WO2020082284A1 - 一种路口是否存在目标道路设施的判断方法及系统 - Google Patents

一种路口是否存在目标道路设施的判断方法及系统 Download PDF

Info

Publication number
WO2020082284A1
WO2020082284A1 PCT/CN2018/111807 CN2018111807W WO2020082284A1 WO 2020082284 A1 WO2020082284 A1 WO 2020082284A1 CN 2018111807 W CN2018111807 W CN 2018111807W WO 2020082284 A1 WO2020082284 A1 WO 2020082284A1
Authority
WO
WIPO (PCT)
Prior art keywords
intersection
trajectory data
parameter information
target road
moving object
Prior art date
Application number
PCT/CN2018/111807
Other languages
English (en)
French (fr)
Inventor
孙伟力
张志豪
杜泽龙
Original Assignee
北京嘀嘀无限科技发展有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京嘀嘀无限科技发展有限公司 filed Critical 北京嘀嘀无限科技发展有限公司
Priority to CN201880002448.3A priority Critical patent/CN111386559B/zh
Priority to EP18812027.3A priority patent/EP3678108A1/en
Priority to JP2018565762A priority patent/JP2021503106A/ja
Priority to AU2018279045A priority patent/AU2018279045B2/en
Priority to SG11201811243UA priority patent/SG11201811243UA/en
Priority to CA3027615A priority patent/CA3027615A1/en
Priority to PCT/CN2018/111807 priority patent/WO2020082284A1/zh
Priority to TW107145159A priority patent/TWI715898B/zh
Priority to US16/221,576 priority patent/US20200134325A1/en
Publication of WO2020082284A1 publication Critical patent/WO2020082284A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/04—Traffic conditions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation

Definitions

  • the present invention relates to the field of traffic management informatization, and more specifically, to a method, system, device, and storage medium for judging whether a target road facility exists at an intersection.
  • the present invention is to provide a method, system, device and storage medium for judging whether there is a target road facility at an intersection.
  • the purpose is to intelligently determine the installation status of road facilities at the intersection and reduce the consumption of manpower and material resources.
  • a method for judging whether a target road facility exists at an intersection is implemented on a device that can include a processor and a memory.
  • the method may include one or more of the following operations. It is possible to obtain the left-turn driving trajectory data of the moving object at the intersection to be tested, extract the characteristic parameter information associated with the target road facilities from the trajectory data, and determine the intersection to be measured based on the characteristic parameter information of the intersection to be measured Whether the target road facility exists.
  • the trajectory data is a data set composed of a plurality of trajectory point information according to chronological order, the target road facility includes a left-turn waiting area, and the characteristic parameter information includes driving parameters after a moving object enters an intersection.
  • determining whether the target road facility exists at the intersection under test based on the characteristic parameter information includes: determining a determination threshold; the number of determination thresholds is consistent with the number of characteristic parameters, and Corresponding to the characteristic parameters, compare the characteristic parameter information with the corresponding judgment threshold to determine whether the target road facility exists at the intersection; wherein, if the characteristic parameter information is in the corresponding If the judgment threshold is within the range, then the intersection has a target road facility.
  • the characteristic parameters include at least one of the following: number of stays, stay time, stay distance, delay time, average speed of crossing, and probability of staying twice.
  • determining the judgment threshold includes: acquiring left-traveling trajectory data of a moving object at a known intersection, extracting characteristic parameter information associated with a target road facility from the trajectory data, and marking whether the target exists at a known intersection
  • the road facility determines the judgment threshold of the characteristic parameter information based on the characteristic parameter information of the known intersection and the labeling result.
  • the method further includes: acquiring a left-turn driving trajectory of the moving object at the known intersection as original trajectory data, and extracting original trajectory data in the original trajectory data that is in a flat peak period and has complete feature parameter information
  • the flat peak period is a period of time when the traffic volume is too high and the traffic volume is too low at the intersection, and the traffic volume is stable
  • determining the judgment threshold of the feature parameter information based on the feature parameter information of the known intersection and the labeling result includes: training based on the number of feature parameter information of the known intersection and the result of the labeling
  • the judgment model determines the judgment threshold and the judgment model.
  • determining whether the target road facility exists at the intersection to be tested based on the characteristic parameter information of the intersection to be tested includes: inputting the trajectory data of the intersection to be tested into the judgment model, The judgment result of whether the target road facility exists at the intersection to be tested is output.
  • the judgment model is a decision tree model.
  • the method further includes: acquiring a left-turn driving trajectory of the moving object at the intersection to be measured as original trajectory data, and extracting the original trajectory data that is in a peak period and the feature parameter information is complete
  • the trajectory data is used as the trajectory data.
  • the flat peak period is a period of time when the vehicle flow is too high and the vehicle flow is too low, and the vehicle flow is stable.
