the particular satellite 1 〇 2 from which the data stream originated. The GPS receiver 106 typically requires spread spectrum data signals 1 〇 8 from at least three satellites 102 to be able to calculate a two dimensional position. The receipt of the fourth spread spectrum data signal enables the Gps receiver 106 to calculate the three dimensional position using known techniques. Thus, the GPS system allows a user of a device with gps receiver 1 〇 6 to determine its position on the earth to within a few meters. In order to utilize this capital. Hole, $ see practices rely on an electronic map that allows the location of the user to be displayed on it. Providers such as TeleAtlas (http://www.teleatlas.c〇m) provide examples of such maps. These electronic maps not only allow the use of the Gps system (or by other means) to display the user's location on the electronic map, but also allow the user to plan the travel route and its likes (the route selection is intended for this route planning, The electronic map is processed by a navigation device provided by the navigation computing device. Specific examples of the navigation device include a satellite navigation device (s), which is conveniently referred to as a portable navigation device (PND). However, it should be remembered that the present invention is limited to Instead, it can be changed to #处理的处理的的处理装置. Therefore, any functionality can be seen from this; Tian T see, in the context of this application J50832.do, -16· 201211508 : Devices: To include (but not limited to) any type of route planning and navigation, the device is embodied as a PND, a transporter such as a car, and a portable computing resource (eg, carrying out route planning and navigation software; Ϊ Personal computer (PC), mobile phone or personal digital assistant (and) or servo@ 'or other computing devices that provide this functionality across the network. • _ 3a and FIG. 4b show an example of a navigation device in the form of a PND, and it should be noted that the block diagram of the navigation device 2 does not include all components of the navigation device' but only many examples A representative of the component. The guide is located in the outer casing (not shown) - the navigation device navigation device 2 includes a processing circuit including, for example, the processor 2〇2 mentioned above, the processing Cry 2〇2 to connect to input device 2〇4 and display device (10), for example, display screen 206): = here refers to input device 2〇4 in singular form, but those skilled in the art should understand that input device 2〇 4 denotes any number of input device keyboard devices, voice input devices, touch panels and/or any other known input device for inputting. Similarly, display honor screens 2〇6 may include various such as liquid crystal display 7^ Any type of display screen of the LCD (LCD). In the case of a Si device 2°°, the processor 2〇2 is operatively connected to the input device 204 via a connection 21° and can be input via the connection - (4) information. And operatively connected to J through the respective output connection 212 And at least the output device is fined to turn the information out to =. The navigation device may include an output device 2. 8. For example, an acoustic reading device (for example, a speaker). While generating audio information for the user of the navigation device, it should be equally understood that the device 204 can also include a microphone for receiving a round-in voice command: Soft 150832.doc -17- 201211508 卜 'Air Show 2 〇〇 can also include Any additional input device 204 and/or any additional output device, such as an audio input/output device. Processor 202 is operatively coupled to memory 214 via connection 216, and = further adapted to input via connection 22 The /output (1/〇) bee 218 receives the message/sends the information to the input/output ^/whistle (10) where (5) the bee (1) can be connected to the I/O device 222 outside the navigation device 20A. External 1/0 device 222 may include, but is not limited to, an external listening device, such as an earpiece. The connection to the 1/〇 device 222 may additionally be a wired or wireless connection to any other external, such as a few car stereo units, for, for example, hands-free operation and/or for voice-activated operations, For connection to an earpiece or headset and/or for connection to, for example, a mobile phone, where the action "connection is available to establish on the navigation device 2" and, for example, the Internet or any of the eight-powered roads The connection between the poor material and/or the connection to the server via, for example, the Internet or some other network. The memory 214 of the navigation device 200 contains a portion of the non-volatile memory ( For example, to store a code and a portion of the volatile memory (eg, to store data when the code is executed). The navigation device also includes communication with the processor 202 via connection 230 to allow A removable memory card (generally referred to as a device is added to the device 2〇〇. In the depicted embodiment, the port is configured to allow the addition of an SD (Secure Digital) card. In other embodiments, the material may be allowed Connect to other formats of memory (Such as,

Fid (CF) card, Mem〇ry, heart, (9) memory card, coffee (general rate bus) flash drive, MMC (MultiMedia) card, 8th position her fairy card, Mic^odnve, or the like). 150832.doc -18- 201211508 Figure 2s illustrates the operative connection between processor 202 and antenna/receiver 224 via connection 226, where antenna/receiver 224 can be, for example, a Gps antenna/receiver And thus will serve as the GPS receiver 1〇6 of FIG. It is to be understood that the antenna and receiver benefits represented by reference numeral 224 are schematically combined for purposes of illustration, but the antenna and receiver may be separate positioning components, and the antenna may be, for example, a GPS patch antenna or a helical antenna. In addition, the portable or handheld navigation device of Figures 3 and 4 can be connected or "connected" to a vehicle such as a bicycle, locomotive, car or boat. This navigation device can then be removed from the articulated position for use with gamma or handheld navigation. In fact, in other embodiments, device 20G can be configured to be handheld to allow navigation by the user. Referring to FIG. 3a, the navigation device 2 can include an integrated input and display device 206 and other components of FIG. 3a (including but not limited to, an internal Gps receiver processing state 202, a power supply (not shown), a memory Units of system 214, etc.). ❹ The navigation device 200 can be located on the arm 252, which can be fastened to the vehicle dashboard/window/etc. using the suction cup 254. This arm 252 is an example of a docking station to which the navigation device 2 can be coupled. The arm navigation device can be rotated on the arm 252 by snapping the navigation device 2GG to the arm 252 to connect the navigation device or otherwise connected to the arm of the docking station. In order to release the connection between the navigation device 2 and the docking station, a button (not shown) on the navigation device 200 can be pressed, for example. The other suitable configuration for coupling the navigation device 2〇〇 to the docking station and the navigation device 2〇0 to the docking station is familiar to the person 150832.doc • 19· 201211508 is well known to the skilled person. Turning to Figure 3b, the processor 2〇2 cooperates with the memory 214 to support the § (Basic Input/Rolling System) 282' which functions as a function of the navigation device 2 and is executed by the device. The interface between the software. The processor 2〇2 then loads the operating system 284 from the memory 2U, which provides an environment in which the application software 286 (which implements some or all of the route planning and navigation functionality described) can be executed. The application software 286 provides a work environment that includes a graphical user interface (GUI) that supports the core functions of the navigation device, such as map viewing, route planning, navigation functions, and any other functions associated therewith. In this regard, a portion of the application software 286 includes a view generation module 288. In the depicted embodiment, the processor 2〇2 of the navigation device is programmed to receive GPS data received by the antenna 224, and from time to time, along with the timestamp indicating the time at which the GPS data was received, Stored in memory 214 to accumulate the series of locations of the navigation device. Each of the lean records thus stored can be considered a GPS location; that is, it is the location of the location of the navigation device and includes latitude, longitude, time stamp, and accuracy reports. This series of locations can be considered as location data. In one embodiment, the data is stored substantially periodically (e.g., every 5 seconds). Those skilled in the art should be aware that other cycles will be possible and that there is a balance between the resolution of the I data and the memory capacity; that is, as the resolution of the data increases by taking more samples, more Multi-memory ^ to keep the data. However, in other embodiments, the resolution may be substantially female: 1 second, 10 seconds, 15 seconds, 20 seconds, 30 seconds 45 seconds, 1 minute, 2 5 \50832.doc -20- 201211508 Minutes (or indeed, in any memory of this memory, any period between the products). Therefore, it is being loaded. Accumulating the traces of the device 200 at various points in time Ο 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二Fixed (approximately 15 seconds period can provide a suitable upper limit). Additionally, the processor 202 is configured to upload the record of the whereabouts of the device 200 (i.e., the data in time) to the vessel from time to time. In the embodiments where the navigation device fan has a permanent or at least commonly existing communication channel that connects it to the feeder, 'the upload of the data periodically occurs (which may be, for example, every 24 hours - times) Those skilled in the art will appreciate that other cycles are possible 'and may be substantially 30 minutes per minute, hourly, every 2 hours, every 5 hours, every 12 hours, every 2 =, weekly or here. Any time between cycles. In fact, only the fine example + 'processor 202 can be configured to upload the track of the track in real time', but this inevitably means that the data is actually transmitted from time to time. While there is a relatively short period between shots, and thus the data can be more accurately corrected, in this pseudo-immediate embodiment, the navigation device can be configured to be in memory 214 and/or The card inserted in the cassette 228 buffers the GPS clamps and transmits the GPS positions when a predetermined number has been stored. The predetermined number may be approximately 2, 36, 1 , 2, or such numbers Any number between them. Those who are familiar with this technology should understand The predetermined number is partially controlled by the size of the card within the memory 214/埠 228. In other embodiments that do not have a communication channel that is typically present, the processor 150832.doc '21 201211508 can be configured to be in the communication channel. The record is uploaded to the server when generated. This can occur, for example, when the navigation device 200 is connected to the user's computer. Again, in such embodiments, the navigation device can be configured to be in memory or inserted The GPS location is buffered on the card in port 228. If memory 214 or a card inserted in port 228 becomes full of GPS positioning, the navigation device can be configured to delete the oldest GPS location and, thus, can be considered Is a first in first out (FIFO) buffer. In the depicted embodiment, the track record contains one or more traces, where each trace represents the movement of the navigation device 2 within a 24 hour period. The hour period is configured to coincide with the calendar day, but in other embodiments 'not necessarily this condition. The server 150 is configured to receive a record of the whereabouts of the device and store this in a large volume data store For processing, therefore, over time, the mass data storage device accumulates the uploaded information of the navigation device 2 from a plurality of δ records. In general, the server will collect a plurality of navigation devices 200. The server can then be configured to generate velocity data based on the collected orbits of the device or each device. The velocity data can include time-varying speeds and/or, showing how the average speed of the map's navigable segments is Time variation. Figure 7 shows an example of an electronic map in which the electronic map is viewed across a network (in this case, the Internet) on a navigation device. The implementation described herein. The first step in any of the examples is to generate this electronic map 19〇〇. Available at hUp://r〇utes.t〇mt〇mc〇m/?Lid= 150832.doc •22· 201211508 l#/map/?center=55.01%2C-3.445&ZO〇m=4&map =basic View the map. In order for the electronic map of Figure 7 to be used for path selection purposes, the electronic map has a series of nodes that are typically not visible to the user of the electronic map. Thus, nodes and the like can often be referred to as map material. However, for ease of understanding, some of the nodes in node 3 are shown in FIG. These nodes are provided at the intersections, end zones, etc. of the navigable segments (in this case, road segments) and are used for path selection purposes. When the user asks his navigation device to plan a route between two or more points, the navigation device plans a route between the relevant nodes of the map. Thus, the map material can be used to generate a route based on at least one criterion set by a user of the navigation device. Those skilled in the art should appreciate that electronic maps also include other nodes that provide the shape of the road segment, the location of the entrance to the building, and the like. Thus, each node 300 has at least an early road segment from which it is issued by the user, and typically a plurality of road segments. For example. , point 302 has four such road segments 3〇4&, 3〇讣, 3〇 complement '3_. In other embodiments, the navigable section may represent a path that is not a path of the road, such as a sidewalk, a pedal lane, a channel, a railway section, a track, or the like. However, for convenience, reference is made to the road segment. Thus, the map displayed by the jin shows a plurality of road segments to its users, each of the = road segments representing a portion of the navigable route in the area covered by the map. I know the route planning method (such as the gamma method, the A* method or the 150832.doc -23· 201211508 M_/Pape method). However, these route planning methods can be extremely slow in terms of calculations. An example of how this route can be planned is shown in us 6 636 8 ,, the teachings of which are incorporated herein by reference. & The embodiments described herein produce a so-called side file containing minimal cost data in a pre-processing step that is used by the navigation device to accelerate the generation of the route when processing the electrons. This communication can be maintained as binary data in the form of a bit vector; that is, a string of 〇 and 1. For the eyeglasses, the side story may also be considered as map material, but the side file (9) may be supplied with or without the electronic map (as shown in Figure 7). Accordingly, some embodiments of the present invention may provide a map tribute as a map that may be separated from a side file, while other embodiments of the present invention may combine map and side file data. However, those skilled in the art should be aware that if a side file is used, the side file should be used in conjunction with an electronic map for which the side file is generated. If this is not done, it is conceivable that an incorrect route will be obtained (for example, a route that does not minimize the cost formula set by its user). Again, different embodiments of the present invention may specify different parameters (in this embodiment, a series of bit vectors) for generating minimum cost data. Thus, if subsequent route planning using the generated map data uses parameters that are different from their parameters used to generate the minimum cost data, the minimum cost data is unlikely to be useful for the route planning. For example, some embodiments may generate minimum cost information for traveling by car through a map. If the follow-up route planning is used to generate a minimum cost data specific for the walking of the 150832.doc -24·201211508 route, it is unlikely to be useful. In another embodiment, the embodiments may produce minimum cost information in the case where the user is willing to travel along a high speed road, a toll road, or the like. In the post-fourth line plan, if the user requests not to use the route of the expressway, toll road, etc., (4) the minimum cost data is unlikely to be useful.八 Eight copies: Some embodiments of the monthly cost can produce a plurality of sets of minimum cost data & each having a different set of 敎 criteria. For example, embodiments of the present invention may approximately produce a minimum cost of 2, 3, 4, 4, 15, or more sets of the following. Therefore, in the case of some implementations and for the production line, in the map preprocessing step, the nodes are divided into a plurality of zones, and thus, any map is divided into a known number of zones, for example, '__ and the determination is made The least cost path between the zones. This pre-processing produces data that can be used with the map to increase the speed of the route planning method. ❹ Normally pre-processing is performed before using the map, regardless of whether the map is used on a web site or on a device such as a chat. Therefore, the preprocessing step is often referred to as a server side handler. Although any of the materials will be suitable for performing the hearing, those skilled in the art should understand that the higher the performance of the device, the faster the pre-processing is. Typically, the architectural calculations are used for preprocessing. Such χ86 gantry devices typically perform operating systems (〇s) such as Micr〇s〇 ftTM Wind〇wsTM, υΝΐχ, LINUX, 〇SXtm, and the like. However, other embodiments may use other computing platforms such as the RISC architecture. And 'as discussed elsewhere, pre-processing can be performed in parallel, and can be performed on multiple computing devices at 150832.doc -25- 201211508 or at least. 芏v in a number of processor cores or real processor cores ,) Perform preprocessing on . For imaginary as the next pre-processing step, the processing is as follows, 304a to 304d) to determine whether the mother-to-force segment of the 1 θ is not in the map.

In each district of the district (Ν), 1 L | a I #匚匚 takes part of the small cost path and generates a quantity (minimum cost estimate). Because of this, for each road segment in a zone, the bit vector contains a bit 70 for each parent & That is, the bit vector is repaired by one bit (except for the area under consideration, each area (10)), and the magic check is determined to be 〇 or whether the road segment is formed to the area indicated by the bit. The cost depends on the road section. - Some embodiments may add additional bits to provide additional information, "headers", visibility areas, adjacent areas 'and the like. Those skilled in the art should understand that, in this sense, The minimum cost path is determined against a number of different cost criteria. For example, the following can be used to determine the minimum cost: the shortest distance; the shortest travel time; the least cost (in terms of environmental impact); the use of minimum gasoline; Produce ^ less co2, f, etc. In the current embodiment, the minimum cost is broken against the shortest travel time. Therefore, those skilled in the art should understand that σ Ν is large for a map covering a large number of areas. Pre-processing can take a long time. 'Risa. Therefore, in the example map covering Western Europe and Central Europe, for example, there are usually 40,000, 0 nodes with 1〇〇, 〇(10), 〇〇〇 Road segment. Suppose there are 1 area in the map, then Ν is equal to 1〇〇, and there is 150832.doc -26- 201211508 100'000'OOOxiOO (N); that is, 1〇xl〇9 bits are required yuan To store 99 (i.e., N-1) bits of each road segment of the map. This map may use approximately 500 Mb of storage in the case of embodiments of the invention described below. WO 2009/053410 discusses a method of assigning a speed profile, which gives the speed of the navigable section along the map as a function of time. This application is hereby incorporated by reference. For the purpose of path selection, the 4 knives of whether a navigable segment constitutes the fastest route to another zone will vary according to time. That is, as traffic density increases/decreases (eg, at peaks) Time and the like), using a navigable segment will become less desirable/more desirable. Some implementations can store a plurality of bit vectors for each road segment. The skilled person should understand that the recording speed The higher the resolution, the higher the number of required bit vectors. However, other embodiments may utilize time-varying functions to evaluate the most determined speed route (much as described in Figure 2〇〇9/〇5341〇). Q. Crossing the electronic ground When planning the route, it is believed that it is necessary to be able to use the time granularity with a precision of one hundredth of a second to perform route planning; that is, it is enough to specify the departure time from a node to the accuracy of 〇_〇1 second.正确 Correctly set the minimum cost route. Therefore, when considering the level of accuracy, it should be understood that the above considerations = 40, _, delete nodes and 1 GM (9), maps of _ road segments have teeth One of them, 〇〇〇 2 possible routes. When the time dimension is further considered, the number of the route # ', the number of disasters is further increased: 7 (that is, the number of days per week) 24 (ie, Daily hours) χ36〇〇 (that is, the number of seconds per hour is 1〇〇150832.doc •27·201211508 (that is, one hundredth of a second) xl〇〇,〇〇〇〇〇 〇 2 = 604.800. 〇〇〇·〇〇〇·〇〇〇·〇〇〇.〇〇〇.〇〇〇 (thousands of seven thousand) possible routes. In the case of the current processing level and using a simple method of exploring each segment at each time interval, 19.178.082.191.780 821 years will be spent. The map of Western and Central Europe currently sold by the applicant has 12 道路 road segments' to further increase this time. Therefore, it should be understood that a large amount of information is required in order to store and process the data in the pre-processing stage. Thus, embodiments of the present invention use techniques to significantly reduce the amount of pre-processing. These techniques are now described in more detail, and some embodiments of the invention may utilize all such techniques, while other embodiments may utilize only some of the techniques. Different embodiments of the present invention may use different combinations of the features outlined below. Thus, the embodiments may utilize all of the techniques described to reduce the amount of information. Other embodiments may not utilize any of these techniques to reduce data. Some embodiments of the present invention do not calculate the =-element vector of the road segment that meets the predetermined criteria - as shown in step 1902. For example, if it is not possible to travel along a road segment with a pre-shipped transport category, then their bit vectors may not be calculated. And ^ Therefore, the bit vector of the following road segments is not calculated in the example of setting the transport mode criteria to the car: • Railway • The segment that exists in the map data that does not correspond to the road segment; 150832.doc • 28 - 201211508 • A one-way road in the wrong (illegal) direction; • a road segment that is in the form of a transport wheel (for example, a pedestrian walking area or a sidewalk when considering the driving sex of a car). A road segment at the outlet (ie, Cannot avoid from the road section; ::::: Reduce the consideration to assess whether the road segment forms the minimum cost network: 7: This technology 'also shows the road road shown in Figure = Figure 6 from Figure 6 The path that is not useful for the minimum cost assessment is removed = this reduction is for the smallest subnet, and the minimum subnet should be evaluated for minimum cost. In the 2 terminology, the network that is left and as illustrated in Figure 7 can be described as the undirected representation of the strongest component of the road network, the component of the two-way connection. The core network is determined by applying standard techniques to two strong connections (also known as two-way connectivity) while adapting to road traffic rules such as 4. restrictions and local access. This is the case. The surgeon should understand that strong connectivity indicates that it is possible to travel between two nodes in two directions (ie, from A-B and B-A). Therefore, referring to both Fig. 5 and Fig. 7, it can be seen that the exit-free road such as a dead end, a few 252, and the like has been deleted. However, it should be noted that there are nodes remaining at the beginning of the road segment that has been removed (for example, 4〇〇 in Figure 7). This situation is due to the reasons discussed below. After reducing to the core network (eg, as shown in Figure 7), the 150832.doc •29·201211508 core network is partitioned to provide the zones described above. In the described embodiment, the network is segmented according to a multi-channel arc splitter (multiway (s) separatGr). Those skilled in the art will appreciate that in this network, if a road segment (i.e., an arc) that is equally divided by a zone boundary is removed, the remaining road segment in either zone will not be connected to any other zone. Knife Warning Network - Step 19 0 6 However, before performing the split, perform the following steps: 1. Determine and divide the island independently. The island tends to be separated from other parts of the network and is thereby linked by means such as a ferry link. These are expressed in a different way than typical road segments (i.e., the average speed is significantly reduced, etc.), and if such links are included in the segmentation, then the method will not perform as well as expected. 2. Shrink the intersections, roundabouts, and other junctions represented by more than one node in the network so that nodes belonging to the same intersection, etc. do not fall into different zones. 3. Shrink all the navigable segments to share a simple path of the same set of features (eg, ferry, NoDriving, Nothrough, etc.) (ie, the connection between two nodes without the possibility of turning) to reduce the transfer to the segmentation system The input size of the network. For example, looking at Figure 6, node 308 can be folded onto road 31. The method of using the divide and treat (divide .and conqUer) 4 for the description of the month is summarized in httP://labri.fr/perso/pelegrin/scotch. The method outlined in this paper divides the map to include several zones, and the X zone is not based on the map area, but is based on the number of sections 150832.doc -30- 201211508 contained in the zone. Having a node so organized has the advantage that the basin can help make the path selection using the zones more efficient because it facilitates the delivery of zones having similar local search diameters. Each node has an identity identification number associated with its associated hID, and thus, the nodes in the same zone have a phase π 叫 叫 ID. The nodes belong to the area at the level of the two levels. According to the grazing case, the special level is a rough level (for long distance path selection). Level L-1 is a detailed level.