  • a system for judging whether a target road facility exists at an intersection includes an acquisition module for acquiring left-turning trajectory data of a moving object at the intersection to be measured, and extracting characteristic parameter information associated with the target road facility from the trajectory data;
  • the characteristic parameter information of the intersection to determine whether the target road facility exists at the intersection to be tested, the trajectory data is a data set composed of a number of trajectory point information according to time sequence, the target road facility includes a left turn to go Area, the characteristic parameter information includes driving parameters after the moving object enters the intersection.
  • a device for judging whether a target road facility exists at an intersection includes a processor and a memory; the memory is used to store instructions, characterized in that when the instructions are executed by the processor, the device is implemented as The operation corresponding to any of the above methods.
  • a computer-readable storage medium characterized in that the storage medium stores computer instructions, and after the computer reads the computer instructions in the storage medium, the computer runs a method for determining whether a target road facility exists at an intersection as described in any of the above .
  • FIG. 1 is a schematic diagram of an exemplary road information system according to some embodiments of the present invention.
  • FIG. 2 is a schematic diagram of exemplary hardware components and / or software components of an exemplary computing device according to some embodiments of the present invention
  • FIG. 3 is a schematic diagram of exemplary hardware components and / or software components of an exemplary mobile device according to some embodiments of the present invention.
  • FIG. 4 is a block diagram of an exemplary processing engine according to some embodiments of the present invention.
  • Figure 5-A is an explanatory diagram of a left turn waiting area at an intersection
  • Fig. 5-B is an explanatory diagram of the intersection without turning left
  • FIG. 6 is a block diagram of another exemplary processing engine according to some embodiments of the present invention.
  • FIG. 7 is an exemplary flowchart of determining whether a target road facility exists at an intersection to be tested according to some embodiments of the present invention.
  • FIG. 8 is an exemplary flowchart of determining a judgment model according to some embodiments of the present invention.
  • FIG. 9 is a schematic diagram of a judgment model according to some embodiments of the present invention.
  • FIG. 10 is an exemplary flowchart of determining whether a target road facility exists at an intersection to be measured using a judgment model according to some embodiments of the present invention.
  • the embodiments of the present application can be applied to a road traffic system and / or mobile device.
  • the road traffic system is traffic lights, traffic cameras, traffic signs, public roads, and pedestrians, automatic vehicles, (for example, small cars, buses, large transport vehicles, Electric vehicles, rickshaws, mobility tools, etc.) public road transportation systems where vehicles and other moving objects travel.
  • Mobile devices are mobile devices equipped with positioning systems, including but not limited to smartphones, smart watches, video cameras, cameras, notebooks, tablets, personal digital assistants (PDAs), in-vehicle computers, navigation, aircraft, etc. used by people in the car Removable device.
  • PDAs personal digital assistants
  • the positioning technology used in the context of the present invention may be based on Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), Compass Navigation System (COMPASS), Galileo Positioning System, Quasi-Zenith Satellite System (QZSS), Wireless Local Area Network ( WiFi) positioning technology, etc., or any combination thereof.
  • GPS Global Positioning System
  • GLONASS Global Navigation Satellite System
  • COMPASS Compass Navigation System
  • Galileo Positioning System Galileo Positioning System
  • QZSS Quasi-Zenith Satellite System
  • WiFi Wireless Local Area Network
  • FIG. 1 is a schematic diagram of a road information system 100 according to some embodiments of the present invention.
  • the road information system 100 may be a platform that provides road information for transportation services.
  • the road information includes but is not limited to road type information, road route information, traffic signal configuration information, road identification information, traffic congestion status information, and the like.
  • the road information system 100 may include a server 110, a data collection terminal 120, a storage device 130, a network 140, and an information source 150.
  • the server 110 may include a processing engine 112.
  • the server 110 may be a single server or a server group.
  • the server farm may be centralized or distributed (for example, the server 110 may be a distributed system).
  • the server 110 may be local or remote.
  • the server 110 may access the information and / or data stored in the storage device 130 and the data collection terminal 120 through the network 140.
  • the server 110 may be directly connected to the information and / or data of the storage device 130 and the data collection terminal 120.
  • the server 110 may be implemented on a cloud platform.
  • the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, between clouds, multiple clouds, etc., or any combination of the above examples.
  • the server 110 may be implemented on the computing device shown in FIG. 2 or FIG. 2 of the present application.
  • the server 110 may be implemented on a computing device 200 as shown in FIG. 2, including one or more components in the computing device 200.
  • the server 110 may be implemented on a mobile device 300 as shown in FIG. 3, including one or more components in the computing device 200.
  • the server 110 may include a processing engine 112.
  • the processing engine 112 may process information and / or data related to road information to perform one or more functions described in this application. For example, the processing engine 112 may determine whether a certain road facility exists at the intersection, for example, whether the intersection is provided with a left-turn waiting area, whether there is a left-turn traffic light, whether a variable lane or other traffic control signs are provided.
  • the processing engine 112 may include one or more processors (eg, single-core processors or multi-core processors).
  • the processing engine 112 may include one or more hardware processors, such as a central processing unit (CPU), application specific integrated circuit (ASIC), application specific instruction set processor (ASIP), image processor (GPU), physical Operational processor (PPU), digital signal processor (DSP), field editable gate array (FPGA), editable logic device (PLD), controller, microcontroller unit, reduced instruction set computer (RISC), microprocessing Or any combination of the above examples.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • ASIP application specific instruction set processor
  • GPU graphics processing
  • PPU physical Operational processor
  • DSP digital signal processor
  • FPGA field editable gate array
  • PLD field editable gate array
  • controller microcontroller unit
  • RISC reduced instruction set computer
  • the data collection terminal 120 may be a video collection device or a mobile device directly equipped with a positioning system.
  • the data collection terminal 120 is a camera 120-1 that is fixed at the intersection or is movable.
  • the camera 120-1 collects driving videos of roads and moving objects on the road, and converts the image information into digital information after processing.
  • the data collection terminal 120 may also be a mobile device configured with positioning information, including but not limited to a built-in device 120-2, a handheld mobile device 120-3, etc., or a combination thereof.
  • the handheld mobile device 120-3 may include, but is not limited to, a smartphone, personal digital assistant (PDA), tablet computer, handheld game console, smart glasses, smart watch, wearable device, virtual display Devices, display enhancement devices, etc. or any combination thereof.
  • the in-vehicle built-in device 120-2 may include, but is not limited to, an in-vehicle computer, an in-vehicle navigation, and the like. Among them, the description includes but is not limited to small cars, buses, large transport vehicles, electric vehicles, rickshaws, mobility tools, and so on.
  • the data collection terminal 120 may send the collected road information to one or more devices in the road information system 100. For example, the data collection terminal 120 may send the road information to the server 110 for processing. The data collection terminal 120 may also send the road information to the storage device 130 for storage.
  • the storage device 130 may store data and / or instructions. In some embodiments, the storage device 130 may store data obtained from the data collection terminal 120. In some embodiments, the storage device 130 may store data and / or instructions for execution or use by the server 110, and the server 110 may implement or use the data and / or instructions to implement the exemplary method described in the present application. In some embodiments, the storage device 130 may include mass storage, removable memory, volatile read-write memory, read-only memory (ROM), etc., or any combination of the above examples. Exemplary mass storage may include magnetic disks, optical disks, solid state drives, and so on. Exemplary removable memory may include flash disks, floppy disks, optical disks, memory cards, compact hard disks, magnetic tape, and the like.
  • Exemplary volatile read-only memory may include random access memory (RAM).
  • Exemplary random access memory may include dynamic random access memory (DRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), static random access memory (SRAM), thyristor random access memory (T-RAM), and zero capacitance memory (Z-RAM )Wait.
  • Exemplary read-only memory may include masked read-only memory (MROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) , Compact hard disk read-only memory (CD-ROM) and digital multi-function hard disk read-only memory, etc.
  • the storage device 130 may be implemented on a cloud platform.
  • the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, between clouds, multiple clouds, etc., or any combination of the above examples.
  • the storage device 130 may be connected to the network 140 to enable communication with one or more components in the road information system 100 (eg, the server 110, the data collection terminal 120, etc.). One or more components of the road information system 100 can access data or instructions stored in the storage device 130 through the network 140. In some embodiments, the storage device 130 may directly connect or communicate with one or more components of the road information system 100 (eg, the server 110, the data collection terminal 120, etc.). In some embodiments, the storage device 130 may be part of the server 110.
  • the network 140 may facilitate the exchange of information and / or data.
  • one or more components in the road information system 100 may send information and other information to the other components in the road information system 100 through the network 140 / Or data.
  • the server 110 may obtain / obtain data information from the data collection terminal 120 through the network 140.
  • the network 140 may be any one of a wired network or a wireless network, or a combination thereof.
  • the network 140 may include a cable network, a wired network, a fiber optic network, a telecommunications network, an intranet, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone Network (PSTN), Bluetooth network, ZigBee network, Near Field Communication (NFC) network, etc. or any combination of the above examples.
  • the network 140 may include one or more network access points.
  • the network 140 may include wired or wireless network access points, such as base stations and / or Internet exchange points 140-1, 140-2, and so on. Through the access point, one or more components of the road information system 100 may be connected to the network 140 to exchange data and / or information.
  • the information source 150 is a source that provides other information to the road information system 100.
  • the information source 150 can be used to provide the system with information related to road information, for example, weather conditions, traffic information, laws and regulations information, news information, life information, life guide information, etc.
  • the information source 150 may exist in the form of a single central server, or in the form of multiple servers connected through a network, or may exist in the form of a large number of personal devices.
  • these devices can use a user-generated content (user-generated contents) method, such as uploading text, voice, images, video, etc. to the cloud server, so that the cloud server is connected to Many connected personal devices together constitute the information source 150.