It is convenient to set an upper limit on the maximum number of nodes and allow some flexibility below this upper limit' instead of setting the absolute number of nodes contained in either zone. For example, the method can specify that the maximum number of nodes per zone is 1 ’ 'this can result in a node spread of typically 60 nodes to 1 node per zone. The upper limit is considered to be advantageous over a fixed number of nodes, as a fixed number can result in zones with forced shapes, which in turn leads to the second best path selection when using the zones. Multiple Ranks In the described embodiment, segmentation is performed to provide a plurality of levels of nested regions, which are conceptually illustrated in FIG. It should be clear from Figure 9 that there is a hierarchical nature in which the lowest level (〇) is subdivided to provide level 1, and level 1 is subdivided to provide level 2. The number of nodes in each zone varies between ranks and has more nodes than the lower ranks (which tend to cover larger geographic regions). Thus, the coarser level (e.g., level 0) of the described embodiment is typically used for long-range path selection purposes, while the finer level (e.g., level 2) is used for short-range path selection. 150832.doc •31 · 201211508 The number of bits per road segment—every less ν knives contain the number of bits for the mother in those ranks. The concept of the capital county to be coded is provided (in an exemplary embodiment, the number of homing will typically be L = 3): • Global level = 0, which will typically have 1 area. • Intermediate level = 1. There will be 10 zones (ie, 100x10) for each zone of the 荨 〇 。 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • , 100x10x5). The use of the hierarchical level is flawed because it reduces (possibly significantly reduces) the amount of processing. In the description of the Fen 4 ·<►田刖贝example, there is i 〇〇xi 0 x 5 zones (ie, 5 zones) U. In a flat structure with the same number of zones, the method outlined in this paper would be required to involve 5000 zones. However, as described in this article In embodiments, this number can be reduced by reference to nodes in the appropriate level. Once the various factors of the map are known, The number of levels and the number of zones per level will usually be adjusted. For example, how many stores can be assigned to maps and path selections should have a speed of execution. Those skilled in the art should have a compromise because As the amount of processing increases, the map size becomes larger to maintain the data 'but the map can be used to calculate the faster route. In the embodiment, there are 3 levels of zones. However, in other implementations, this may exist. Any number of levels. For example, there may be approximately one of the following number of levels: 1, 2, 4, 5, 6, 7, 8, π 150832.doc • 32- 201211508 or ίο. Using 100 above, 〇〇〇, the simplest encoding for each road segment will be coded for each size, and for each meta-vector (arp flag). So as follows: Example (usually with 4 〇, 〇 〇〇, in the case of a node and a map of Western and Central Europe, the area ID of each node of the level is fixed using a fixed size bit for each road segment of the fixed level. Count

Each node area ID at level 0 will use i〇g_2(100) bits. • Each node area ID at level 1 will use l〇g-2 (i 0) bits = 4 bits. • Each node area ID at level 2 will use log-2 (5) bits = 3 bits, etc., and each bit vector (arp flag) will use 1 unit of bits ( 100 areas minus 1) for the current zone. Each bit vector (arp flag) at level 1 will use 〇 one bit (i 〇 zones minus i for the current zone) • level 2 Each bit vector (arp flag) will use 5 bits (5 areas minus 1 for the current area). Figure 9 shows the most coarse level of area 6〇〇, which provides 6 areas in the figure. To 6. Each of these zones is subdivided into additional zones (in this embodiment, nine subzones), as indicated by the dashed road segments within the coarse grade zone 6〇〇. In the described embodiment, another level of subdivision is used, which is also subdivided into each of the zones provided by the dashed road segment, thus providing a three-level segmentation 'but not for ease of reference in FIG. Show these levels. Other embodiments may of course use more levels (e.g., 4, 5, 6, 7 or more levels) 150832.doc -33.201211508 or less (e.g., 1 or 2 levels). Thus, embodiments of the present invention introduce a so-called visibility region for a finer level zone, step 1908. The visibility area of the k zone is a collection of (匕丨) zones in which the zone can be distinguished on its own bit in the flag set. Naturally, the visibility area always includes the (k-丨) region to which a given k-region belongs. In addition, it may contain some adjacent (k_1} regions. The visibility region should be counted during preprocessing, and the visibility region should be stored in association with the bit vector. It can also be seen from the opposite side that the k region and its (k- 1) The relationship between the visibility areas of the level: each (k-Ι) area knows its outskirt consisting of the neighborhood level area. In this way, it can be seen that the flag set of the road section is no longer running through. The entire map has a fixed length; the fact is that the bit vector in each region has its specific length A+B+c+N periphery 〇+N periphery 丄. Where A refers to the most rough level, B refers to the middle The granularity region, and c refers to the region of the finest granularity (in the embodiment where there are three levels). Subsequent preprocessing calculates the visibility region to generate a bit vector before the shortest path computation. The method for finding the visibility region can be described as follows : For each k zone, start the breadth-first search in the adjacent map of the zone; then for each visited k zone (the location is close enough to the start zone), add the (k-1) zone it contains to Visibility area. Close The relationship should take into account both geographic metrics (eg, the distance between the median points) and the theoretical distance of the graph. (Thus, if a region is geographically distant, it is linked to the current region by a quick connection. , the area can be close to the same wood κ, if it is difficult to travel between the areas, then a zone can be considered "distant" (even if it is geographically close). The threshold can be subjected to experimental adjustments because it depends on the specific characteristics of the map (such as the average diameter of the metropolitan area), which is most susceptible to the negative effects described above (such as increased pre-treatment, etc.). Navigable segments of exceptionally long travel time (such as ferries or NoDriving road segments) are hidden during adjaceny graph traversal, such that the visibility region is always limited to a single island or a small island belonging to the same region. The area of the area visited during the priority search period. The area including the smallest (inclusion-minimal) set that completely covers the area below the neighborhood is called the visibility area. The inverse relationship is called vicinity. · The neighborhood β (the list of ranks) consists of all the regions at level 丄+7 with g in its visibility region. To provide a specific example, and see Figure 9, the example of the visibility area is as follows: Right determines that the minimum distance between each 1 area and the edge of its visibility area should be at least 1, then the visibility area of some selected i areas will be composed of the following areas (the own 0 area can be It is omitted because it always exists): • 15: (ie, Level 15 No. 15 does not have a visibility area because it is at the center of its Level 0 area.), • 28: 5

(That is, the No. 28 of the No. 1 area has a single-D-visibility area of the No. 5 of the district (grade 〇)). • 37: 2, 5, 6 150832.doc -35· 201211508 (ie, 'Level 1 District No. 37 has three visibility zones, which are Grade Zone 2, Grade Zone 5 and Grade Zone 6 6) The level 1 level of the neighbors, also determines the level 1 neighbor of each level. Therefore, as an example, • 1: 21, 24, 27, 28, 41, 42, 43, 51 (ie, for level 1 area, level 1 area: 21 24, 27, 41, 42, 43, 51). Thus, in this example, it can be seen that zone 28 is listed although it is not in the leftmost row of zone 2. This may, for example, be because zone 28 has a fast link to zone ^ and is therefore close when considered in terms of time rather than distance. Those skilled in the art will appreciate that proximity can be judged against other measures such as greenness (minimum c〇2 or the like) and the like. Encoding of Neighbor Lists Figure 9a shows more details of the levels that can be used compared to Figure 9. Level 〇 is the most coarse level and can be considered as level k-1. For simplicity, in Figure 9a, only Zones 1 through 4 (i.e., 602; 604; 606; 608) are shown. Each of these k-Ι regions is further divided into nine regions (丨 to 9), which may be referred to as k-level regions or level 1 regions. In addition, each of these level 1 areas is divided into four k+1 level areas (i.e., level 2 areas). Generation of a bit vector Once the network has been partitioned into three levels of zones, it is processed to determine the least cost path to each of the zones and to generate a bit vector for each segment of the zone 'Step 1910. Therefore, as discussed above, the cost formula for 150832.doc -36-201211508 is used to analyze the minimum cost path for each road segment in the Ren-Zone, to all other regions, whether it is the bit vector of the road segment, as shown in the figure Shown in 7 is a bit vector that is not simplified for each of Hi. ', 'solution, Figure 7 show should also understand that the minimum cost path is determined, between: pairs: for a number of time periods to calculate (as shown in the following two Ο 。 calculation. The network has been removed Hi reduced network Road) to perform this node with " effectively fingering onto the source from which it originated. In the heart of the word, the area (4) shown on Figure 5 is placed on node 400. Oral: Yes, the per-bit vector contains three bars · the leftmost column 700 containing the identity identification number of the node at the beginning of the track under consideration; the node containing the end of the road segment under consideration The second column t of the identity identification number and the third column of the bit vector containing/the road segment. Therefore, it should be seen that each road segment is identified by two nodes, one at each end of the road segment. Any suitable path selection method can be used to determine if the road segment is part of the lowest cost route. In this particular embodiment, the entire network is explored using the well-known Dijkstra method. ' And by using various strategies to reduce the amount of processing time. Those skilled in the art will appreciate that the techniques described above for reducing the network will also reduce the amount of J processing time. Embodiments of the invention may use any one or more of the following: • 1^3 (4) f* 言10 4ki, the well corresponds to all of the bit vector entries of a zone. In the reverse 150832.doc -37- 201211508 diagram, the shortest path tree for each boundary node. This method is advantageous because the parent zone can be processed in parallel, thereby reducing processing time. • Reduce recalculation of subtrees like the shortest path. • Do not recalculate the bit vector that has been generated. Thus, the 'in total' embodiment may perform the following operations to generate a bit orientation step: As will be appreciated by those skilled in the art, a directed acyclic graph (DAG) representation. The cost and long extension table) are reversed, and the search through the electronic map can be used to map and accompany the data structure (turn 2) shrink the simple road segment about the finest level (ie, fold the end nodes onto each other, such as As discussed in the section), if the road segment consists of the number of times = 2 - or more nodes (all nodes are in the same zone of the given level) 'and the navigable segments have the same attributes: for example ".f=, "W_" and " IsNGDriving", Bay "; The road segment is called simple. Wishing (that is, fr〇m_no e) and / or tail (that is, :, 岣, whether the node belongs to the district, the road network The section of the road is divided into three groups: the exit, the entrance and the inner road section. That is, both the old and the tail are in the same area, then the road segment p + is called the inner road shoe and the right head is in the area. However, the tail is not in the area, then, the road section is called the speech, the road section, and the right tail is in the district but the head is out of the zone is the entry road section. Bay Road 4 · For the fresh one &amp of the NoThrough and NoDriving areas There are road sections set up 150832.doc -38- 201211508 special turn restrictions. The processing routine is in a bottom-up manner > .B 仃 ' at the finest level of 鈒 p p, starting at level 2 in the described embodiment. At each zone, for each zone and repeat The following steps: Level 1 - determine the exploration area of the 。. At the top level (for example, the area in the described embodiment), which is the entire picture,

~ 圚 at a finer level, which is a subgraph limited by the visibility area of R (as described above)

/ Therefore, at the intermediate level (that is, 'level υ, only for the slabs containing the grading area of the grading area 1', see the 厪& field to perform the following steps. At level 0 (should be remembered, used for beans) W is used for long-distance path selection), the road segment of the entire map. The heart collects the entry road section of the visibility area, that is, the point from the outside of the area to the area within the area. Then collect the border road section, That is, the road segment after the road section is broken. If the head (♦ tail (L2) does not prohibit the turn from luL2, the road segment L2 is called after (and L1 is before L2). In order to find the border road segment To shrink a complex intersection into a single node. The ferry road segment and the road segment with the "Nerving" attribute are not considered as border roads. The road segments collected in this way can be reduced (possibly significantly reduced) in the following The amount of processing required in the described exploration steps. The exploration step can reduce the exploration of the map and consider their route to the given route including the entry route. 2. For each exit road segment, Find the root road segment and The number of steps is as follows: In this case, if the exit road segment is 丄 and a single exploration step is performed. If no, the road segment is taken as the direction of the root road of the single exploration step in both directions. If there is a pre-emption: If the itching road segment is a single: the road illusion acts on the root road/I of the separate exploration step, the most unruly road segment is the most 疋, what to wear Passing out the road to the mother of a slave road - the extended route ® dry road (if it is located in the starting road section of the take: the root road section of the early stepping step. At a finer level (ie 'except the rank. ^Special treatment of the ferry road section. As mentioned above, in the neighborhood of the disc, the ferry road section is ignored. If the head section of the ferry road section does not belong to the visibility area of R, a single exploration step will be performed. The ferry road section is the only-root road section and the exploration area The domain is constrained to the head zone itself. 3. Perform the scheduled exploration step (described below). 4. Chasing the shortest path from the road segment to the root road segment in the exploration area. In addition to the highest level (also That is, at all levels except level 0), the ordering of the road segments from which the individual tracking starts will affect the result. Assuming that R is the level L area (where L> 0), the level (ε) area is processed first. The road segment, then the remaining exit road segment at level L, and the like to the finest level; 150832.doc •40, 201211508 α is the internal (and has not been removed) road relative to the finest level The segment is finally processed. When is the brake ρ + ± hot 'To X# The road segment that was visited earlier in the tracking, there is no one, the mouth is the path. At the same time as the memorial, set each arrival (the appropriate target bit vector in the segment. At all levels except the highest level, and 〇), perform additional actions··, have crossed the level L for the first time. After the boundary of the area, mark the boundary road section and all other road sections going to the root road section with the flag of the transport section, and mark the shortest path with the u-turn flag to make the corner of the curve. At all levels except the fine level, the fill phase map is simplified for the lower-level after all the areas of the ΪΓΓ, and (4) is recorded as all the road segments of the transport road segment. The node of the new simple path and protection point accumulated by the receiver. Second, the edge is marked as a u-turning section. Finally, the road bh removed before the bit is expanded according to the correlation matrix to the relevant matrix in the processing of this level = two: the example uses the correlation matrix 'its mismatch The relationship between the two road sections. In addition to the finest level of the temple level L, a new correlation matrix is generated and used ^ ^. At the level, the matrix is coded according to the level 0 + area ^ 21 , 〇 ^ ^ ^ 索 丨 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , .doc •41 - 201211508 Combined. At lower levels, most matrix entries are equal to the empty set, ie, the matrix is sparse. ^ The purpose of the off matrix is to help set the bit vector for a road segment that has been deleted at an earlier stage. These road segments are not included in the directed-acyclic order resulting from the level L exploration step, but if they are not deleted, they will be included in the directed non-cyclic graph. More precisely, the correlation matrix is used to rank a road segment in the exploration area of ^ (which is deleted during the calculation of a certain level t of [not the finest level]) (8) Set the bit inward for the level π + 〇 region R in the search region of S, the matrix element M[U] will eventually be the level correction at the boundary of the visibility region of & Passing all of the shortest paths from the SiR (in the reverse view) through one of their zones. A matrix 4 is generated that initializes all entries to an empty set. Next, for all rank L zones S, the following two actions are performed. L matrix establishment: for each root step segment and each of the obtained directed acyclic graphs in or on the weighting step, for each level 0 + 0 region R contained in the search region of s, and for each of R Entering the road segment Γ 'Update the matrix element Μ [Foot 51] as follows: & Use Α to indicate the exploration area of R (as calculated earlier for the level) From D/Follow the way to / and check if it is Leave a: If the road segment on this route is from the level Μτ to the level called). If T is still in a' but not included in A, add Ding to the collection [foot S] and treat the lower one. • Item fetch matrix. For each road segment in the exploration area of S that has been deleted before the start of level B calculation I50832.doc -42- 201211508, assume that R is the level 0 + 7) area where the road segment ends (again Once, relative to the inverse graph, the logical vector OR of the bit vector bit of the region S on the road segment is now set, where T includes all elements of M[foot q. 'The bit vector bits of each T will be set directly at an earlier level according to a directed acyclic graph or a similar procedure involving a correlation matrix at a lower level. Exploring Ο y_ ^ 1 person J sentence is very It consists of a directed acyclic graph of the small-sensitive cost path k. This 彳I4 nqx is based on the variant of the Dijkstra method used for the minimum cost calculation. Objective function, for example, such as line-entry time or estimated fuel consumption, or any other factor described elsewhere. - One embodiment uses travel time, machine length, and suppression of improper turn and Weighting of other penalty items with.

Make the following changes to the classic (Dijkstra) method:

1. It works on the road segment map. The items that B (4) use to visit, relax, etc. are road segments, not the section turning restrictions and/or traffic extension. The force allows the method to take into account that the 2.== function is a vector of the roots of the root road segment for the arrival time slot = (travel time, cost). When you choose, you will have a brother-in-law mode (free flow, work 早 early peak time, secret 丄々 ^ f night peak time, etc.). This is further explained in the following discussion of the i, , and cost of evaluating a plurality of time periods -43· ^0832^ 201211508. 3.1 = Store more than one label for a given road segment. If the set of two unfinished (unfinished) traffic extensions are not equal, then the picks themselves are referred to as independent ', and both are continually propagated on subsequent road segments. Otherwise, if there is a difference between the cost function values of the different arrivals, the parent is not replaced (that is, a standard iron discards the poor label. Otherwise, borrows two; good) 'goes. The wrong one is the mother and the time slot. The better value of the production =: two instead of the original standard to spread. The merged label 4 is the union of the original set of pre-sets in each: Γ. • Produce a special (10)” on both segments. The code starts in both directions. The true #(disambiguation) route broadcasts a u-turn label that does not pass the 17-turn to the standard and the general label/M... ',, the retrospective phase when the set bit vector is affected. When the start tag is not the same as the same track, the shortest path is added to the flag when the shape of the turn is poor. It will explore the limited area of the area to monitor the border road sections as defined above: : mouth? There are border road sections for a long time, according to the time slot;; search surface to reach the established monitoring comb (Zhan Can (3) kiss. Value range The above-mentioned standard is hidden only on the road outside the domain when it is piled up. ^Mutong privately weaves at least one time. The value of this function is lower than that of the current monitoring comb A. If the exploration area extends the time-varying function + maintains for each island - a separate monitoring comb. Some embodiments of the invention can be calculated to be displayed over a plurality of time periods rather than at 150832.doc • 44 « 201211508 single-time crossing The bit vector of the minimum cost path of the network. Familiar with this: the technician should understand that The lowest cost route through the road network can be changed over time due to the influence of traffic density, etc. Because a, for any node, there may be two or more minimum cost paths, each-minimum cost path For different times. In this embodiment, the bit vector is not encoded with a time reference applicable to the minimum cost path of the material. The bit vector is only set to identify the navigable segment as part of the least cost path or not

Ο is part of the minimum cost path. Because &, when using minimum cost data for routing: the selection algorithm will have to consider all possible minimum cost paths from the node. This process will now be briefly described with the aid of Figure 1A. In the standard Dijkstra exploration of the Internet, when the network is explored, the method uses the total cost incurred to date to reach the point in the network plus the expected cost to be incurred. Thus, some embodiments utilize a function rather than a discrete value to make a cost estimate at each node. Thus, at (iv) _, Figure 751 is being probed. "Per-node (for example, 75 〇) has a cost function associated with it that identifies the parameters being explored (eg, time, cost of fuel, or the like) The cost function can be associated with a road segment rather than a node. The method then processes the cost of the cost to add a new cost by summing the function at the current node with the cost that has been accumulated so far to generate a new function. The example shown in Figure 1a shows at 752 the cost function generated by the Sogou method at node 750 and shows how the travel time varies with the departure time (ball). It should be seen that the cost function is Increased at points 7 5 4 and 7 5 6 during morning and evening 150832.doc -45 - 201211508 rush hour. In a particular embodiment, the cost function is stored at 5 minute intervals (eg, average speed over road segments) That is, it is a function of a quantization function with a time period of 5 minutes instead of a continuous change. If a road segment is part of the lowest cost route at any time period, then the road segment is set. Bit vector. Projecting core data to the entire network The above describes how to reduce the network contained in the map in order to reduce the number of road segments and nodes that must be considered by the segmentation method. However, further consideration should be given to the "salty steps". In the deleted section, the point, the more path selection method can still generate routes to or from the deleted road segments and nodes. Thus, the nodes and road segments to be deleted are assigned to and in the core network. The same area as the node to which the connection is connected. The size of the bit vector generated by the compression as described is very large, and therefore the information needs to be compressed. The path of the yoke of the moon is exemplified in different ways. Pressure, two ',, and '- embodiments utilize various techniques to compress, coalesce, and/or align bit vectors, followed by Huffman coding of the bits. Jt 匕, —jtt The two examples can try to ensure that there is a distribution of bit directions, because this can have unevenness in the neck. It helps to ensure that the sputum code is more sloppy than D in other cases. The bit vector distribution is as follows: 150832.doc 46· 201211508 (49% of the time) (49% of the time) (2% of the time) == Fuman - manipulates the bit vector to have less (79% of the time) (19% of the time) (2% of the time) Decrease the resulting bit vector Ο To reduce the amount of bit vector that needs to be generated, embodiments of the present invention use any one or more of the following strategies: • Not for All nodes make the (4) ID and generate the zone 1D only for the navigable nodes (for example, ignoring the nodes corresponding to the railway). 〇00000____ 11111..., 00000. 11111. Ο ? Ο -5 ? • and = all road segments are required A bit vector, and the bit vector can be used for decision road segments around the decision node. Decision nodes and decision road segments (as described in this document) can be determined by viewing road segment data. • Although there are many possible bit vectors, some bit vectors are much more frequent than others, 7L is frequent, so special encoding can be used for the most frequent bit vectors (for example, 〇〇〇.〇〇〇 and U1 1U) . • =〇: Most of the bit vectors of _ or (1)...111 are still "set to 1' or set to 〇. So the Huffman code of a partial block should encode the bit vector quite efficiently. • In the close to each other, the bit vector is the same, so the difference encoding can effectively encode the bit vector. One::: by: the source area reorders the destination (10) so that the different areas have more similar bits The vector pattern (the idea is described in this document). 150832.doc -47- 201211508 • All bit vectors around the road segment of the node should always be given. It can be used appropriately to encode the bit vector more efficiently. Some of these situations are discussed in further detail below. = It is worth noting that 'the technique described in this paper is intended to reduce the size of the bit vector. However, 'note that' the map is used for the purpose of path selection. The device of the material needs random access to the data. Usually, the effective programming of the data requires a variable size. However, this will prevent random access to the data.

Thus, embodiments of the present invention may use a trade-off in which the data is G-coded into a series of indexed pages, and then using the variable edits in their pages = 'in these embodiments' For a random access (via a foot index) of each page, a page has been accessed, and the embodiment can then decode the entire page. This provides a trade-off between efficiency and map size - increasing the number of nodes per page reduces the size of the map, but slows data access. One particular embodiment of the invention uses 16 nodes per page. It should be understood that any one node may also have a different number of road segments leaving the node. Thus, even if there may be the same number of nodes, there may be a variable road segment per page. In addition, different compressions can occur for each of the bit vectors stored on each page. λ This structure can result in a map format as shown in Figure 11. In this figure, the Arp flag is The McCaw is synonymous with the bit vector. The number "η" is stored in the constructor and the number "η" can be changed for different maps to minimize the performance of the map. β weeks, ie η" is a compromise between the map size and the access speed of 150832.doc -48 - 201211508 degrees when decoding map data: • A large η value will group many nodes in _, which is for the map? It is beneficial, but it is not good for the speed of random access to data. • Small η values will group a small number of nodes, which is good for the speed of random access to data, but is not good for ground closure. • The history of “η” can be defined as (for example) 4, which is the page of '16 originating nodes (the originating node is at the beginning of the road segment, ie, the column in Figure 7), but it should be remembered that Each-initiating node has several arriving road segments (9)_road segments, so in the case of assuming an average of three arriving road segments, each means that the storage amount per page is equivalent to ~ one road segment. In this case, depending on the coded area (four) level, the data has a different format. Figure 12 shows an example of the format of a level 〇 (i.e., the most coarse area), and Figure 12 shows an example format of areas of other levels.

The bit 70 vector and related information are stored in the "data structure", and the format of the data structure is shown in the figure " The format contains the following: a header is used in the Huffman coding described later. Fuman's discussion; the number of districts in each stratum is 804 (in the case of a certain number of zones per level • is empty), the zone neighbors 806 (empty in the case of zoneless neighbors); _: Remap table 808; bit vector page (10) nodes) 810; and region ID and :: vector 812. The data structure of the hold bit vector can be maintained in a single file, or it can be held in multiple files. In some embodiments, the ground icon header 800 is configured to contain additional information indicating each of the following: 150832.doc -49- 201211508 • The maximum number of levels. • The length of the shortest neighbor list. • The length of the longest neighbor list. The bit offset of the section containing all neighbor lists. . Keeping the files of the λ or each file may have the size of the list " The per-list length is encoded relative to the shortest list length, a fixed number of bits of ^BitsRequired(l〇ngestListLength - shortestListLength)^1. Note that if all lists = have the same length, no bits are needed to encode the lengths. - Then encode the contents of all the listings. Each-list consists of a number of 7L groups of neighboring zone IDs: pairs at level 、, 3 element tuples at level 2, and so on. f α, does not encode the from-regi〇n tuple (the part of the ASCII file that precedes the "·"). It is implied because the list of all districts is stored in ascending order. For example, if the map has 3 levels of H)()x1Gx5 ((10) areas at level °, 1 area at level 1, and 5 areas at level 2), Bay·· at level 0 At the beginning, the wrong originator j 7 t Like ^ 1 2, 3, ···, 1〇〇 list (1 0 0 list in this order).丨tBu stem, main suppression l month early) Each of these 4 early mornings contains several pairs. • Store the originating area at level 1 '2.1, 2.2, ..., 2.10, 3.1 (1 〇 〇 list in this order) 3 element tuples. List of U, I·2, 13, ..., l.io, ..., 100.9, I 〇〇. Each of these lists contains because it is the last level. At level 2: Nothing is stored, 150832.doc -50- 201211508 So there are no neighbors. Each component in the tuple is stored as n bits A. The number of bits per level is determined based on the number of zones in the corresponding level. Therefore, it is possible to randomly access the list. In the case of 3 levels 1 〇 (4) (4), the coding tuple He will use 7 bits for & (because there are ι 区 areas at the level )), for 4 bits of b (because There are 1 area at the level ), and 3 bits for c (because there are 5 areas at level 2). Example: Assume that l〇〇Xl〇x5 areas have the following divisions: rough level

100 districts, 1 district at the intermediate level i, and 5 districts at the detailed level 2. In the file at level 0, the section containing the neighbor list will contain: i (9) the number of the list of the 100 units in the 4th level. The number of bits π is calculated according to BitsRequired (longestListLength-shortestListLength). Each number is relative to the shortest list at that level (the shortest list is stored in the header). • Followed by the contents of 100 lists (1〇〇). The first element of the parent pair is encoded on 7 bits (because there are 1 area at the level )), and the second element of each pair is encoded on 4 bits (because it exists at level 1) 10 districts). In the file at level 1, the section containing the neighbor list will contain: • 100 Χ 10 = 1000 number 'which indicates the length of the list of 1000 zones at level 1. •• Next is the content of 1000 lists (1000 3-element tuples). The first element of each tuple is encoded on 7 bits (because there are 100 regions at the level )), and the second element of each tuple is encoded on 4 bits (because 150832.doc -51-201211508 is There are 1 level 丨 in each level, and the third element of each tuple is encoded on 3 bits (because there are 5 levels 2 in each level 1 area). In the file at level 2, there is no need to store anything because it is the last level. For encoder input files, a list of any form can be used. Therefore, once the division has been performed into three levels, each node is assigned to one of each level; that is, three areas. The header 800 is in communication, and the header used in the embodiment of the present invention is small, and thus, it is not necessary to optimize the size in order to reduce its size. Usually, for convenience, all content is byte aligned or word aligned: • (4 digits) encoded version (increased each time the map format changes) • (4 bytes) Map flag (to turn features on or off, initially 0, but can be used later if you need to add optional features) • (4 bytes) Total number of nodes in the ground • (4 Bytes) Total number of road segments in the map • (4 bytes) Segment Huffman tree byte offset • (4 bytes) Segment region Binary large object (bk) b) Bytes offset • (4 bytes) Segment area page information byte offset • (4 bytes) Segment bit vector page information byte offset • (4 Bytes) Segment Variable Size Recording Byteset Offset • (4 Bytes) Maximum Byte Vector (4) Flag) Page (in bits) (can be used by route planning method at startup) Pre-allocated bitstream decoder 150832.doc •52- 201211508 worse condition) • (4 bytes) average bit vector (arp flag) page size (in bits) () to interpolate the bit vector page position) • (4 bytes) minimum bit vector (arp flag) page difference (to make all the differences >=〇, thus avoiding the storage of the sign) • (2 bytes) The maximum size of the bit vector (arp flag) history (used by the route planning method to pre-allocate the history buffer at startup) • (2 bytes) of the road segment per page Maximum number (currently unused) • (1 byte) Apollo (Ap〇ll〇) level of this file • (1 byte) Bit per bit vector (arp flag) • (1 Bits) bits per bit vector (arp flag) page difference (fields in fixed size records of bit vector (arp flag) page) • (1 byte) per binary large The bit of the object index (the field in the fixed size record of the area page information) • (1 byte) the bit counted per area (the block in the fixed size record of the area page information) • (1 bit) Tuple) Each nontrivial bit vector (arp flag) block bit (1 byte) area node page large Log- 2 () Huo (1 byte) bit vector (arp flag) page size 1 〇 [2 () (1 byte) used to encode the region 1] 0 Hoff The number of man trees (1 byte) is used to encode the number of bit vectors (arp flags) of the tree. 'Hoffman tree 802 150832.doc -53· 201211508 is used to encode around each node. Hoffman's number of road segments, only about 10 yards, only exist in the archives of the ranks / the non-trivial bit vector (& flag) Man-t's largest waffle tree, the larger the compression, the better, but the more memory you need in ϋtj $ & + (between map compression and road: memory usage in planning methods) Eclectic). Tian Liao Shi Da, 霍夫 〇 位 位 位 位 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( Tree: very small, only 4 codes when the historical size >=n, the bit vector (King flag) difference code Huffman tree. Very small (the number of Huffman trees is stored in the header) There are 3 zone time zone ID Huffman trees in the zone page: very small, only 3 codes when there are 4 zone time zones in the zone page Hoffman tree: very small, only 4 codes when in the zone page Huffman tree with >=n zone time zone (1): very small (the number of Huffman trees is stored in the header). The other parts are small, but they are also available. Wherein the geographic phase region remapping table 804 and the region id list 8〇6 can be used. Although the region ID 806 is larger than the rights format compression region ID 806 of FIG. 11, as illustrated in FIG. 14 to reduce the amount of data used. In most cases, each area page stores a list of distinct areas in the area page 150832.doc -54- 201211508 It is expected that this list is small (in fact, many π reports may only contain i at least level 0 Area). The list of zones has a variable size. The data in the page (i.e., random access) should be stored at random and, therefore, a fixed table size is used to allow this. Each of the distinct zones is sorted by the frequency of the nodes in each zone: the list, the first element corresponds to the zone with the largest number of nodes in the page, and the last element of the list has the smallest number of pages The area of the node. Ο Ο For each node in the zone page, the zone ID (local to the page) can be used to encode its zone ID. Too much of this local ID is the area in the page (which is a small integer ', often because G corresponds to the most common area in the area page). The ID of the area of the lang point is stored as follows: • All possible overlaps of the area in the array storage page of the area 1D Early in the morning. The list of zones is the contiguous zone m in the array. Lists can (and do) overlap. The array does not store the beginning and end of each list (this is done by the area page information table). The area page is a fixed-size record table (so you can randomly store the number of items included in the list at the beginning of the list + the number of items in the list. ', •:) lang points contain - local The node ID (as opposed to its page for each of these concepts in the context of this article). The area array area array encodes the page & 'Bei's list of all possible areas. It is a simple array of area 10, 150832. Doc -55- 201211508 column, where the list of area IDs can overlap. The size should be small because the list overlaps. The area array is further described in the section area page information. The area page information uses the fixed size of the area page information table. A list of the zone IDs in the two clamp tables in the record: • Zone count (the number of zones in this page, expected to be small) • Offset in the array of zone lists (where the zone list begins In an embodiment, this is described in Figure 2. The offset intercepts in the array of regions: (for example) assuming that there are always less than 256 regions in each level (which is a reasonable assumption) ), each zone 1 has a fixed size of 1 byte The recording system is sufficient (but if each level of 256 bits is considered too restrictive, then each zone is more than 8 bits is easy to implement). The zone count in the 1 page table record indicates the specified offset. How many zones are in the array to be considered. If several zones have the same list, they can point to the same location, which should be = two's. We can expect many pages to share the same list or share the same early morning part of the page. To explain this in more detail, Figure 12 (for example, nr=9 to divide 512 nodes in the starting point, 2lU points, ie, and together), note the extent of the array ID of the array 900 is also compact, Example: The same in the column - the location or overlap Z stomach can point to large objects in the array). In fact, the number of pages is increased (= the area of the mark is doubled, because each page then decreases the energy of the array by a small number of areas. Therefore, this array should allow the generation of the eve $ so the combination It can be used as a lot of memory on the device to which the generated map data is loaded without the need for too much 150832.doc 56 - 201211508. The array of IDs is as small as possible. This method is intended to reuse the same list of regions in the array as often as possible. This method freely reorders the last two elements of the list because of its Does not affect the size of the Huffman code.