  • FIG. 2 is a schematic diagram of an exemplary computing device 200 according to some embodiments of the present invention.
  • the server 110 and the storage device 130 may be implemented on the computing device 200.
  • the processing engine 112 may be implemented on the computing device 200 and configured to implement the functions disclosed in this application.
  • the computing device 200 may include any components used to implement the system described in this application.
  • the processing engine 112 may be implemented on the computing device 200 through its hardware, software programs, firmware, or a combination thereof.
  • the computing functions related to the road information system 100 described in this application can be implemented in a distributed manner by a group of similar platforms to spread the processing load of the system.
  • the computing device 200 may include a communication port 250 connected to a network for data communication.
  • the computing device 200 may include a processor (eg, CPU) 220, and may execute program instructions in the form of one or more processors.
  • An exemplary computer platform may include an internal bus 210, different forms of program memory and data storage including, for example, a hard disk 270, and read only memory (ROM) 230 or random access memory (RAM) 240 for storing and processing by the computer / Or various data files transferred.
  • Exemplary computing devices may include program instructions executed by the processor 220 stored in read-only memory 230, random access memory 240, and / or other types of non-transitory storage media. The method and / or process of the present application may be implemented in the form of program instructions.
  • the computing device 200 also includes an input / output component 260 for supporting input / output between the computer and other components.
  • the computing device 200 may also receive the programs and data in this disclosure through network communication.
  • FIG. 2 For ease of understanding, only one processor is shown in FIG. 2 as an example. However, it should be noted that the computing device 200 in this application may include multiple processors, so operations and / or methods described in this application that are implemented by one processor may also be jointly or independently performed by multiple processors achieve. For example, if in this application, the processor of the computing device 200 executes steps 1 and 2, it should be understood that step 1 and step 2 may also be executed jointly or independently by two different processors of the computing device 200 (For example, the first processor performs step 1, the second processor performs step 2, or the first and second processors collectively perform step 1 and step 2).
  • FIG. 3 is a schematic diagram of exemplary hardware and / or software of an exemplary mobile device 300 according to some embodiments of the present invention.
  • the trajectory data can be collected on the mobile device 300.
  • the mobile device 300 may include a positioning unit 301, a communication unit 310, a display unit 320, a graphics processor 330, a processor 340, an input / output unit 350, a memory 360, and a storage Unit 390.
  • the mobile device 300 may further include a bus or a controller.
  • the mobile operating system 370 and one or more application programs 380 may be loaded from the storage unit 390 into the memory 360 and executed by the processor 340.
  • the application 380 may receive and display information about image processing or other information related to the processing engine 112.
  • the input / output unit 350 can realize the interaction of the data information with the road information system 100 and provide the interaction-related information to other components in the road information system 100, such as the server 110, through the network 140.
  • a computer hardware platform may be used as the hardware platform for one or more elements mentioned here.
  • a computer with user interface elements can be used to implement a personal computer (PC) or any other form of workstation or terminal device. With proper programming, a computer can also act as a server.
  • PC personal computer
  • a computer can also act as a server.
  • FIG. 4 is a block diagram of an exemplary processing engine 112 according to some embodiments of the invention. If shown, the processing engine 112 may include an acquisition module 410 and a judgment module 420.
  • the obtaining module 410 can obtain data.
  • the acquisition module 410 may be selected from one or one of the road information system 100, the data collection terminal 120, the storage device 130, the network 140, the information source 150, or any device or component disclosed in this application capable of storing data Get the data above.
  • the acquired data may include one or a combination of one or more of the moving trajectory information of the moving object, the moving object information, the environment information, the traffic jam situation information, the algorithm, the model, and the like.
  • the obtaining module 410 can obtain the moving trajectory data of the moving object at the intersection.
  • the moving object is a movable object that can travel on the road, including but not limited to vehicles, bicycles, carriages, rickshaws, movable robots, and the like.
  • the moving trajectory data of the moving object may be obtained by converting image information collected by a fixed or moving video collection device into digital information after processing.
  • the driving trajectory data may be collected by a mobile device directly equipped with positioning information.
  • the trajectory data is a trajectory data set composed of several trajectory point information according to time sequence, and includes all data information related to traveling of a moving object. For example, travel route, travel time, speed information, location information, etc.
  • the acquisition module 410 may extract feature parameter information associated with the target road facility from the trajectory data.
  • the target road facilities include, but are not limited to, traffic lights, traffic signs (including prohibition signs, travel mode indication signs, etc.), left-turn waiting areas at intersections, variable lanes, and other road traffic facilities and their random combination.
  • the target road facility is a left-turn area.
  • the characteristic parameter information includes driving parameters after the moving object enters the intersection.
  • the characteristic parameter information includes information capable of distinguishing characteristics of intersections with left-turn waiting areas and intersections without left-turn waiting areas.