For example, when a list contains 2 regions and 34, the storage list 10, 34 or 34, 1〇 (independent of frequency) is equivalent, because when there are only 2 nodes, the Huffman tree is Only one bit is used in all cases. In other words, for the last two zones, the requirements for sorting by frequency are relaxed. Similarly, the list of 3 elements 10, 34, 4〇 can also be stored as 1〇, 4〇, 34, because only the first code 10 (most frequently) will use the “retention bit and other code 40 and Each of the 34 bits uses 2 bits (independent of the order). Thus, looking at Figure 12, it can be seen that the array 900 stores two values: the length and the offset from the beginning of the region data. Therefore, in the first column (3. 〇) For example, this refers to the three pieces of data offset from the beginning of the file (ie, μ, 34, 4〇). As another example, the array input (丨: 2) refers to the list— The area (that is, the length is 1), wherein the offset from the beginning of the file is two; (also 40). In an alternative embodiment, 'code the area page information according to the following method: The number of sub-zones. The number of sub-zones can be - variable in the rank. However, the number of sub-zones of each rank is constant, relative to the minimum number of zones at each level and using bg 2 ( Max_number_of_regions-min_number_of_regi〇n) 仿 _ 4立来150832.doc -57. 201211508 The number of coding sub-areas. So if If the number is constant, then one bit is used for the code area count, and this section is empty. The minimum number and the maximum number of areas are stored in the header of the side file. And Figure 6a discusses the encoding of the neighborhood of this section for a list of zone neighbors at a more detailed level L+1 at a given level of each gram. For example, the zone at level W D may have the following list of neighbors at level L=2: 3.5 4 3·6·3 3.6.4 4.7” 4.7.3. As discussed elsewhere, the list of neighbors can be used to increase the speed of the pre-processing used to generate the side or side standard. This section has been split into 2 subsections: • Subsections that are used to encode the length of all neighbor lists (as many as the number of zones at a given level). The length is encoded relative to the shortest list and then the number of bits is calculated as 1〇g-2〇ength 1〇 / Sh〇neStJiSt). If all lists have the same length =. The _ bit is then used to encode the subsection of the length (and therefore the section is empty • used to encode all neighbors to have more samples).匕 , 数目 数目 数目 数目 疋 疋 疋 - - - - - 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目 数目The code-dimensional #^ two-dimensional table is used to reorder I, the quasi-table, which (more effectively encodes the bit to the _ position of the rendition and the bit reordering) in this document, the following - Step-by-step new sorting and coalescing to optimize the Huffman encoding of the replenishment of the bit vector of the bit in the 兀 alignment. This table is used by ί 50832.doc -58· 201211508 route planning method to find the key in the bit vector when knowing the position of the bit to be decoded:

• The originating zone ID of the current node • The original arriving bit (t0_bh) is the index of the bit (that is, the bit index after the quantizer). The two indexes of the two-dimensional table are: • Originating zone ID • Destination bit Index (area of destination) This section consists of 2 sub-sections: the segment where the de-aggregation bit is aggregating to each of the regions used to encode the level 〇. The area used to encode the number of bits per number number_〇f_c〇alesced_bits) is l〇g_2 (max is used to encode the segment of the bit remapping table. Because the destination bit is used for path selection No change (the destination remains the same when the path is selected), but the origination area (depending on the node being explored when the path is selected) 'matrix is stored in the column of destination bits. In each column , storing the number of bits reordering for each originating area. The route planning method should generally only load one column corresponding to a given path selection destination into the memory when performing path selection. The route planning method does not need to be the entire The two-dimensional matrix is stored in the memory. The route planning method can randomly access the list. The number of bits per code-remapped input is calculated as l〇g_2 (the maximum number of coalesced bits is shown in Figure 13 for level 0) The zone (see Figure 9) extends the I50832.doc -59-201211508 bit 70 inward section 812 of the block shown in Figure 11. As can be seen, each page contains the road segment count, ID and bit vector〇 ° ° * Figure 14 for Levels other than the level 扩展 extend the bit vector segment 812 of the talent case shown in Figure u. It can be seen that for other areas, the bit vector is stored and the road segment count or region m is not stored ( For the level, they are categorized.) In view of the difference in the area covered by the area of each level, the number of nodes per page in each level may vary. For example, because of the level, the area within the area covers a large area. For level 可能 there may be more nodes per page per zone. For example, 'for level 〇, some embodiments can store M2 P's per page per page, and the finer level can have a smaller number. Node_e.g., 256 nodes '128 nodes, 64 nodes, 32 nodes, 16 nodes or the like. In contrast, an embodiment may utilize 1 node per page, 2048 nodes, or 4 〇. 96 nodes. The encoding table 81 〇3 of the bit vector 810, 812 has a fixed size record. The originating node IDs are grouped together in 2n pages. 'Grouping data into pages with multiple consecutive nodes For convenience, because of the period, 7L to $ for the same neighbor Several road segments in $ have similar types. By using pages, it is possible to encode several road segments in delta and achieve good compression. Similarly, 'there may be differences in the node's area': another aspect of the page , which means that accessing the data of a road segment requires de-encapsulation of the right-hand road I (not directly random access). It may be necessary to decapsulate several nodes or road segments to access - road segments or nodes. The funding of 150832.doc -60-201211508 is considered acceptable because the data can be cached, so it will be useful soon after the #data"not a useless road segment is read. This additional information (four) - ahead) parameters). Pre-reading on disk cache • When compared with the Dijkstra A* path selection ratio, the path selection using the bit vector reduces the size of the extended search by -1 to the order of magnitude. By knowing the page as a group of materials, only the map is small

Ο J port P knife still needs to be decoded (along the page of the actual path). • Thanks to the difference compression between the zone ID and the bit vector, it should significantly reduce the size of the encoded data. • The page reduces the size of the index 'because the data will be stored in the side file as described. Each record in /810 contains a - "Difference" field, which is used to find the position of the variable size of the parent page (relative to the difference between the interpolated positions). The number of bits for each difference is stored in the header. In order to access the zone ID and the bit tl vector 812, the decoding 计算 can calculate the estimated offset of the start page by performing the following linear interpolation: interpolated_offset=fr〇m_node_id*avg_pagesize 八 avg_page_size is the bit stored in the header Average page size (possibly fixed to improve accuracy). The offset of the data can then be calculated as follows: offset shifted-offset+min-delta+delta where min-delta is the minimum of all differential intercepts for all pages (store 150832.doc '61 · 201211508 in the header) And the difference is Zen 10, ... the unsigned stop stored in the page. The min-delta value ensures that all lambs are ordered to be positive (do not store the positive sign). A record of a variable size is obtained by the placement of the previously described bit to the difference in the ribbed. Each record contains 2n metrics (the zone ID of the originating node and its bit vector attached to the road segment at all levels). The same indexing scheme will therefore be used for all levels. Variable size record storage: • Page number - for the entire hundred _ /ι4> ^ Whether the node in the page is the part of the same area; • On the bit vector page φ 夕 # ^ 母至郎点The number of road segments (stored only at level 0, as it will be the same for all levels); the starting point in the page is the point m, one for each level (1) (for all levels of information stored in the level) In the case of (4) 〇*zhen; • only the bit vector of the road segment around the node in the page at level i (only around the nodes with > 2 attached road segments). This is the largest part of the data. • The encoding of the number of road segments around each node. For the two of the two points in the bit vector page, the Huffman code: Ma is in the The number of road segments around the node. This information is not specific to all levels, and it is only stored in the file at level Q (4) q * Shenzhen. Knowing the number of roads around the node is also used to decode the bit vector (see 1000, Figure 13). Bessie, the information that already exists in other buildings, weighs 150832.doc -62- 201211508 but makes it easier and faster to decode the page ten-bit vector, and no/page looks for information in another place. This small increase in size provides an increase in performance. Encoding of the zone ID in a variable size record

• Encode the zone ID 1002 of the node in the variable •• size record after encoding the number 1000 of road segments around each node (see Code Layout). In order to perform path selection using the bit vector generated in the preprocessing, it is usually necessary to access the area 1] 0 1002 at all levels of the given node, and the area IDs C) of all levels are stored in the same file close to each other. Instead of being split into different files at each level. A bit vector page with 2n nodes (eg, n=4) and having 3 Apollo levels will therefore store the node ID as follows: node#0: I〇cal_region_id_level_0 local_region_id_level 1 local—region_id”evel_2 node#l: Iocal_region_id_level_0 Local_region_id_level^l local_region id level 2 〇~ ~ node#2: l〇cal_region_id_level_〇local_region_id_level^l local region id level 2 node#15: local—regi〇n—id level—〇i〇cal_regi〇n—id—level l local_region—id—level_2 In addition: • No code is coded around nodes with 1 or 2 attached road segments (ie, for storage 〇 as the number of attached road segments 150832.doc -63· 201211508 points). When the bit in the page number is set at a given level, then it is known that all nodes are in the same zone at the same level, and the domain only encodes the number of times (^) _ times (for having >=3 attached) The first node of the road segment). The number of bits used to encode the zone ID is i〇g-2 (regionCount-l) ° • In addition to the first node of the zone code in the page, a bit is also encoded before encoding each zone 1D. This bit indicates whether the area m is the same as the previously encoded section _ at the same level. When this bit is set, there is no need to encode the zone 1D because it is the same as the previous code of the other class. When this bit is 〇, it is called the code area ID by (4)-Chong Xin (10). Since many consecutive nodes are in the same zone, it is often only necessary to encode the zone ID by 1 bit. The Huffman coding of the index of the region is valid because: • The regions are sorted by frequency in each zone page (so the local index 0 is more frequent than the local index 1, ...). There are different Huffman trees in each of the ten-pages (one Huffman for the three regions in the page, and one Hoffman tree for the four regions in the page). and many more). The Huffman tree is small, and thus it is possible to store several Huffman trees without using a large amount of memory. "," 'How to 'at least at level 0' has 3 zones or more than 3 zones should be quite rare (but not at other levels). The code of the bit vector in the variable size record 150832. Doc -64- 201211508 Second:=::=1:- Segment) The coded bit vector segment around the node (the node of the road segment, the road grip selection method; the attached U1.·.111) implicitly Assigned to their nodes. Values are not coded to reach the node. See Figure 10 'It should be noted' that two nodes can specify a road segment ❹ There is a node at each end of the road segment. Because &, when considering the direction of the Japanese:: The node at the beginning can be called fr〇m_n〇de, and the point at the end: can be called to_node. The various properties f about the bit vector are used within the encoding to make the encoding: • Bit vector Many of them are 〇〇〇〇〇〇 or j丨i丨丨丄. • For other values of the bit vector (ie, non-〇〇〇〇〇〇 or lu ui) 'very likely duplicates and the same value Will probably be repeated. 〇• Again, the logic of all the bit vectors around a given node Or should be 111...111. The bit vector is encoded as the first Huffman code and optionally other Huffman codes. The first Huffman code indicates whether the bit vector is: • For a trivial bit vector 〇〇〇〇〇〇之〇 • For the ordinary bit vector 1 11... 1 11 code 1 • Code 2 to indicate that a non-trivial bit vector has not been encountered in this page. Only in this case, there are other Huffman codes followed to encode the newly encountered bit vector. 150832.doc 65 · 201211508 • When the bit vector is the same as the bit vector previously encountered in the current page ( Ignore the trivial bit vector 000...000 and the code >= 2. This encoding therefore uses the history of the previously encountered code in the page. The code is thus actually indexed by the history of all previously encountered codes. In addition to this Huffman code, only in the case of non-trivial bit vectors not found in history, 'need to encode more information (code = 2). In this case' immediately following Huffman code == 2 After that, the encoding is: • No locating element • A number of bit vectors used to encode n regions by blocks of N bits The Fuman code (Ν and η are given in the icon head). For example, using 11 blocks of block encoding 1 区 area (99 bit vector) requires encoding 9 Huffman codes. (9 X 11 = 99) Since most of the bit vectors contain 〇 or contain 1, the locator indicates whether the bit vector is stored as negative. This makes it possible to store most of the contents in the Huffman tree. 〇 code (hence, Huffman coding of the improved block) ° Only if the size of the block is smaller than the number of zones, there is no locator 'this is actually the situation at the level ,, but at level 1 or At 2, the monthly bit b encodes the entire bit vector in only 1 block, so no positioning element is unnecessary. If there are 100 zones; N=100 (hence, a bit vector of 99 bits), then the first block encodes the bits for the destination area 1 to ,, the second block coding area 12 to 22, etc. . In the first block, the LSB (Oxl) corresponds to the destination area 1 'the next bit (0x2) corresponds to the area 2, the next bit (0x4) corresponds to the area 3, and so on. 150832.doc -66 - 201211508 For the bit vector using history, the depth of the historical array is the number of distinct bit vectors previously encountered by the page (not considered _ 'willing ten bits vector 〇〇〇 ..·〇〇〇 and 111...111). Different Huffman trees are used when the history vector contains one element, one element, two elements, three elements, and the like. ^ The Huffman tree multiplication is acceptable because all Huffman trees are both 'and thus do not require a large amount of storage:

• When the history has 0 elements, the Huffman tree has 3 codes: for 〇〇〇...〇〇〇, for 111."1111, for the new bit vector. • When history has one element, the Huffman tree has 4 codes·· for 〇〇〇···_0, for (1)...(1)! For the new bit vector 2, for the same bit vector as the element #0 in history. • When the history has 2 elements, the Huffman tree has 5 codes: 0 for _...000, for (1)... Chuanzhi, 2 for the new bit vector, 2 for the element in history and #0 3 of the same bit vector, 4 for the same bit vector as element #1 in history. •and many more. It is expected that the size of the bit it vector page is small (for example, 2 „original nodes), so the number of Huffman trees is expected to be small. However, it is possible to limit the size of the Huffman tree to a maximum value: for example, Whenever history contains more than one element, a single-Huffman tree (the value η is stored in the map icon header) will be used. This code encodes only the distinct bit vector _ some codes in each page. In the case of an index page, the encoding is more efficient in size, but at the cost of slowing down the decoding in order to use the bit vector for path selection. 150832.doc -67- 201211508 (There are more bit vectors in the page to be Decoding) Statistics Here is a detailed statistic of the file format in 254 regions (ι grades > encoding Netherlands, Belgium and Luxembourg. Use the following input table number • Number of bits per bit vector block: 11 Number of nodes per bit vector page: 2 8 4=16 Number of nodes per zone page: 2Λ9 = 512 Provide statistics to give a map in terms of map size and a description of the map format based on some actual data: concept, - Global statistical counter: 1SSS632 -598632 22〇1δ〇72'?138 (45,5|; 87437736 13l〇9i4 yi3348 36243? 607717 235171 87437736 4 Q f 2880a ^5 G64 5621s 249788. 84§〇l672 2847844 211245^ 79841370 1639322 i 8 2 b69 6 I668053 i463l83 607717

74g57299

Node count.............................. Line words................ .............. Skip the line around the node with 1 line. - Skip the line around the node with 2 lines. , --- at level = [0] Statistics: Map counters........................ Coded ordinary arp flag 〇〇〇.., 〇〇〇 encoded ordinary Aip flag in... 111 encoded arp flag in history....... not encoded in the history of the arp flag... negative block.......... ......... Map size (in bits)................... t domain standard #.......... ............ Hoffman Tree................... District II carry large objects.......... .............. Area page information................. arp flag page information-* *........ ..... Variable size record............... Line count around the node.......... ίρ Point area ID....... ,.......... ............................. Code ordinary 000...〇〇〇. ........ Code ordinary 111...Ill ......... The code found in history.....:...... Not found in history....... ..*. Not found in history.. No positioning 7L ... ......................... 150832.doc -68- 201211508 All sizes are in bits. The total map size is 87,437,736 bits (1〇, 929, 717 bytes). Indentation reflects the hierarchy. The variable size record information is the maximum piece of information (96.975% of the map size). Sub-items (indented) give more detail in variable-size records. The bit vector is the largest piece of information (91.3120/.) to be stored in the variable size record. And in the bit vector, the non-trivial bit vector that has not been encountered in the history constitutes the largest part of the map (83 〇 15%) C. Statistics about the Huffman tree

This section gives statistics on the Hoffman tree when encoding the Netherlands, Belgium and Luxembourg in 255 zones (i.e., for the map data shown above). Huffman tree-I Hoffman tree of the number of road segments around the stomach of each node: NrLines Bits: 1 Value: 3_.Ma 10 110 1110 11110 111X10 U11110 1111111 Bit: .2 .Value: 2 Code bit: 3 Value: _ 1. Code bit: 4 Value: 4 code bits: _ 5 value: 5 _ code bit: 6 Value: 6 code bits: 7 Value: 7 Station bit: 7 value : 8 _ yards Most nodes have 3 attached road segments, but in the 2nd and 3rd positions in the Hofhe tree, we found that there are 2 attached: road segments and 1 attached road segment The node (which is not the decision node). Huffman tree of non-trivial bit vector block This is the largest Huffman tree, because the storage block of the trivial bit vector is the largest mantle size contribution factor (in the case of the Netherlands, Belgium and Luxembourg 255) , for 83.015%). 150832.doc •69- 201211508 Bits: 1 Value 0 Bits: 6: Value I Bits: 6 Value 2 Bits: 6 Value 4 Bits: 6 Value δ Bits · 6 Values 16 Bits: 6 Values 32 Bit: 6 Value 64 bit 7C · ^ Value 512 bit 7C _ g value 1024 bit: 7 value 128 bit: 7 value 256 bit: _7 value 384 bit: 8 value 5 cut, too large Huffman Tree: Non-trivial Arp flag block Ma 0 code 100000 code 1Q0001 100010 code 100011 code 100100 code 100101 code 100110 code 1001.11 code 101000 code 1010010 code 1010011 code 1010100 code 10101G10 bit 70: 24 bit: 24 bits: 24 bits ::2.4 Bits: 24 Value Value Value 1534 1717 1741 1830 1973 Code Code Code Code i 11111111111111111 xiqxi 1111H11111U1IX11111101 Store all the blocks formed by 〇 as the most frequent type, and according to the Huffman tree above Only one bit is encoded (this means that even if the trivial bit vector 〇〇〇···000 is not block-coded, 50% or more of the blocks encode the value 〇). This is because most non-trivial bit vectors contain • most 〇 (and a few 1) • or most 1 (and a few 0). The encoding scheme negates (~) contains most of the 1 bit vector, so the last 'bit In most cases, the coding block of the meta-vector encodes a block containing 〇〇〇... 000 in only one bit. The next most frequent block is a block of only 1 bit (1 ' 2 ' 4, 8, 16 ' 32, 64.·.). It has more or less the same frequency' and therefore has the same (or nearly the same) number of bits. The Hoffman tree of the ID of this area is stored on each page by frequency, so it can be seen that compared with the area ID of 150832.doc •70· 1211508 other places, the storage area of the area is smaller than that of the area. Less bits (in fact only 1 bit). Different Huffman trees correspond to pages having 3 zones, 4 zones, 5 zones, and the like.霍夫 Hoffman Tree: Regions ---[ Bits: Bits: Bits: Elements Meta-Bit Bits 1 Value 0 Code 2 Value 1. Code 2 Value 2 Code Huffman Tree Regions 1 Value 0 Code 2 Value 1 code 3 Value 2 code 3 Value 3 code 0 10 11 10 110 111 —f Huffman tree bit: 1 Value: Bit: 2 Value: Bit: 3 Value: Bit: 4 Value: Bit :4 value:

2 ns.-code. Code code 10 110 mo 1111

Ns' code code code code

TJ element, element, bit, bit

接曼............ 值值值值值值值霍〇.10 110 1110 11110 11111 · · Shear · * · Huffman tree code 位 of the bit vector history code It is said that the trivial bit vector 〇〇〇...〇〇〇) is the most frequent (and in the case of large eves, only one bit is encoded). The code 丨 (trivial bit vector u i · · · 丄 1 1) is the next most frequent (and only 1 bit coded). Again frequent code (2) is a non-trivial bit vector encoded by block. The other code (>2) is for the bit vector found in the history of ▼. 150832.doc •71 · 201211508 bit·_ bit '· bit: tree man... fut value value.' rp bit: bit: bit: bit: tree@又. ....... Fu value value value Huo F..~ " Code code ag code code code bit: Bit: Bit: Bit: Bit: Bit: Bit: Bit: Byte: Bit: Bit: Bit: Bit: Bit: Bit: Bit: Bit = Bit: Bit = Bit = Bit: Bit Bit = Bit: Bit: Bits: Tree Μ........... Af value value value value Α value value value value value ΑΙ value value value value value value value A3 value value value value value value value value hohohoho tree Manshuman 2 3 code code code ο 10 210 1110 1111 code code code code code code code code code... Eyv UifT> exhaustion of exhaustion ο 10 110 1110 1.1.110 11111 ο 10 110 1110 11110 111110 111111 00 01 10 1100 1101 1110 11110 11111 .··Scission-·. Effects of input parameters on map size There are several input parameters that control the file format shown in Figure 8 and can affect the map size. Adjustments to these parameters may be a compromise between map size and memory 150832.doc -72 - 201211508 body usage or decompression speed (depending on parameters). The following input parameters may be used in embodiments of the invention: • Area page Size • Bit Vector Page Size • Bit Vector Block Size • Bit Vector Huffman Code Count • Area Huffman Code Count Bit Vector Block Size Effect __Color_. Figure △ Less (bits) ) 4 35548192 5 33648792 6 32290344 7 30853616 8 31103200 9 3043669S (default) 10 30051792 11 2S266784 12 28934696 Increasing the Huffman block size of the bit vector improves map compression. The higher the size of the soil, the better the compression. However, increasing the block size is at the expense of storing more memory for the block to store larger Huffman trees with 2,8 11 values. The above table illustrates this situation. When optimization is introduced to remap the zone ID for each source zone, the impact of this parameter becomes even more significant: optimization when using large bit vector block sizes will hopefully result in significant map size Reduced. Bit vector page size effect 150832.doc •73- 201211508 Map size (bits) 2Λ1 55563944 2'2 42502936 2Λ3 34898840 2Λ4 30436.696 (default) 2Λ5 2.7389952 2Λ6 25165032 2Λ7 23635936 页Page size helps to better compress map. Unfortunately, the large page slows down the decompression of the data in the file format by the way the path selection method is accessed, since accessing the bit vector of a random road segment requires decoding all the road segments in the page. The table above illustrates this situation. Bit vector Huffman code count effect value map size (bits) - — — 1 1 30866320 2 30748024 3 30634168 5 30504504 7 30467944 5' 3〇4366S6 (default) 11 30423688 Add bit vector huffman code The number helps to increase the compression slightly, and this is illustrated in the table above. Increasing this value has almost no disadvantages, because the Huffman trees are small anyway. Adding to more than 9 Huffman trees (preset values) does not provide any significant improvement. Increasing this parameter may be more efficient when the bit vector page is large. The bitwise vector in the coalescing and reordering bit vector has a pattern. Each of the source regions (i.e., the regions of the nodes storing the bit vectors) are significantly different in type. The number of bits used to store the bit vector of (10) elements can be reduced by storing a small translation table for each source region. These tables perform two functions, which are further described in this section: ^0832.(30( -74· 201211508 • Bit coalescing • Bit reordering θ ::? Understand this idea as follows: When in Spain, clear ^.% of the road section (for the destination region of Sweden bit = 1) Can> to Norway (that is, 'the destination for the destination of Norway is also 1). The second road 'no access to Sweden (bit = 0) 'in most cases its Ο Ο ^ bit ^!&quot威. So # in Spain, the destination area of Sweden and Norway 7G to the value of the 罝 position a few hands ~ 疋 phase 4. In fact, for many purposes: and δ 'It is always strictly equal. Which bits The full correlation between the yuan and the - 很大程度上 depends largely on the originating area. For example, 'When in Hex, the target regions of Norway and Sweden are much less relevant. Another ^ —. Finland, the destination region of Spain and Portugal... is likely to be 100% relevant (or at least very close to the surface). Take advantage of the nature of his H - 疋, , , 悤疋 equal (completely related). They can be clustered into a single -, where. The coalesced bit reduces the number of bits to be encoded in each region. The number of bits is reordered using a property of this mn〇/iB two-position 相当 which is quite high (but not 100 〇 〇), with ^ + maximizing the Huffman coding of the bit vector ( And shuffle the bits in the manner of reducing the Hoffman tree. 150832.doc -75- 201211508

The reordered position a丨 by the regional ship Hoffman compiled by the bit ~ Λ

BeorderEits() (2) Agglomerated bits I completely related to the group of bits + are agglomerated into a single bit"

I

CcciIe.3ceB.it3 (} \ . The original bit to be coalesced + reordered - ^ 一 · · ^ - secret - ^ + Before calculating the table used to coalesce the bits and reorder the bits, Calculate the correlation of all pairs of bits in all originating zones. The number of distinct pairs is: C(N, 2)=Ν!/(2!*(Ν-2)!)=Ν*(Ν-1 ) / 2 ... where N is the number of zones. Therefore, when the number of zones is high, this number is quite high. Calculating all bit correlations is the slowest way to generate the file format shown in the figure. The complexity of the method is ηχΝχΝ, where η is the number of bit vectors, and ν is the number of regions. For the parent bit pair in each region (ie, the different rows in the bit vector) To), calculate the two-dimensional table: bitCorrelation[fromRegionId][bitI][bitJ]. Each entry in the table contains a structure of 4 fields, and its count: • bitl=0 and bitJ in all bit vectors of fromRegionld = 0 times • Number of bitl = 0 and bitJ = l in all bit vectors fromRegionld • Number of bitl = l and bitJ = 0 in all bit vectors fromRegionld • fromRegio The number of bitl=l and bitJ=l in all bit vectors of nld is 150832.doc -76- 201211508 Although the calculation is costly (and therefore slow) in terms of processor time, it is easy to parallelize this process because each calculation The bit correlation of an originating zone is completely independent of the other originating zones. Embodiments of the present invention use multiple threads to speed up the calculation of the correlation of all bits to effectively use SMP (Symmetric Multiprocessing) machines. System: In a single CPU machine, it takes about 8 minutes to calculate the bit correlation of the Netherlands, Belgium and Luxembourg with 300 zones. But the parallelism principle is well scaled, and the thread is enabled and used. When 4 c cpiJ, it takes 2 minutes to calculate the bit correlation (the original one-fourth).

When a number of bits are fully correlated (ie, always all 〇 or all U), they can be aggregated into only one location without releasing any information. Will be fully correlated in a given zone. The bit set is called a "group", and the bit vector of a given area is composed of several groups. If the bit vector of the N bit is composed of n groups, the bit vector of the encoded N bit uses n A bit Q + small table to indicate which bits in each region are equal. This bit coalescing has the advantage of reducing the size of the resulting file in a lossless manner. This advantage is likely to follow The number of zones increases and increases because more bits can be agglomerated. Therefore, the map size increases with the number of zones: Sexually. People, in one embodiment, obtain the following information: Minimum number of groups The group's eagle ia He, Belgium and Luxembourg 255 districts 12 84 '~ The largest number of groups 152 - Helan, Belgium and Luxembourg 300 districts 13 90.017 150832.doc -77- 201211508 thus 'has 25 5 In the case of the Netherlands, Belgium and Luxembourg There are at least one even 12 groups before the Huffman coding (that is, only 12 bits are needed to encode the bit vector of 2 5 5 bits (ie, the length of the bit vector is The number of zones is the same)) (and therefore, there are even fewer groups after Huffman coding). On average, for the Netherlands, 2 5 5 districts of Belgium and Luxembourg, the zone needs 84 bits. In this example, the worst zone requires 152 bits (152 groups) before Huffman coding. The zone Id=3 in the example of the Netherlands, Belgium and Luxembourg as another example and with the above 255 zones For example, it has the following 18 groups obtained from encoded data. Area Id=[3] has [18] groups tx ^ α ) *1 ξ 1 I» *1 ί 2 5

( 3 ί 5 7 Β :l 12 13 16 ΐϊ 2C 21 Ζ 2 23 24 2h 2i 27 2§ 29 3D •3: 32 33 m 37 3.Β 35 4S 4.2 43 4』45 4* 52 5: 55 62 63 £ 7 6.9 "?:! 72 73 ·?4, :75 7 6 BS: δ: S2 S] 56 culvert 39 5D 35 96'^·· »8 l〇£: 1C1 102 ID3 124 105 :C:€ IQi 11G 112 113 ΙΓ& 120 123" 127 129 13D 133 1/23 136 1« 14.3 LH1«: U9 .13.2 LS: 135 13 LM 13? 15Ϊ L6C iis :67 ITi 172 173 :17·3 Γ?6 iT? 173 IS : 132 :B1 I3*Q 13⁄4 2DG 2Q3 2D5 2C:T 2LD 2:U 212 2L3

^15 218: 219 220 221 222 226 22? 229 235 23T 2^3: 2-12 243 24* MS 24"? 2« 23C 252 J tl i € ;. ♦ 1 ·; 9 *1 I IS:) ♦ 30 ( 13 U 35 «a 49 5T 68 i5 33 IDS 115 1:7 128 131 :S4 I3g _ i53 151 164 1€B 1'6S 137 IBS 193 2Q9 217 224 23B 233 } ♦ Ξ3 ( 17 ia 53 5« = €. 65 31 S2 99 107 :14 li6 12:1 124 123 :2€ -3" 140 1Λ1 144 24S H6 152 161 1S2 163 17Q 1~?5 13ΰ ι04 -36 191 133 294: 1« igg :93« 20 : 202 2Si 2C6 2Ca 216 223 233 2-2B 23i 231 232 233 2.3-4 236 2iC 2:4S 2« 251 ;. ' 啻i 4 3*1) *= ί ·ίΙ 7» ~9 ::1 I9T } ♦3 ·: 52 77 gi j tl i 35 ] ♦ 3· ·! 5& 61 9,i ;| tl { €*1 ) *1 { 7C } ♦1 ( 122 ) *1 ·: :53 :, - 78 - 150832.doc 201211508 Every road slave is not - group (18 groups with area Id=3). The parentheses are the index of the bits that are coalesced in each group. The number after # is the number of bits in each group. In this example, one group is extremely large and contains up to 141 zones. This is unusual. Generally, [the area on the edge of the map is more than a bit in the middle of the map. The number of bits has been reduced to the average on average. In this example

(255/84). Although, it does not mean that the map size is reduced to the original ^' because the remaining bits to be encoded have more nets than the original bits (which is more difficult to encode by the Huffman block). However, the bit coalescing bit can still reduce the map size (possibly reducing the map size). When the bits in each zone are coalesced, the bit vector in each zone therefore has a different number of bits (but the bit vector of a given origin zone has the same number of bits), as in Figure 15 Explained. The shading indicates the originating area ID of each bit vector. One has. It is quite easy and fast to identify the group of fully related bits by the bit correlation of the ten-out table (bitc〇rreiatj〇n). All bit pairs in which no 01 or 10 type can be found are fully correlated and can be coalesced into 0 bit 0. To give another example of bit 兀 coalescence, the following table gives 6 bit vectors ( Each meta-vector has an instance of 15 bits). These bit vectors have not yet been coalesced. 150832.doc -79- 201211508 C - - ΐ» ie, 乂二二二二cca':ec:::::;:ccrj This A 1 5-bit TL bit vector can be clustered into only 4 groups - Therefore, there are 4 bits per bit vector before Huffman coding. This is shown in the table below: 1 2 3 4 5 6 7 8 1 υυ 11 1 1 0 11 1 0 00 00 1 1 0 11 0 11 00 1 0 1 00 1 11 11 0 0 1 00 0 0 11 00 0 0 1 00 0 00 00 0 1 0 11 Τ' 9 1+3+9 2+11 5+7 4+8+10 1 0 1 1 0 0 1 1 0 ”1 0 1 1 1 0 0 0 1 0 0 0 0 1 0 Μ 00 π_ ΤΓ 1 Which of the left-handed points in the left-handed clothing constitutes a 4-bit coalescence vector. For,, the main dry 1 a _ # , the moon side, the 4-bit coalescence The vector brother 兀 indicates the left part of the certificate ^ brother 1 block, the third column and the ninth block. Because of the left part of the mouth in the above example, 'Y I knife (that is, 15-bit original vector) is located, The third and the ninth total are the same, so they can be represented by the first block on the upper side. Therefore, the right side of the table, in summary, the following: The table of consistency between the left side and the right side is as _ The original bit coalescing bit 0 0 1 1 2 1 3 0 4 0 5 3 6 2 7 1 8 2 9 2 10 3 150832.doc -80. 201211508 The original position of the initial bite guilloche 11 0 12 3 13 1 - 14 1 A coalescing element effectively fanned the nearby destination area and the same angle from the originating area (an The group of destinations in the gle sector is grouped as shown in Figure 16.

The shading in Figure 16 indicates the group of zones that are clustered together. The two maps (i.e., 'Fig. 16a and Fig. 16b) show the same area clustered from the perspective of two different origin regions: a and B. Which zones are clustered together depends on the zone ID of the originating node, as shown by the figure. In this figure, the eight zones are agglomerated into four clusters, and from the viewpoints of the originating zone A or the originating zone B in the figures, the coalescing bins are different / the main forbearing vertical strip zone (1300) It is agglomerated because it is in the same direction when viewed from the originating area A. In the mouthpiece, an example is illustrated in Fig. 17, which illustrates that the coalesced cells will be grouped in the area to which they are attached and from the point of view of each of the originating areas. The number of dead and the number of fans = the position of the bit vector to be coded in each bit area, for example, the bit of the bit vector of the parent g in the condition of 255 regions of the Netherlands 'Belgium and the Fitzgerald The number is: origination area-> coalesced bit 1 -> 57 2 -> 93 3 -> 18 4 -> 22 5 -> 63 6 -> 75 II··cut. .. 251 -> 21 252 -> 46 253 ->117 254 -> 8〇150832.doc -81 - 201211508 The method is accurate (it is not trial and error). Therefore, it finds the ideal number of groups of bits. In some embodiments, the coalesced bits are implemented in a lossless manner: the material is only coalesced when the region always has (four) bits. In other embodiments, however, this can be extended to cause loss of coalescence bits, i.e., such that complete restoration of the original material is not possible. Such lossy embodiments may introduce a threshold. If the difference between the bits is less than χ, then some of the bit vectors in the bit vector are replaced with 1 - or more than 〇 to allow their bits In the case where the vector is coalesced, it can be coalesced. And the case of agglomeration - the example will be to make 〇〇〇〇〇〇 1100 and 〇〇〇〇〇〇 〇 〇 π amp — 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点 优点The teacher selects the m point and the method of selection will evaluate the other road segments 'because the extra m will indicate that the road segment is part of the fastest route. In the case of no loss coalescence, the 15 limit is zero. Increasing the threshold will: agglomerate other zones' and the cost is to introduce a few bits into the bit direction but the (4) limit is kept low, then it will use the data: Speed has almost no effect. For example, if the two bits in the inward direction are almost always the same (except for (for example, only 1 bit), then this is only possible - it is possible to coalesce more bits. May be accessible! The difference is to make the map more compressed. The higher the limit, the pre-bit reordering. In some embodiments, once the bit can reorder the bits. The bits have been coalesced (such as meta-reordering and described above), not (in Hoffman, ed., 150832.doc • 82, 201211508 but helps to make Hoff' and can also help to reduce To reduce the number of bits to be encoded before the Huffman tree code is reduced, the MAN coding is more efficient, thereby reducing the memory of the number of dissimilar codes in the large, j, bit vector Huffman codes of the map vector. Body requirements. Different from the bit coalescence, u π w 』 is a few trials and columns, and a good reordering can be found, but its eight order. j find a good re-arrangement

The bit vector is encoded by block. Block A is small customizable and remains in the icon head 800. In the explanation given below, and as an example: Set the block size to " one bit. The bit vector, the number of bits may not be divisible by 11, so the last block may be smaller than U bits. The last block uses a different Huffman tree than the Huffman tree used for the complete block with i i bits. Encoding each bit vector is thus encoded: • Uses several complete blocks of Huffman for 11-bit blocks (in this example) • Uses another Huffman when the remaining bits are less than 11 The last block of the code. The following picture depicts the bits to be Huffman-encoded by block, where the bit vector is in 2 regions (which have a different number of bits after coalescing): 150832.doc -83- 201211508 Complete block ->< — Complete block <-last block,...1------------------4· .! xxxxxxxx S .i XXXXXXXXI - :- +---------------+ .I x >: χ χ χ χ χ χ I • lx XXX X >: XM j .I χ χ x 丨--- - Complete block-><--- Complete block->< Complete block--- Each zone has a Hoffman tree that will most likely be too costly in terms of memory. So in some embodiments, all zones share the same Huffman tree. Since the associated bits are different in each zone, it is advantageous to reorder the bits in each zone so that the associated bits are systematically placed at the same location in all zones. So sharing Huffman trees across all zones becomes more efficient. A table that will be used to reorder the bits of each zone is stored in the map material. Note that instead of encoding a separate table for agglomerating bits and a table to reorder the bits, instead of encoding only one table, it is two conversions (aggregation + reordering). Composition. The method of reordering the bits in the complete block is as follows: For each zone, find the most relevant pair of bits. Since the fully correlated bits have been coalesced, the remaining bits are not fully correlated. However, some bits are much more relevant than other bit pairs. In the first complete block, the most relevant bit pair (a, b) is remapped to bits #0 and #1 : 150832.doc -84- 201211508 | a b........ j . ..........I...........!xx xl The Huffman tree of the complete block is less different due to the most relevant bit pairs of the grouping Code (this has the advantage of using less memory to store data), and the statistics are more skewed (making Huffman coding more efficient). In most cases, the first 2 bits (for example) contain 00 or 11, but almost never • do not contain 10 or 〇 1 (as do other bits), rather than containing a random sequence of bits. In the case where the bits are not reordered, the first 2 bits will contain a variety of patterns 00, 01, 10, 11. After remapping the first 2 bits, the method then finds the next most relevant (c, d) bit pair and remaps it to store it in the 2 bits before the second block: la B........ .led.........I...........1 xxx ί +菜柴;箱絮结梁媒篇'^篇ΐ^ The article is written by the media; the sample. Chang Xiong + male bully tower media solution can be found in the media; - black: media; article German articles box media + article media purple media box articles articles; sigeais; ^ intercept box Fu Media-level thin + ----1_ Next, the method finds the next most relevant (e,f) bit pair again and remaps it to store it in the first two bits of the third block: + «Χ:3Σ82:·2ε4ΕΕϊ5:Κ;ε=ΙΚί;2Ε83 3:85»55:ί!£;Κ:3=£5;·ε»2!ί;Κ: +3=·2ε=2: :ΒΕ5ίϊ;»:«:δ=::=ε32;3Κ2Κ8Κ·β:;=:ΪΒ:5«»··«5ΪΤ3Κ + Χ2:8είΚ:ΚΪ:5Ζ:ΙΪ:ί38:ε:·3Β·55;: 3!ϊ3Εϊε·«5 3Κ·Μ:; ί:2Χί»:Κ«Κ-|-------}- iab,........I c d....... ...!e f.........lx x xi This method returns when the last complete block is reached The first full area - remapped block and the next most relevant bit pair (g, h). If several pairs have the same correlation, the tie breaker selects the pair that is most relevant to the previous pair in the same block (i.e., the pair most relevant to (a, b)).