  • the characteristic parameter information may be the number of stays, the stay time, the stay distance, the time that the moving object passes through the intersection, the delay time, the average speed of passing through the intersection, the probability that the number of stays is equal to or greater than twice, and so on.
  • Fig. 5-A is an explanatory diagram of an intersection with a left-turning waiting area
  • Fig. 5-B is an explanatory diagram of an intersection with no left-turning waiting area.
  • the moving object 510 when the intersection has a left-turn waiting area 530, the moving object 510 usually waits twice when waiting for the left-turn green light. The first time is at the stop position after the stop line 520 at the beginning of the red light, and the second time is after the straight green light is on, the moving object 510 enters the left-turn waiting area 530 and waits for the left-turn green stop position.
  • Position A2 is the second stop position of the moving object 510.
  • the distance that the moving object 510 stays twice is the distance from A1 to A2, that is, the length of the left turn waiting area 530, and the length may be the straight-line distance from A1 to A2 or the trajectory distance from A1 to A2.
  • the time that the moving object 510 stays at A1 is about the remaining time of the straight red light
  • the time that the moving object 510 stays at A2 is about the time of the straight green light.
  • the staying time is the remaining time of the red light.
  • the characteristic parameter may include at least one of the number of stays, the stay time, the stay distance, the delay time, the average speed of crossing the intersection, and the probability of staying twice.
  • the intersection may be a pavement road with a distance from somewhere in the left-turn lane to the entry end of the next lane, wherein the somewhere in the left-turn lane may be On the left-turn lane, there is a certain distance before the stop line ahead of the intersection in the direction of travel.
  • the length of the intersection should include the sum of the distance from the left turn lane to the intersection parking line and the length of the turn from the intersection parking line to the entry end of the next lane.
  • the intersection may be a section of pavement road with a length of 300 m from a certain point on the left-turn lane to the entry end of the next lane.
  • the stay is the trajectory data, and when the speed values of at least two consecutive trajectory points are less than a set value, the stay is considered to be once.
  • the stop may be that when the speed values of three consecutive track points are all less than 0.67 m / s, it is considered that a stop has occurred.
  • the residence time is the duration of one stay.
  • the dwell time may be the remaining time to go straight to the red light, or the time to go straight to the green light.
  • the stay distance is a distance traveled by the moving object between two stays, and the distance may be a linear distance or a trajectory distance.
  • the stay distance may be a straight line length value or a track length value of a left-turning area.
  • the delay time is the difference between the time it takes for the moving object to actually pass through the intersection and the time required for the moving object to pass through the intersection without the stop. If the delay time is within a certain value range, it is considered that the intersection has a left-turn area.
  • the ratio of the delay time and the signal period can be compared with a certain value range. If the ratio is within a certain value range, the intersection is considered to have a left-turn area.
  • the signal period may be a traffic signal change period.
  • the signal period may be the interval between the current straight green light and the next straight green light.
  • the time required to pass through the intersection without the stop can be obtained by acquiring the left-turn trajectory data of the intersection within a period of time, and extracting the Trajectory data, and calculate the average time to pass through the intersection through the extracted trajectory data.
  • the time required to pass through the intersection without the stop may be updated, for example, it may be specified to update every other month.
  • the average speed of the crossing is the average speed of the moving object through the intersection. If the average speed through the intersection is within a certain value range, the intersection is considered to have a left-turn area.
  • the probability of staying twice is the ratio of the number of travel trajectories of the moving object with the number of stays twice and greater than the total number of selected travel trajectories of the moving object. If the probability of staying twice is greater than a certain value, the intersection is considered to have a left-turn area. It should be understood that in some cases, even if the intersection has a left-turn area, it may not stop or only need to stay once to pass the intersection.
  • the characteristic parameter information may be statistical values. For example, the statistics of the number of stays, the mean and variance of the stay time, the mean and variance of the stay distance, the mean and variance of the delay time, the mean and variance of the average speed through the intersection, etc. Using statistical data such as mean and variance as feature parameter information can reduce the influence of individual feature parameter information on the result and improve the accuracy of judgment.
  • the obtaining module 410 can also obtain the left-traveled moving trajectory of the moving object obtained at the intersection to be detected as original trajectory data, and filter the original trajectory data.
  • the original trajectory data in the original trajectory data that is in a flat peak period and the feature parameter information is complete may be extracted as the trajectory data.
  • the flat peak period may be a period of time when the vehicle flow is too high and the vehicle flow is too low, and the vehicle flow is relatively stable. For example, in a general city, the peak period is usually between 10 am and 16 pm.
  • the acquisition module 410 may acquire the trajectory data over a period of time, which may be one month, one quarter, one year, etc., to increase the number of samples.
  • the obtaining module 410 can obtain the trajectory data within one month of the intersection to be measured, extract the characteristic parameter information of the trajectory data, and obtain the statistical value of the characteristic parameter information, and determine whether the intersection has a left-turn pending according to the statistical value of the characteristic parameter information Area.