The algorithmic approach continues as described above until all bits in the complete block 150832.doc -85- 201211508 have been remapped: + =:=;;:=2 2 5=£Ϊ; 2Ϊ·= 1=: :=5 = 2=: :&=::=:=:;:=;:=: two.== +=work; Si 1S: weeping two S: + two; =;=2: two male 5 ^二==二35;二2=;二:^二工工=二-j-------1-]a to ghmns fc yz . jcdijopuv AB . jefk 1 qrwx CD * ixx κ | In the example, since the block size has an odd number of bits (11 bits), there are still unmapped bits in each complete block. The method then finds the bit most relevant to "z" and stores it in the first block. Then find the bit most relevant to B and store it in the second block, and so on, until all the complete blocks are remapped: vw .jg. ―~ 议 an anϋΆ 攻 攻 软 这For sst + this + : a total of iis sss. articles: Wu as ijjisa-sasiKiasssi.sisiaiSiKsss « ί;»·.ϊ!ϋ+ ^ ^ istr off fy.· ag, sst tas asa. .3⁄43⁄4 g» aa ss.; 2;..;is.is;.SBiitsiai:iii. II-J-——————————j- iabgh.mnstyzE;cdijopuvABF|efklqrwxCDG!xxx| At this point, all the bits in the complete block are remapped . The method then remaps the bits in the last block, attempting to group the most relevant bit pairs (as done for the full block) reordering the bits in the last block can help reduce the map size However, it is not as helpful as reordering the bits in the complete block for two reasons: • The full block is more important. In this example, each code uses 3 Huffman codes for the complete block, and it uses only 1 Huffman code for the last block, so the complete block is the last incomplete last block pair. The contribution of the total map size is greater as normal, and it is more useful to optimize the Huffman coding of the complete block. • Since all the most relevant bits in the complete block have been picked to remap them, the remaining bits to be remapped in the last block are less relevant. Therefore, the entropy of the bit in the last block is therefore higher than the entropy of the bit in the complete block. In other words, the Huffman coding of the last block is not 150832.doc *86· 201211508. Huffman coding such as full block is efficient.

Labghmn s tiyzEj cdi j opuv A BP|e flcl q rwxC DG 丨Η I + Jie poor Zen far media; sales articles Liang Yi black a% articles articles. + articles media articles; "box articles SK media: black miscellaneous Chapters Jiejie Xirong and + a piece of food: media. Media. Miscellaneous. The box is sold in black; ^ soft articles. Parents +_ - should remember, for all the complete blocks of all districts Use the same Hoffman tree. In the above example, the coded bits are thus encoded with the same Huffman code. • All complete blocks, and finally the last block is encoded with a different Huffman code:

(using 3 complete blocks of the same Huffman code) (using the last block of another Huffman) When seeing the above figure, the method remaps the first bit of each block ( Instead of remapping all the bits in the first block before jumping to the second block, the reason should be clearer. Since the same code is used for all complete blocks, it is necessary to make all the codes of all blocks as identical as possible. If you remap all the bits in the first block and then remap all the bits in the second block (and so on), each block will have a very different pattern: the first block will have the most The associated bit, the second block will have less relevant bits, the third block will have more irrelevant bits, and so on. It will be possible to generate a number of Huffman codes for each of the complete blocks, but it is believed that the cost of memory is too high. The method outlined so far works well while having all of the complete blocks share the same Huffman code. Possible Other Optimization Pages 150832.doc -87- 201211508 : 兀 Vector Pages Save a Difference Field, which is used to find the starting offset of each bit vector page. Insert (r〇ad near mterpolat〇r) to store the difference 榀] bit. However, the partial roads (4) =, because the small node ID (level 〇, highway) (4) 2 requires more bits to the bit vector around the $ node ID (level 5, destination road). Bit, encoding of the s page information is not as advantageous as may be desirable because the interpolator cannot accurately predict the actual offset. By modifying the interpolator, it will be possible to improve the encoding of the bit vector page information table to make it more efficient. Some embodiments of the present invention may use a bit vector page (and possibly a zone page) having 2 levels instead of only (4) levels. A page having 2 nodes (called a sub-page) for grouping data can be grouped into pages having 2, solid pages. Thus 'linear interpolation parameters will be stored page by page rather than globally (in the header). For example, 'index level 2 may have previously grouped 24 = 16 nodes into sub-pages and index 1 may group 210 = 1023 sub-pages into pages. Then for 1〇24xl6 = 16K nodes instead of all the nodes, the point count (4〇, delete, _ nodes in the map of western_and-Central_Eur〇pe (Western and Central Europe)) is linearly interpolated, so The linear interpolation of the variable size offset becomes much more accurate and the difference block in index 2 is therefore smaller. In the case where the page is large, the size of the extra index 1 is small (small enough to fit into the memory). It is advantageous to be able to fit into the memory of the device as it should not slow down the path selection using the data. Instead of storing the average page size for the per-input in the index i table, instead of storing the average page size, 150832.doc •88· 201211508 is stored in the header. 1 page with 2 subpages

Index 2 subpage with 2Λιι nodes

I page 0 I +----------+

丨页1 I +------ + i 12 1 +-------10>+--------- | 2Λη starting node variable size information +- --------- Ο + I Page 256

+ i page 257 I

Interleaved Bit Vector Page Road 2 Non: Offset Interpolator is inaccurate because important

The standard and secondary roads (many ordinary flags) are different inward. A simple way to make the interpolator more linear is to interleave pages of different network level 4, and this can be used in the implementation of the present invention. The files described in the above embodiments store the following pages. #0 #2 #2 #3 #4 #n-3 #n-2 #nl It is possible to store in an interlaced manner in other embodiments that may be more effective. .doc -89- 201211508 Save as follows: #0 #π-ι #1 #η-2 #2 IFH —3 #3 fn-4 To access page #X (for example by route planning application), by loading Page #χ' accesses the page, where: • Χ' = 2χχ (where X is an even number) • X'=2x(x-(nl)) (where χ is an odd number) This embodiment should be advantageous because It will reduce the size of the index by 7L per page. However, the data may be poorly grouped for the cache, this

May slow down data access (less hits in the file system cache memory card). The zone ID of all nodes is not stored. It is not necessary to store the node of the road without exit and the zone ID of the node with 2 attached road segments. For path selection, these nodes can be ignored. One of these nodes can be converted into entering its neighbor decision node. Store additional information at the page level and/or node level to view the map data. There are many TLs that are only included in the ordinary bit vector (or 11···11 1 to the S page. - Some embodiments may be for each Page, save 1 bit to mark their pages' to store the bit vector in their pages; more efficient, because for each - '一甘 & Λ 兀 兀 才 only need a single bit It is not 〇〇〇...〇〇〇 or i n. lu. It will not only reduce the size of the page containing only the flat/bit, but also the non-trivial bit 150832 .doc • The Huffman tree of the bit vector code on page 90-201211508 is better optimized (because the frequency of these codes will increase significantly in percentage, so the number of bits used to indicate nontrivial vectors Will be reduced.) In finer network levels (eg, level 3), most pages contain only trivial bit vectors, so there may be only 1 bit vector per page in about half of the pages. Store only the bit vector at the decision node Ο Ο As discussed above, some embodiments may not A bit vector of a node having (4) attached road segments is stored. However, other embodiments may promote the idea to store only the concept of a bit map to a w map node and a decision road segment around the decision node. Because it can be advantageous in coding the land θ θ, private: there is no need to encode the position 70 of the non-decision road segment, as discussed now. Two: 卽: is a node, where there is an entry for path selection The choice of the non-decision node that leaves the node (without making a U-turn) is not the node of the decision node. Taste - where is the choice, there is always only $ s road (10) • whistle, money,怿The road that leaves the node. 殳, 殳 is the method used to leave the decision-making node, Zeng defeat # • The non-decision road segment is #非,五... The way to stop the slave. All decision roads = not the road to decision The road segment of the segment. Therefore, all road segments around the decision node are ^ ° but not all road segments in the decision section are non-decision road segments 又 #又. Around the non-decision node: only, flat code decision Road e position The position of the non-decision road segment is 150832.doc 201211508 The vector is implied (〇〇〇... 〇〇〇 or 111...111) because the path selection technique using the data can be based on the existing road Determine the information in the segment. How to determine if a node is a decision node? Reduce the criterion to: isDecisionNode = (lineCount >= 3) && (lineGoingOutCount >= 2) where: lineCount: is attached The total number of lines to the node ignores the type of line (railway, reference line) where path selection is not possible, ignoring lines that are closed in both directions and ignoring prohibited roads (residential areas). lineGoingOutCount: The number of lines that are legally selected to be attached to a node for leaving the node. Select a road segment to leave the node as a legal sight segment attribute: • Road segment type (railway and reference road segments are always illegal) • Road forward/backward flow (stored in road segment flag) • Prohibited Passage Attributes in Road Segment Flags (Prohibited Passage Road Segments Do Not Have Any Bit Vectors) In some embodiments, ignoring non-decision road segments may discard approximately 40% of the road segments. It has been found that this percentage is fairly consistent and not related to the map. It is advantageous to avoid encoding a 40% bit vector, but it saves less than 40% of the map size because it primarily removes the trivial bit vector. Removing a bit vector around a node (dummy node) with less than 3 attached road segments removes the non-trivial bit vector, so the map size savings for non-decision road segments of this category are comparable to for non-decision road segments The map size is large. On the other hand, the screening requires a decoder (such as using the path selection method of the map) to decode the road segment type of the road segment and the road 150832.doc -92- 201211508 flag & and apply logic to it to clarify the implied Bit vector, which slows down the process. 〇 ‘ Real examples can also look at the movement (ie, the turn limit) to determine whether the road segment is legal, but this technique adds complexity. It is neglected that the embodiment may encode a more bit vector than the strictly necessary bit vector' but achieves a simplification of the method. Examples of non-decision nodes 匕 In the example, (b) is attached to 2 road slaves that can pass in both directions. (b) Not a decision point because there are <= 2 attached road segments. Therefore, the bit vector of the two road segments leaving node (b) will not be encoded. The decoder can implicitly set it to completion. A—>—b—The arrow > in this example shows the legal flow direction. (b) Not a decision point because there is only one exit. So 0) all road segments around do not need a bit vector. The bit vector of the road segment (b)_>(c) leaving node (b) is not encoded, which will be implicitly 11 . Nor does it encode the position of the illegal road segment that leaves (b). It will be hidden. The ground is 〇〇〇..〇〇〇. Example of decision node a—<->-b__<_>__c 150832.doc -93- 201211508 (b) For the decision section, point, because when there is a choice from u & (d), there is a choice : The path partner continues toward (a) or toward. Note that in this example, when the clip can only be looked up, there is no choice: path selection ^ heading (c) to continue. Make node $ to white M, R main 卽 (8) is still the decision point · point, because there is at least one choice from (d), it should store the position of the two road segments around node (b) Meta vector. Its::::::)TM segment (b) does not store the road segment (b) _> (d), ^ is located at the direction 'because it is flowed according to the direction of traffic/direction, select this road〇〇〇 ...〇〇〇〇 "The bit vector is logically or assumed that the node has 3 attached roads p ^ amps, and the first 2 decoded road segments have the following bit vector: OOOQQOOQQQOOOQOOQqqq

aOOOOGOODOQDOOCOOOOOC The third bit vector does not need to be encoded, as it can only be: 1111111111111111111.1 This can only be used when the bit vector of the road segment of the node happens to be in this order: all bit vectors are 000.·000 and The last _ bit vector is 111...111. In practice, it appears that it occurs quite frequently in a finer level (e.g., level 0) network (which is where most road segments are located). Take the first 2 bit vectors as: 00000000000000000110 00000000〇〇〇〇〇〇〇〇〇〇!〇150832.doc -94- 201211508 except for the 2 bits that are unknown and need to be encoded in some way All bits of the third bit vector can only be set to 1. 11111111111111111??! Since most of the bits are known in the above example, it should be possible to use this information to encode the 2 unknown bits more efficiently than encoding the entire bit vector. Therefore, in this example, only 2 bits must be encoded. A possible fast decoding scheme using this property would be to calculate the bit mask of all decoded bit vectors in the road segment of the current node by logically ORing all previous bit vectors in the road segment. In the case of using the same example as the previous example, if a node has 3 attached road segments, and if the previous 2 road segments have the following bit vector: D0000000000000000110 ODDOOOOODOQOOOOOQ010 ... then the logical or bit mask is : 00000000000000000110 The third and last bit vector to be encoded around the node is taken as: :11111111111111111001 The encoder can freely encode any other code (otherCode) as long as the following conditions are met, instead of encoding the 111 111 1 111 1 11 1 111001 Huffman code (which can be rare): value_to...encode = -bitMask | otherCode In this example, otherCode=0000000000000*0000000 meets the requirements because: value_t〇_enc〇de = IllllllllllllllllOOl - -0000000000000* QG〇〇11Q | 0000000000000*0000000 The code 000000000000000000000 is much more efficient than the code 11111111111111111001 150832.doc •95- 201211508 because 〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇〇 is much more frequently. Decoding is fast because decoding only needs to compute a bit mask (logical OR operation) whenever it decodes the bit vector and apply the bit mask to the last decoded bit vector: Actual bit vector = Bit Mask & Decoded Bit Vector ==00000000000000000110 & 00000000000000000000 = 11111111111111111001 Where the road segment around the node is stored such that a road segment with a maximum of 1 in the bit vector is at the end zone, this The optimization works well. However, this can prove to be difficult: • Perform a classification of the road segments around the node before calculating the bit vector (unless it is re-encoded using the bit vector information by a posteriori) • has been based on the road name Classify road segments. It is possible to classify road segments around a node when the road name is identified. There should be a strong correlation between the bit vector of the road segment around the originating node and the network level using the network level: the path selected to the other zone will generally prefer to select a coarse level network (1 level will usually be a rough level) Network, and 0 for finer level networks). The following example depicts intersections with road segments at different network levels: 4 Lines (2)-> (1) At Network 4 (Important) Lines (2)->(3) at Network 5 ( Minor) 1---=2 == -=-3 Line (2)->(3) At Network 5 (minor) 150832.doc -96- 201211508 The bit pattern is likely to be as follows: 2 1 ???????????? 2 3 ???????????? 34 000〇〇〇〇〇〇〇〇〇 is not stored in the node of the finer network level In addition to being close to the destination or starting point, the path selection almost never goes through an unimportant road segment network level (such as network level 5). Therefore, it is possible that the bit vector at level 5 is not stored, and therefore, the influence on the path selection should be minimal (even if the number of road segments at level 5 is large). In most cases, the path selection will only explore a few road segments at the least important network level at the beginning or end of the route, as it will quickly reach some of the more important road segment network levels, and The search acceleration data about their nodes will almost always tell the router to skip back to the least important network level navigation segment and use to or continue to stay at the more important network level. Navigation segment. This is true at the most important network level 'because it is used at the destination level when the destination is still far away. 〇 In addition to discarding the bitwise vector at the least important road segment (eg, level 5), the zone ID at the network may not be encoded. Converting a small number of zeros in a bit vector into 1 is also a lossy compression scheme. If the bit vector contains only one 〇 (or possibly, less than - small threshold), it may be possible to convert it to 丨 if the following result is produced (ie, set the road segment to Part of the fastest route): Ordinary bitwise Xiang (because the ordinary bit vector is encoded in a compactr way than the coding non-trivial bit to 150832.doc -97·201211508) • or already The bit vector found in the history of the tilde vector page (because their bit vector is also more efficiently encoded) converting the bit to 里 in the middle does not affect the path selection result, it will only be due to Make the path selection consider more road segments (because it is now set to be part of the fastest route) and make the path selection slower. However, the second embodiment of the present invention can use this method - especially in the case of a large map savings, while at the same time having a performance impact on the path selection speed (the pedormana is smaller. The path selection map 18 does not map 1600 'its The coverage-area has a start node 1602 and a destination node 1604 and a plurality of road segments that have been explored by the prior art A* search method. The selected route is displayed by the dotted line 16〇6. Figure 1-9 shows the map 165〇, It covers the same area as the map of Fig. 18 and 'shows the same start_point! 6〇2 and destination node (10) 4. This figure also wakes up and mentions the road that is not explored using the embodiment of the present invention, which utilizes The same road network and the same criteria as those used to generate the route of Figure 18. Both of them travel by car, hoping to use speed as a function of cost to minimize, Rong Wang:, $, - people The dotted road section 1 606 shows the roads selected by the method, the temples, the temples, and the road sections are the same as the road sections calculated in Fig. 18. I ^ By way of an embodiment of the invention The route ° ', Jin Ming's mathematically accurate, and can find the best / best route for the model that has been selected with the minimum, σ cost. Therefore, it should be obvious that the sue > When the field is compared with the prior art, the way of exploring the embodiment of the present invention is 150832.doc -98-201211508, which is less obvious. Therefore, compared with the prior art and the method, the present invention The method of the embodiment is relatively fast, and usually significantly faster. It should also be noted that 'as the route 16G6 approaches the destination 16Q4, the route with the starting home is searched; that is, when the route "(10) advances past the point "n" The temple is to explore the road except the lowest cost route. A solution to this situation is that the method has been self-contained at point 1652 and switched to the level! area. It should be seen that it is 'exploring other road segments near destination 16G4.再 Γ Γ = = By explaining that the route planning method will switch from level 1 to level 2 to explain this situation. Therefore, when making long-range path selection, the speed route., "With the method in- The fastest nt η $ , , ; Levels (for example, self-equal, and 〇 to 1) may exist as "most, explore more than one route. 帛 low cost" and therefore need to be available 2:::: has been generated and the bit vector is calculated And store it, then its J is used to calculate the route between two points. Dantu 20 shows the method by route planning, where the at-node is made to advance; = the previous technology of the road. In order to achieve this , the decision to move to the next two road segments - the estimated cost of moving the node to the target. Then add -I to the custom path with the least cost. These methods, self-sufficiency, point line has been incurred so far The sum of costs, and the cost of keeping the approach small. The mother-reverse is considered to be, for example, starting at the node, where – Β (starting node) is 2 cost units apart. Knowing points C and D are both ~, starting from node D, reaching 150832.doc -99- 201211508