  • the trajectory data of the intersection to be tested can be used as the trajectory data of the known intersection, and the result of the judgment can be used to mark whether there is a left turn The result of the area is used.
  • the track data and the marked result of the known intersection are used as training samples to determine whether the other intersection to be tested has a left-turn area or to update the data of the model.
  • the processing engine 112 may save the trajectory data and the judgment result of the intersection to be measured in the road information system 100 for use in training samples Data call.
  • the processing engine 112 may store the trajectory data and the judgment result of the intersection to be measured as the trajectory data and labeling result of the known intersection in the storage device 130.
  • the judgment module 420 may be used to determine whether the target road facility exists at the intersection to be tested based on the characteristic parameter information of the intersection to be tested. For example, according to the left-turn driving trajectory data of the moving object at the intersection, extract the characteristic parameter information related to the left-turning waiting area, and determine whether there is a left-turning waiting area at the intersection. In some embodiments, the determination module 420 may also compare the characteristic parameter information with a corresponding determination threshold to determine whether the target road facility exists at the intersection. If the characteristic parameter is within the corresponding judgment threshold value range, there is a target road facility at the intersection. If the characteristic parameter is not within the corresponding judgment threshold range, it is considered that the target road facility does not exist at the intersection.
  • the number of the determination thresholds is consistent with the number of the characteristic parameters, and corresponds one-to-one with the characteristic parameters. For example, when the characteristic parameter is the number of stays, the corresponding judgment threshold can be greater than or equal to 2 times; when the characteristic parameter is the stay time, the corresponding judgment threshold can be two durations, one can be the remaining time of the straight red light, and the other can It is the duration of going straight to the green light.
  • the judgment module 420 may also input the left-turn trajectory data of the moving object at the intersection to be measured into a judgment model, and output the judgment result of whether the target road facility exists through the constructed judgment model. For example, if it is necessary to determine whether there is a left-turning waiting area at a certain intersection, the acquired track data of the intersection within a certain period of time is input into the judgment model, and after the judgment model is calculated, it is output whether the intersection has a left-turning waiting area the result of.
  • the judgment model may be a judgment model obtained in advance through machine learning.
  • the processing engine 112 may further include a training module 430.
  • the training module 430 may be used to determine a judgment threshold.
  • the training module 430 may also be used to obtain left-travel trajectory data of moving objects at known intersections, extract feature parameter information associated with target road facilities from the trajectory data, and mark known intersections Whether the target road facility exists, the determination threshold of the characteristic parameter information is determined based on the characteristic parameter information of the known intersection and the labeling result.
  • a number of known intersections may be selected as samples to obtain trajectory data and annotation results.
  • annotating the existence of the target road facility at a known intersection may collect the results of the existence of the target road facility at a known intersection through manual site surveys, traffic photo cameras, electronic maps, and the like.
  • the trajectory data and labeling results of known intersections may be obtained from the road information system 100, for example, the trajectory data and labeling results of known intersections may be the previously saved intersection trajectory data and corresponding judgment results Converted known data.
  • the training module 430 may obtain the corresponding judgment threshold after processing, statistics, or other calculation methods according to the obtained feature parameter information and the marked results.
  • the judgment threshold may also be an empirical value determined manually according to actual requirements. For example, the judgment threshold of the number of stays can be reasonably inferred according to the actual situation. As mentioned above, when the number of stays is less than 2 times, there is a high probability that there will be no left turn waiting area at the intersection, and the number of stays is greater than In the case of 2 times, there is a high probability that the intersection should have a left-turn waiting area. Therefore, the judgment threshold for the number of “stops” can be directly determined by humans as “ ⁇ 2”.
  • the training module 430 may also be used to train a judgment model based on the feature parameter information of the known intersection and the result of the annotation, to determine the judgment threshold and the judgment model. For example, to obtain the left-turn driving trajectory data of a moving object at 100 intersections, extract the characteristic parameter information related to the left-turning waiting area, and mark whether there is a left-turning waiting area at the 100 intersections, and convert the characteristic parameter information of 100 intersections And the annotation results are used as training samples, and machine learning is performed to obtain the judgment threshold and judgment model of the feature parameter information.
  • the judgment model may be a decision tree model, but is not limited to classification and regression tree (Classification And Regression Tree, CART), iterative binary tree three generation (Iterative Dichotomiser 3, ID3), C4.5 algorithm, random forest (Random Forest), Chisquared Automatic Interaction Detection (CHAID), Multiple Adaptive Regression Spline (Multivariate Adaptive Regression Splines, MARS), Gradient Booster (Gradient Boosting Machine, GBM), etc. or any combination thereof.
  • the verification set may be used to verify the model, and the model parameters may be adjusted according to the verification result to achieve the best state of the model.