This is 3 units, and from the node ❻, to the node R unit. Therefore, the method will be compared to the road segment that explores the route to the node D more preferentially than the node C, and the 'to the next node' is too strong 5 is the estimated cost for the target to the printing point. Comparison (ie, B〇2+5 = 7, and BD=2+3 = 5). The minimum cost bit vector of the embodiment of the present invention is used to perform the segmentation of the segment of the road segment (ie, Figure 7〇4) to identify the portion of the minimum cost path from the road slave to the destination area. Candidate. Thus, the method described with respect to Fig. 20 is modified to be shown in the figure. \ Typically, route planning is performed using a bit vector, as described in relation to Figure 。. However, some embodiments of the present invention may allow the route gauge d to be returned to Α (or other prior art methods) under various circumstances. Different embodiments may use any of the following: missing bits to 4 information ( thereby allowing the system to function even in the absence of a bit vector); using a cost function; in addition to calculating for it Path selection outside the cost function of the bit vector (usually the fastest route, but not necessarily); the user specifies criteria that are different from their criteria for pre-computing the bit vector (eg, the user wants to avoid the highway) ); the type of transport is different from the type of transport used to pre-calculate the bit vector (for example, the user walks without taking the car); and the route is below the threshold length (for example, the start and end of the route are in the same area) . When compared to Figure 20, two additional numbers 18A, 1801 are shown on Figure 21. In addition, on the figure, the dotted road segments 丨8〇2 and 丨8〇4 are used to indicate 150832.doc -100- 201211508

G Ο area. "1" 1800 indicates that the road segment BD is part of the minimum cost path from the node 6 to the zone in which the target node is located. "G" 18G1 indicates that the road segment BC is not part of any minimum cost path from the node B to the zone in which the target node is located. Thus, when starting from node A, embodiments of the present invention will use the m outlined above to explore nodes from a. However, once the node has been reached, the exploration is used progressively because there is a clear indication that only the road segment BD is a possible part of the least cost path. Thus, once the A* method has been used and the least cost path is found, it will then select the least cost path by looking at the associated bits within the bit vector. However, if there are different minimum cost paths between a pair of zones at different times, then the minimum cost profile can identify more than one minimum cost path between the zones. Therefore, the path selection between the starting node and the target node (starting point and destination) may find more than one minimum cost path, and must choose between the minimum cost of competition, for example, the number in the road segment BD is deleted. The number of positions in the segment BC (4) indicates that each A 4* -ΐ* α* ΒΒ ^ θ . . . is the value of the portion of the minimum cost path at a specific time. j In this embodiment, the position vector does not identify each _ road segment. Part of the minimum cost path (four). Therefore, the path selection algorithm performs a cost analysis of the two possible least cost paths based on the time traveled on the segment of the path. This time can be determined based on the departure time from the start/start node. The speed profile is associated with each segment (as shown in the figure), which is about how the expected speed of travel through the segment changes over time: Based on these speed profiles, the expected speed at the relevant time can be determined, the plant

In the case of J 150832.doc 201211508 determines the cost of traveling along the road segment at that time. The cost of the parental small cost path can be determined by the cost of the road segment of the parent-minimum cost path. The t-shoulders of the right-handed stalker can then choose the route with the lowest cost path to take the low cost at that time. The determined route can be displayed on display 206. _ How the route is not visible on one of the map materials. In this embodiment, the display includes an indication of the expected speed on at least two of the navigable segments of the route. For this embodiment, these are: the time series for the color of the route and the side of the image of the map are not the expected speed along the route. For example, if the expected speed exceeds the first threshold, the navigable segment can be colored to a first color (eg, green) and if the expected speed is below the first limit, the navigable segment can be colored Is the second color (for example, red). An alternative method of determining a route will now be discussed with reference to Figures _ and 24 through 31. The cost of representing a cost that changes over time can be determined rather than determining the cost of the route at a particular time. In this embodiment, multiple values of the velocity profile of the parent-navigable segment, rather than a single value, are used for the cost profile of the Shihuoding route. For example, referring to Fig. 21, the cost per route is a distribution of how the cost changes over time, rather than having a single-turn of each road segment Bc, ce, and the like. These cost distributions are summed to determine the cost profile of the route. The distribution of cost profiles via trees is also shown in Figure l〇a. In this embodiment, 'each navigable segment has a velocity profile associated therewith that contains 12 time bins per hour, per-time group 150832.doc -102 - 201211508 containing at some time The expected speed of the shackles. The cost of the expected speed for each group is determined, η g, and the cost of each relevant group of navigable segments of the route is summed to determine the cost profile. The sum of the maternal slaves may not be from the same day-to-day group, as the expected time to travel along each segment will vary depending on when the user is predicted to arrive at each-navable segment. For example, it can be predicted that the user will spend 5 minutes to travel along the navigable section BD, so the acupuncture and BD and DF summed to the group value of -& can be a group value of $ minutes apart, also ρ The value of time group τ of Bc is added to the value of time group τ+5 of DF. Search only for navigable segments that are part of the least cost path, without exploring and

A competing navigable segment of a portion of the non-minimum cost path. Therefore, for both paths BDF and BCEF to be considered, the two paths must be identified as the least cost path by the least cost data. If the two paths BDF and BCEF are the minimum cost paths at different times, then at the node F where the two routes intersect, the cost profiles of the BDF and BCEF need to be preserved to some extent and may be further propagated. In order to achieve this, the processor performs the calculation of the two cost profiles (C丨 and C2) on top of each other and determines the upper bound distribution UB (shown in solid lines) and the lower bound distribution LB (shown in dotted lines). This is shown in Figure 30 and Figure 31. The UB and LB distributions are then used for further propagation, allowing for the determination of the maximum and minimum cost profiles for each route. Further propagation of the upper bound distribution and the lower bound distribution occurs in a manner similar to the cost value of subsequent navigable routes being added to such profiles. If the routes being explored are separated, and then intersect at a node similar to node F, then by overlaying the two upper bound profiles of each entry path to determine a single 150S32.doc • 103· 201211508 new on The two upper bound profiles are again processed by overlapping the two lower bound profiles of each entry path to determine a single-new lower bound profile. Hai two lower bounds overview. In this way, the increase in the number of profiles is maintained at a level that does not unduly obstruct the painful route or does not require a general (because the memory on the PND is limited). ° μ In another embodiment, only the upper bound profile and the lower bound profile are retained for further propagation along the route. In another embodiment, determining an average of the two entry profiles to generate an average profile for further propagation along the route 0 - once the cost profile for the entire route has been determined, then the upper profile or the lower profile is The person chooses the final cost profile, which is used to determine the display on the navigation device, as will now be explained. Figures 24 and 25 illustrate the manner in which multiple values of the cost profile can be displayed. In Fig. 24, the cost profile is illustrated by displaying the departure time from the start point and the time interval from the destination to :: and separating the predetermined amount (15 minutes in this embodiment). Similar to the time column 2_ in Figure 23, the 'every-time column is color coded to indicate that the speeds two along the different parts of the route provide arrows 2〇01, 2°°2, when the user selects the 1st day: 'make The display shows the pair of departure time and arrival time, and the time column of = or later start time (select left arrow · make display ^ start _ ' and select right arrow 2 (10) so that the display time column can be selected later to make the display select The display area between the opening materials (similar to the one shown in Figure 23). In Figure 25, the cost profile is displayed as a set of time periods (in this case, 150832.doc -104·201211508, in (10) A continuous graph of hours (equivalent to one week). This Z map is displayed in conjunction with the @ % graph, and the map description helps to form this ❹: brother: the route from the second starting point to the destination. And route color horses (as illustrated by the line and dotted lines) so that the user can determine when: which route will be used. For example, a plot of 7 hours 4 minutes and 9 small buckle IS II Coloring to show at the beginning of the route 'should Follow the dotted line = win instead of the first part of the dotted line line 2004. This display allows the user to easily see how the cost of traveling between the starting point and the destination changes over time and how this affects the route. An alternative embodiment is illustrated in Fig. 28. In this embodiment, it has been determined that the present profile is as described above, but only images of the route determined for the Z time specified by the user are displayed. , ^ The date and time when the route not displayed can be applied. The user can select the day/day of the day when the image of the route is not determined by the user: the arrow on either side of the width changes the time and/or time. Changing 曰 and/or at: may cause a route change, and the display is updated to display the new route and = any time spent traveling on the route and any distance traveled - as shown in Figures 27 and 28 An example of an updated display. / Immediately (eg, on the order of milliseconds) occurs when the display is more 'eg' less than milliseconds and preferably less than (10) milliseconds. Information and/or prisoners are additional functions for the use of the navigation device for the use of the profile search for these other time-based routes that have been searched by the profile and to achieve such rapid recalculations. -105- 201211508 = 'When a journey time (eg, start time) is selected, the processor of the navigation device can compare the cost of traveling at that time with the window around the time of the trip (10), for example, 30 minutes of journey time) The value of the cost profile. If the LUC finds that the journey time within the window has a lower cost, the processor can = display the indication of the journey time giving the better result. By way of example, in the illustrated embodiment, a note 2〇15 is displayed, which notifies the user that if he or she travels after 15 minutes, the travel time will be shorter (in this case, 1 short) minute). To perform the path selection described so far, the bit vector information is obtained from the encoded side standard described with respect to Figures 11 through 13 using a decoding process. The output of the pre-processing described above includes: • Hierarchical zone assignments for most nodes, and • Pixel vector assignments for outbound road segments at these nodes. Map Loading Consistency Check When loading map data, the collection of side files should be presented in the map directory. Otherwise, the decoder will undo the search acceleration data so that the route search can be performed without search acceleration data. There are several checks that help ensure the integrity of the data, which are now listed. Side files follow a naming convention. The number of levels is given by the number of side files, but the number of levels is also stored in the header of each side file and equals the number of side files. The number of areas on each side of the file stored at their respective levels is the same as the number specified in the side name of the file 150832.doc -106- 201211508. In addition, the side rights file stores the following information (which should be the same for all side files): • Side file version. The number of nodes per "bit 70 vector page" (explained below). There is also a sum checkmark in the parent file, which identifies • a specific entire set of side files • the entire map. For the associated side file of a given electronic map, such information should be correct. If any of the above checks are lost, the search acceleration data feature will not be launched for this map. The data structure loaded into the memory The denier horse reads the side standard in the implementation of the external memory. This means that the bit vector: the content of the second floor case is not completely loaded into the memory, but only according to the needs. However, 'some general information is loaded into the memory at the beginning' and as long as the map is loaded, the data is kept in the middle. ^ Header Information

U Each side of the file begins with a header section containing the information described above with respect to Figures 8-10. This information on each side of the file is stored in the memory application. Huffman Tree Definition The side trough case contains the Huffman tree. The per-Huffman tree definition gives a complete description of the specific Huffman coding and is later used to decode the side file m π into the original data (ie, the bit sequence from the side standard) Decode into a number or some other specific value). From the per-side standard 150832.doc •107- 201211508 Read the following Huffman tree definition and keep it in memory. • The Hoffman tree used to decode the road segments at each node. (The number of coded road segments can be smaller than the number of road segments in the base map. Note that the Horn Coder is independent of the base map, which means that it does not need to read the map.) This Hoffman tree is only stored in the file on the side of the temple, and is used to decode the Huffman tree of the page (explained below). • Several Huffman trees for decoding bit vector blocks. The number of tree definitions is given by the length of the regular bit vector block as stored in the side file header; it is equal to the block length minus -. (The entire bit string of the bit segment of the road segment is split into blocks. If the length of the bit vector string string is not a multiple of the regular block length, then the last block is shorter for the rule from 2 until the rule For each block length of the block length, there is a Huffman tree. The Huffman tree is not used for blocks of length _ because this is only one bit and is stored directly). • Several Huffman trees for selecting the decoding method of the bit vector. Specify the number of such trees in the header; explain their use below. Neighbor List The side slot encodes the neighbor list at each level except the fine level. The list of neighbors in the zone at level k is zero or more than the list of zones (four)) known as rank k R neighbors. The neighbor list section of the grade k case has two parts. The P-knife 3 has the number of neighboring areas per sub-level k sub-area. For example, °(k〖)' right has 4000 zones at the global level 0, and each - 150832.doc -108- 201211508 global zone is subdivided into 10 ranks, then the level file contains 4〇〇〇χΐ A level of scorpion area. For each of these sub-zones, the length of the individual neighbor list (which consists of level 2 zones) is given. • The second part contains a list of actual neighbors, each of which is known from the first part. Each neighbor list is a list of rank (k+Ι) subzones. Area remapping table

The side slot case stores a bit vector in a compressed format by coalescing and reordering, and the like. The class side tile has a zone remapping table for solving the bit vector of the block code. It consists of two parts: The first part is an array indexed by level k sub-areas. For the level, • 卞 ^---- in the < section: no. This is related to the outgoing road segments of the nodes in the respective sub-areas of the block-coded bit vector. The second part is a two-dimensional array indexed by (1) the bit position in the uncompressed bit vector and the (10) level. The array entry fingers are: the position of the location element in the uncompressed bit string read by the block in the compressed bit vector. The two parts are only in the solution only at the global level. Note that only the first part is kept in the memory; the first period is used as needed because it is large. Currently, the storage file is remapped at the side file (k=〇). Start route query The decoder is used to speed up the route from the ^ ^ ^ 夯 location to the destination node set. (The destination is defined by an instance of a point-by-point point that defines the entire road path as a destination, or a collection of multiple targets 150832.doc -109-201211508. The target area is a sequence of area IDs, each partition level having a zone ID (starting at level 。). At the beginning of each new search, by clearing the previous target zone set and passing the new destination node to the decoder A set of target areas is created. The decoder will determine the list of target areas that are not repeated. The mechanism for finding the area of a given node is the same as during the search; see below for details. 0 Scanning nodes during the route search Set the set of target areas. The decoder is characterized by the segment of the road network that is provided to return the road segment that leaves the node: the indexed bit array (this node and the given person's mouth E The function of the position vector. The order of the road segments at the nodes and nodes is defined by the map. Whenever the value of the bit is zero, this means that during the search, the corresponding road segment is ignored. (This usually leads to the search. A large amount of money V and therefore a large reduction in the operation of the search.) For a few nodes, the area is 你 _ _ θ 匕 a a 或 or a bit 疋 vector data are not: in. For these nodes 'decoder Returns the = string of all bits. (This will prevent the route search from skipping any road at this node = right, ''疋' is the status of the Brin ask for the month b. In addition, there is an indication given丄 ie,··············································