  • the data in the verification set and the training data of the judgment model are independently and identically distributed, and there is no intersection.
  • system and its modules shown in FIGS. 4 and 5 can be implemented in various ways.
  • the system and its modules may be implemented by hardware, software, or a combination of software and hardware.
  • the hardware part can be implemented with dedicated logic;
  • the software part can be stored in the memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware.
  • processor control code for example, on a carrier medium such as a magnetic disk, CD, or DVD-ROM, such as a read-only memory (firmware Such codes are provided on programmable memories or data carriers such as optical or electronic signal carriers.
  • the system and its modules of the present application can be implemented by not only hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc. It can also be implemented by, for example, software executed by various types of processors, or by a combination of the above hardware circuits and software (for example, firmware).
  • the process 700 may be performed by processing logic, which may include hardware (eg, circuits, dedicated logic, programmable logic, microcode, etc.), software (running on the processing device to perform hardware simulation Instruction) etc. or any combination thereof.
  • processing logic may include hardware (eg, circuits, dedicated logic, programmable logic, microcode, etc.), software (running on the processing device to perform hardware simulation Instruction) etc. or any combination thereof.
  • One or more operations in the process 700 for judging whether a target road facility exists at an intersection shown in FIG. 7 may be implemented by the road information system 100 shown in FIG. 1.
  • the process 700 may be stored in the storage device 130 in the form of instructions and executed and / or executed by the processing engine 112 (for example, the processor 220 of the computing device 200 shown in FIG. 2, the mobile device shown in FIG. 3 300's central processor 340).
  • obtaining left-traveling trajectory data of the moving object of the intersection to be measured Operation 710 may be performed by acquisition module 410.
  • the driving trajectory data can be obtained by processing and converting image information collected by a fixed or moving video collection device into digital information.
  • the driving trajectory data may be collected by a mobile device directly equipped with positioning information.
  • the trajectory data is a trajectory data set composed of several trajectory point information according to time sequence, and includes all data information related to traveling of a moving object. For example, travel route, travel time, speed information, location information, etc.
  • the moving object is a movable object that can travel on the road, including but not limited to, bicycle, carriage, rickshaw, movable robot, and the like.
  • the trajectory data may be collected by a handheld mobile device 120-3 installed with a positioning unit.
  • the handheld mobile device 120-3 may include, but is not limited to, a smartphone, personal digital assistant (PDA), tablet computer, handheld game console, smart glasses, smart watch, wearable device, virtual display Devices, display enhancement devices, etc. or any combination thereof.
  • feature parameter information associated with the target road facility may be extracted from the trajectory data.
  • operation 720 may be performed by acquisition module 410.
  • the target road facilities include, but are not limited to, traffic lights, traffic signs (including prohibition signs, travel mode indication signs, etc.), left-turn waiting areas at intersections, variable lanes, and other road traffic facilities and their random combination.
  • the target road facility is a left-turn area.
  • the characteristic parameter information includes driving parameters after entering the intersection.
  • the characteristic parameter may include at least one of the number of stays, the stay time, the stay distance, the delay time, the average speed of crossing the intersection, and the probability of staying twice.
  • the intersection may be a pavement road with a distance from somewhere in the left-turn lane to the entry end of the next lane, where the left-turn lane may be left On the turn lane, there is a certain distance before the stop line ahead of the intersection in the direction of travel.
  • the length of the intersection should include the sum of the distance from the left turn lane to the intersection parking line and the length of the turn from the intersection parking line to the entry end of the next lane.
  • the intersection may be a section of pavement road with a length of 300 m from a certain point on the left-turn lane to the entry end of the next lane.
  • the stay is the trajectory data, and when the speed values of at least two consecutive trajectory points are less than a set value, the stay is considered to be once.
  • the stop may be that the speed values of three consecutive track points are all less than 0.67 m / s, it is considered that a stop has occurred.
  • the residence time is the duration of one stay.
  • the dwell time may be the remaining time to go straight to the red light, or the time to go straight to the green light.
  • the stay distance is the distance traveled by the moving object between two stays.
  • the stay distance may be a length value of a left-turn area.
  • the delay time is the difference between the time it takes for the moving object to actually pass through the intersection and the time required for the moving object to pass through the intersection without the stop.
  • the time required for the moving object to pass through the intersection without the stop can be obtained by acquiring the left-turn trajectory data of the intersection within a period of time, and extracting that the occurrence has not occurred.
  • the trajectory data of the stay, and the average time of passing the intersection is calculated by the extracted trajectory data.
  • the time required for the moving object to pass through the intersection without the stop may be updated, for example, it may be specified to update every other month.
  • the average speed of the moving object through the intersection is the average speed of the moving object through the intersection.
  • the probability of staying twice is the ratio of the number of travel trajectories where the number of stays is two or more times to the total number of selected travel trajectories.