, & ― 幻 功. Returned by the decoder in the directory F

The bit vector of the node in the middle _^4 J隹目拾Q Calculate "this... is a bit string of all bits 1. This 荨 查 δ δ δ function is used in the best of the road ridge green search brother According to the side standard format, the bit vector of the given node is decoded according to the + half-filled right-hand step. 150832.doc -110. 201211508 In the per-side file, the information of the area and the bit vector is organized into so-called ' The number of nodes of the page is a fixed power of 2 (for example, 16), and is the same for each side _ and outer j file in the side file set. This means that for a given . The bit shift is used to calculate the page index of the bit vector page. The information of the point is stored continuously by the node ID. • Find the page offset of the given node for each side, prepare a hundred + a - Λ In the header, ί is stored in the side of the 被 eight is used to multiply the page by the average size and the nearest two is the byte offset. The correction is stored in the index by page. In the table of indexes. When this table is stored in a separate section of the side slot, look for the correction in the table and add the correction to the approximation. The page offset page thus gives the position of the page in the side file. Decoding page cache When the first query page is decoded, the decoding (relative to the bit γ of the node in the fixed page) is fast. ° ± The self-deprecating-page query node's Bellow Temple' can use the cached data without any side-slot file. =Decoded bit The special tag bit in the vector bit string is used to remember Whether the node is a no-information node. - The page number specifies the same area ID for each page in the so-called page number bit. The page number contains - only the element for each level, and the level = the level. The beginning of the page is stored as a common Hoff 150832.doc -111- 201211508 Decoded outbound road segment count as mentioned above 'Each page contains a fixed number of points 鍅 six from a point of the tribute. The thrift is stored in the header of the parent side file.— (or for level 0, + I—the page begins after the page number), which lists the number of road segments out of the page t. Whenever the road segment is out This means that the number of shares of the corresponding node is not stored at all. In the temporary array. When the page is known, the decoding area is given by the zone segment column after the road segment counting section, and there is one _ per level. The specific equal = zone 1D sequence corresponds to the side file. m is stored in the zone. When all nodes of the unfinished node have stored information on the nodes at all levels (ie, zeros), the 'supplement of the zone's % pills' section is counted as beta. Read the zone ID sequence of the brother-node of the road segment. If the given level is greater than the order level, the new female r ", the page of the page specifies that all the zone IDs are the same 'is all at the level The section is the same (four). Otherwise, the reading has a positive road. At the end of the process, the solution of the solution = point (where the sequence can be a null decoding bit vector to find the set of relevant bit positions for a given node and the target area, the special - leave this The value of the bit vector bit of the road segment of the point. The above target (4), there will be a starting or a series of operations. For each I50832.doc -112- 201211508 one node, the decoder side, - 彳舁A set of related bit positions. For a parent-out road segment at the node, such as - the relevant position is the same as the set of positions. It depends only on the set of & and the target area of the node 2. If there is only one target area , then there will be only one relevant bit position at a level; in other words, for this node, the information stored at other levels can be ignored. In the case of more than one target area, one; ;: Nissin A - II The position of the fl position can be coincident, so it is always possible to store at most the position of the relevant target area - 楛 & In the following, it will be described how the decoder determines the position of the associated bit of the _ sleeve g 4Φ fields. For more than one target area, the relevant bit positions are found in the same way and the related bit positions are combined into a set. When the neighbor zone is not defined, there is one bit vector bit (per outgoing road segment) for each target zone at each level. (For the sake of Jian Cui, ignore the fact that (10) does not store bits for the node itself). The target bit is at the first-level (counting from level 0), where the node (1) is different from the area 1D of the target area. For example, if the zone ID of the node at the global level is equal to the target (four)' at the global domain but the two zone IDs are different at level W, then the relevant bit location is at level i and is equal to the target zone ID. The bit position in the uncompressed bit vector string at the level; this string contains one bit for a possible target area 1D. The zone remapping table is used to convert this location into a location in the compressed bit vector (in the actual string of the class). The finest" and so on, and the assumption that the node defines the neighboring zone, then the relevant bit is determined at the level (in the neighborhood where the target zone is the zone of the node). For the example, suppose the target area ID sequence is (a, b, c, sentence, 150832.doc -113- 201211508 area ID sequence is (e, f, g, h). If (a, b, c, d) For (e, f, the neighbor (as defined in the neighbor list section), and (a, b) is the neighbor of (e), the relevant bit is determined by (a, b, c, d) and located At level 2, that is, the level of the (e, f, g) first code as the neighbor's (a, b, c, d). More precisely, the relevant bit position is the area in the level 2 side file (e , f, g) neighbor's (a, b, c, magic index. In this case, the rule bit h of the zone ID (as explained in the previous paragraph) is irrelevant. Again-time, this correlation The bit position is relative to the uncompressed bit direction. The decoder causes the material 4 to be newly mapped to convert it to the bit position in the compressed bit string at level 2. More information about the neighbor list Contained in the file "Promotion of enhanced multi-level graph partitioning - for enhanced multi_levelgraphpartiti〇ning). Decoding bit vector bits in the case of a fixed set of target regions, each section of the page The 2-yuan vector of the point will be composed of the bits of each outgoing road segment. If the road of the node: twenty zero is zero (that is, the node is no information node), each bit will be reduced to 1, For one of the target areas, if you do not enter one of the target areas, you will be able to decode 7/day. The right node is located in the ^ item "A, then the position 兀·向向盎 into αγτ 1 .丄』 will be all 1 ; In this case 'may have to skip the encoded data in order to decode the data of the subsequent nodes. 0 If the current node is 亓 and 4 τ is not trivial, then the decoder confirms the set of corresponding bit positions, ia , ^ ^ ^ 々 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早 早The bit/string is reduced, but the decoder only needs to be in the 疋. During decoding, the information that is not used by I50832.doc 114· 201211508 is read (skip only), but only after it is actually needed. In case)), otherwise, the unused portion of the bit stream is not read at all. The level is grouped into a set of related bit positions. The relevant position is converted into a position in the (four) metastring instead of decoding (four) abbreviated 4 and solving it, and % is buying the bit at the relevant position with f. # + ' A _ If there is a correlation 右 at a certain level on the right, then the first time (when the ripper remembers how far it is processed in the side file), skip the sequel of the predecessor node Ο 到达 reach the node being discussed in the file It: when on a given side, for each out-of-way—stores one bit; two = the result of a logical OR operation of the decoded bits of the associated bit position. Information on the specific road segment at this point is specified: · Start. It has the following values: ^^*1 symbol open • 7 all bit vector bits of the road segment at this level It is not necessary to decode other bits. ) said. For this road segment, • all the neighboring bits at this level are assigned to all the neighboring bits at this level. 70 is 1 (including the hand 5 tiger. For this road segment, it is not necessary to decode other bits. • Specify the bit vector The bit string is explicitly given by the block. The decoding of the compressed bit to the _ 曰 兀 按 。 。 。 。 。 。 。 。 。 。 。 。 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码 解码In the whole "history stacking" t. Stomach, which is established during the code period, • specifies the 2 Μ 且 且 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史 历史Meta-vector bit value. History 150832.doc • JJ5- 201211508 Depending on the current size of the history stack, use the same Huffman tree for the decoding method selector. There is a specified number of Huffman trees in the header t of the side file. No limit value; when the size of the history stack exceeds this value, Bayer re-uses the Huffman tree of the directory. The upper right method selector symbol specifies that the bit vector string should be explicitly encoded, then the decoder is block by block Read the compressed bit vector bit string. Collect the bit = value at the relevant bit position (relative to the compressed bit vector string) at this level. Perform a logical OR on the value of the bit. The logical or intermediate result: you can skip The remainder of all other blocks in this road segment: Bit 兀. The Huffman tree used to decode each block depends on the number of bits in the block. For the specified in the side file header There is a Huffman tree in the length of the rule block. The last block of the road segment can be shorter; depending on its size, the appropriate Huffman tree is used for decoding. The block of length one is only one. It reads directly without Huffman decoding. If the number of blocks is at least 2, then white. The bit stream generated by the self-position 兀 is contained in the Hoffman θ days of the blocks. An extra bit of the other. The value specifies whether the entire decoded string is resolved as a bitwise negation of the continuous value. (The purpose of this negation is better Huffman compression [pattern Brief Description] Figure 1 is an illustrative part of a navigation device that can be used by the AJ King 5 Garment System (GPS). Figure 2 is a schematic diagram of a server and navigation device that spans the communication channel and is set up; Figure 3a is a schematic diagram of the navigation skirt; 150832.doc •116-201211508 Figure 3b is a diagram Figure 3 is a perspective view of a navigation device; Figure 5 shows a portion of an example map generated by an embodiment of the present invention; Figure 6 shows a map of Figure 5, shown thereon Figure 7 shows a map of Figure 5 after processing; Figure 8 shows an example of how a node is assigned to a map in some embodiments; Figure 9 illustrates how a map can be segmented into An example of a plurality of nested regions; Figure 9a illustrates an enhancement of the partitioning of Figure 9; Figure 10 illustrates an example set of bit vectors utilized by embodiments of the present invention; Figure 1a illustrates how to utilize during Dijkstra exploration of the network Information on time profiled; FIG. 11 shows an example file format of a file according to an embodiment of the present invention. FIG. 12 shows an embodiment of how a region remapping table and a region list can be encoded; FIG. 3 shows an example file format of the coarsest nested area as illustrated in FIG. 3; and FIG. 14 shows an example file format of the nested area except the coarsest nested area; FIG. Figure 6 (which includes Figures 16a and 16b) illustrates the effect of coalescing bits; Figure 17 also illustrates the effect of coalescing bits; 150832.doc -117- 201211508 Figure 1 8 shows the use of previous The technical path selection technique is to highlight the map of the route in question; FIG. 19 shows a map of the route considered by the embodiment of the present invention to highlight the prompt; FIG. 2A shows an example of the A* search method (prior art); Modifications to the figures used by at least some embodiments of the present invention are shown; 'Figure 22 shows a flow chart summarizing the steps of the method; Figure 23 shows a display of the navigation device illustrating the expected speed on the navigable section of the route; ~ Figure 24 Displaying a display of navigation devices that illustrate the cost profile of the route between the start point and the destination; Figure 25 shows the cost of the route between the start point and the destination - or multiple routes Display of the navigation device; display; Figure % shows the navigation between the start point and the destination for the specific journey time and the control device for changing the travel time. The user has changed the update at the time of the journey; Figure 28 shows the update of the travel time display for the subscriber of Figure 27 after using _p, &; Figure 29 shows the navigation device time; Figure 3 The description is given to the user-recommended travel-series display, where the two points are at the cost of the two scented _ _ J J 眺肀 route cost profile 150832.doc • 118- 201211508 graph; and Figure 31 shows from Figure 30 The upper bound cost profile derived from the cost profile presented in the example and the lower bound cost profile. [Main component symbol description] 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 6 Zone 7 Zone 8 Zone 9 Zone 15 Zone 21 District 24 District 27 District 28 District 37 District 41 District 42 District 43 District 51 District 100 Global Positioning SystemΟ Ο 150832.doc -119- 201211508 102 Satellite 104 Earth 106 Global Positioning System Receiver 108 Spread Spectrum Global Positioning System Satellite Data Signal 150 Server 200 Navigation Device 202 Processor 204 Input Device 206 Display Screen/Integrated Input and Display Device 208 Output Device 210 Connection 212 Output Connection 214 Memory System 216 Connection 218 Input/Output (I/O) 埠 220 Connection 222 External I/O Device 224 Antenna/Receiver/Internal GPS Receiver 226 Connection 228 埠 230 Connection 250 Dead End 252 Arm / redundant ring 254 suction cup 150832.doc -120- 201211508

280 Functional Hardware Component 282 Basic Input/Output System 284 Operating System 286 Application Software 288 View Generation Module 300 Node 302 Node 304a Road Segment 304b Road Segment 304c Road Segment 304d Road Segment 308 Node 310 Road 400 Node 600 Region 602 604 District 606 District 608 District 700 Leftmost Barrier 702 Brother II Shed 704 Third Barrier 750 Node 751 Figure-121 - 150832.doc 201211508 752 Cost Function 754 Point 756 Point 800 Ground Icon Head 802 Hoffman Tree 804 Area Count/Zone Re Mapping Table 806 Area Neighbor/Zone ID List 808 Area ID Remapping Table 810 Bit Vector Page Index/Table 812 Bit Vector 900 Array 1300 Vertical Strip Area 1600 Map 1602 Originating Node 1604 Destination Node 1606 Selected Route 1650 Map 1652 Point 1800 Road segment BD 1801 Road segment BC 1802 Dotted road segment 1804 Dotted road segment 2000 Time column 2001 Left arrow 150832.doc -]22· 201211508 2002 Right arrow 2003 Dotted route 2004 Point line 2010 Block 2015 Remarks Cl Cost profile C2 Cost Profile LB lower bound distribution on UB Distribution ❾ 150832.doc -123 -

Claims (1)

201211508 VII. Patent application scope: 2. 4. A method for determining a route using map data comprising a plurality of navigable paths, the map material being divided into a plurality of zones, the method comprising using at least one processing device to: receive a starting point and a purpose on the map data And a journey time, using the map data and the minimum cost data of the least cost path between the areas identifying the map data to determine the route from the origin to the destination, characterized in that the zones are at different times Where there is a different minimum cost path between a pair of zones, the minimum cost profile identifies more than one least cost path between the pair of zones, and the route comprises the zone from which the origin and the destination are included The minimum cost path identifies the minimum cost path for the journey time with a minimum cost. The method of claim 1, wherein the minimum cost path is used to identify the least cost path for deriving the identified minimum cost path from the journey time, and determining that the route includes a minimum cost path independent of the time root pair And one or more relevant times and a cost analysis of the operations to determine a method of selecting a destination or a claim 2 at the time of the journey, which causes the minimum I-minimum cost path to identify the path The -costed and determined route for the least cost path includes selecting a minimum cost path corresponding to the derived from the 4 journey time - or a plurality of associated time reference times based on the minimum cost profile. The method of claim 1 or claim 2, wherein the map data includes a speed of the time of the time of the navigation on the navigable section of the material path 150832.doc 201211508, and at least partly according to the path The speed profiles of the navigable segments determine the cost of the path. 5. The method of claim 4, comprising determining the one or more by determining a cost based on a value of each of the speed profiles of the navigable segments forming one or more routes between the common starting point and the destination A cost overview of the route. The method of claim 5, comprising receiving an overview request to determine a best journey time between the far start point and the destination for the defined criteria and in response, determining the cost profile and displaying the result to The user. The method of claim 6, comprising causing the processing device to determine an optimal journey time for the defined criteria based on the cost profile, and wherein the result is not displayed for one of the optimal journey times. 8 • The method of claim 6 f, , earth, and the result of which is shown as a display of the cost profile. 9. The method of claim 5, wherein the cost profile request includes a time period, and searching for the route is limited to the time period. 1 〇 · A method of requesting an item, an item, comprising receiving a request for one of a route between a point and a destination at a particular journey time, determining a time window including a journey time of the other party a cost profile, and depending on the cost, the cost of the route at another time within the time window is: at the cost for the particular time, and if the cost is small, then: The user notifies the message of one of the other travel times. The method of claim 1 or claim 2, comprising receiving a selection of the time of the journey from a user via a user interface 150832.doc 201211508, and causing the display to display the map of the determined route One of the images. 12. The method of claim 11, the direct sentence evaluation, the deduction of the travel time, the user is able to change the travel time while viewing the image of the map material that is not determined by the exhibition. And in response to the change in the travel, the display is updated with an image of a new route for the changed travel time. </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> </ RTI> <RTIgt; A data carrier that, when executed by a processing device, causes the processing device to perform one of the methods of any one of claims 1 to 13. ' 15. A type of data carrier' that stores a plurality of navigable Map data of the route 'The map data is divided into a plurality of zones; and the minimum cost data identifying the least cost path between the zones of the map data at different times. 16· - Contains memory and processing equipment a computer device, the memory storing map data comprising a plurality of navigable paths, the map data being divided into a plurality of regions, and identifying a minimum cost data of a minimum cost path between the regions of the map data; The device is configured to receive a point on the map material and a destination and a journey time and use the map data and the minimum cost data to determine from a route to the destination point, characterized in that, in the case where there are different minimum cost paths between a pair of zones in the zones at different times, the minimum cost 150832.doc 201211508 data identifies the pair of zones One or more least cost paths, and determining a route includes identifying a minimum cost path of the route between the zone containing the origin and the destination based on the minimum cost profile when considering the journey time. 1 7. Navigation The device includes a positioning system, a display, a user interface, and a computer device as claimed in claim 16, wherein the processing device is configured to derive a starting point and a destination via input on the user interface and will determine The results are displayed on the display. 18_A method for determining a route using map data, minimum cost data, and a speed profile, the map material including a plurality of navigable segments representing segments of the navigable path of the map data; The data identifies the least cost path between the areas where the map is poor, and the time is in the areas at different times. Where there are different minimum cost paths between the zones, the minimum cost profile identifies more than one least cost path between the zones; and the velocity profiles of the navigable segments identify the speed on the navigable segments Varying over time, the method includes using at least one processing device to: search for "or multiple routes" from the point of view to a destination on the map data, the search comprising determining a set of connections to a node Whether the one or more navigable segments in the navigation segment are segmented by the minimum cost data for one of the minimum cost paths of the region containing the origin and the destination and one of the navigable segments of the group or In the case where a plurality of parts that can be identified as - the least cost path, the part of the group that is the part of a minimum cost path, or the plurality of navigable I and the root (four), etc. The speed is indeed 150832.doc 201211508 成本 One of the cost profiles of one or more routes over time. 19. 20. ο ο 21. A data carrier storing instructions that, when executed by a processing device, cause the processing device to perform a method as in claim 18. A computer device comprising a memory and a processing device, the memory storing map data, the map data comprising a plurality of navigable segments representing segments of the navigable path of the map data, and a speed profile of the navigable segments, The speed profiles identify how the speeds on the navigable segments change over time; the processing device is configured to search the map - one or more routes from the point of interest to a destination The search includes determining that one or more of the set of navigable segments connected to a node can be identified as one of the minimum cost paths of the zone containing the origin and the destination. Partially, and in the case where one or more of the navigable segments of the group are identified as part of the least cost path, only the portion identified as a least cost path is explored from the group One or more navigable segments, and determining one of the one or more routes cost profiles over time based on the velocity profiles of the explored navigable segments. A method for determining and displaying a route using map data, the map material comprising a plurality of navigable segments representing segments of the navigable path of the map material and a speed profile of the navigable segments, the method comprising responding to the user Determining, along the input of the interface, a route along one of the navigable segments, determining an expected speed of one of the navigable segments along the route based on the associated velocity profiles, and displaying the determined route and the pair The pre-150832.doc 201211508 22. 23. 24. 25. 26. An indication of the speed of the period on at least some of the navigable segments of the route. A data carrier storing instructions that, when executed by a processing device, cause the processing device to perform a method as claimed. a computer device comprising: a display, a user interface, a memory and a processing device, the memory storing map data, the map data comprising a plurality of navigable k and a segment of the navigable path of the map data a velocity profile of 4 voyagable segments, the velocity profiles representing predicted travel speeds along one of the segments at different times, the processing device being configured to determine along the input in the user interface to determine along the Navigating another route, determining an expected speed of the navigable segment along the route based on the associated speed profiles and causing the display to display the determined route and the pair of routes on the route An indication of the expected speed on at least some of the navigable segments of the navigation segment. If the computer is configured to cause the display time to be different, the user can change the journey time by interacting with the user interface while viewing the route. Responding to a change to the time of the journey' the processor determines an updated expected speed for the navigable segment along the route for the new journey: and updates the display of the edge route to indicate the updated Expected speed. 1 Please use the computer device of ^24, wherein the journey time is private from the 0 cup of the display, and the user can move the slider to adjust the travel time. The computer device of any one of clauses 24 to 25 wherein the expected speed is shown as a color code of the navigable segments. 150832.doc