  • the characteristic parameter information may be statistical values. For example, the statistics of the number of stays, the mean and variance of the stay time, the mean and variance of the stay distance, the mean and variance of the delay time, the mean and variance of the average speed through the intersection, etc.
  • the determination threshold may be determined first, and then the characteristic parameter information is compared with the corresponding determination threshold to determine whether the target road facility exists at the intersection. If the characteristic parameter is within the corresponding judgment threshold value range, there is a target road facility at the intersection. If the characteristic parameter is not within the corresponding judgment threshold range, it is considered that the target road facility does not exist at the intersection.
  • the characteristic parameter is the probability of staying twice.
  • the probability of staying twice in the trajectory data is between 15% -50%.
  • the probability of staying twice in the trajectory data is ⁇ 5%. Therefore, it can be judged that when the probability of staying twice in the trajectory data is 3%, the intersection does not turn left.
  • the probability of staying twice in the trajectory data is 30%, the intersection has a left-turn area.
  • the number of the determination thresholds is consistent with the number of the characteristic parameters, and corresponds one-to-one with the characteristic parameters.
  • the corresponding judgment threshold can be greater than or equal to 2 times; when the characteristic parameter is the stay time, the corresponding judgment threshold can be two durations, one can be the remaining time of the straight red light, and the other can It is the duration of the straight green light; when the characteristic parameter is the stay distance, the corresponding judgment threshold can be the length of the left turn waiting area; when the characteristic parameter is the probability of staying twice, the judgment threshold can be greater than 15%.
  • the characteristic parameters may be sorted to determine the order of comparison and judgment. For example, the number of stays is used for comparison and judgment. If the number of stays is greater than or equal to 2 times, then it is further compared whether the stay time is equal to the two thresholds. For another example, compare the probability of staying twice. If the probability of staying twice is 10%, then the intersection is considered to have no left turn area. If the probability of staying twice is 75%, which is greater than the threshold of 15%, Further compare whether the residence time is equal to the two thresholds.
  • FIG. 8 is an exemplary flowchart of determining a judgment threshold and a judgment model according to some embodiments of the present invention.
  • the process 800 may be performed by processing logic, which may include hardware (eg, circuits, dedicated logic, programmable logic, microcode, etc.), software (running on the processing device to perform hardware simulation Instruction) etc. or any combination thereof.
  • One or more operations in the process 800 of the judgment model for determining whether a target road facility exists at an intersection shown in FIG. 8 may be implemented by the road information system 100 shown in FIG. 1.
  • the process 800 may be stored in the storage device 130 in the form of instructions and executed and / or executed by the processing engine 112 (for example, the processor 220 of the computing device 200 shown in FIG. 2 and the mobile device shown in FIG. 3 300's central processor 340).
  • Operation 810 it is possible to obtain left-tracking trajectory data of a moving object at a known intersection. Operation 810 may be performed by training module 430.
  • a number of known intersections may be selected as samples to obtain trajectory data and annotation results. For example, 100 intersections are selected as known intersections, and the left-turn trajectory data of these 100 intersections is acquired.
  • the left-turn driving trajectory data of several known intersections within a period of time may be acquired. The period of time may be one month, one quarter, and one year.
  • the left-traveling trajectory of the moving object obtained at the intersection to be measured may be the original trajectory data, and the original trajectory data may be filtered.
  • the original trajectory data in the original trajectory data that is in a flat peak period and the feature parameter information is complete may be extracted as the trajectory data.
  • the flat peak period may be a period of time when the vehicle flow is too high and the vehicle flow is too low, and the vehicle flow is relatively stable. For example, in a general city, the peak period is usually between 10 am and 16 pm. It should be understood that if the intersection is in a peak period of congestion, the number and duration of stays of moving objects will usually be more than the peak period, and the driving speed and the time through the intersection will also be significantly different, resulting in poor data regularity. Not conducive to the accuracy of the calculation. On the other hand, abnormal conditions such as discontinuous and interrupted trajectory data caused by abnormal working of the data collection terminal 220 may also cause interference, resulting in inaccurate calculations. Therefore, by screening the trajectory data, the stability of the data and the accuracy of the judgment result can be improved.
  • feature parameter information associated with the target road facility may be extracted from the trajectory data. Operation 820 may be performed by training module 430.
  • the target road facility may be a left-turn area.
  • the characteristic parameter information may be statistical values. For example, the statistics of the number of stops, the mean and variance of the stay time, the mean and variance of the stopping distance, the mean and variance of the delay time, the mean and variance of the average speed through the intersection, etc.
  • annotating the existence of the target road facility at a known intersection may collect the results of the existence of the target road facility at a known intersection through manual site surveys, traffic photo cameras, electronic maps, and the like. For example, it is possible to mark whether the known intersection has a left-turning area through the real scene information in the electronic map.
  • a judgment model may be trained based on the number of characteristic parameters of the known intersection and the result of the labeling to determine the judgment threshold and the judgment model. Operation 840 may be performed by training module 430.
  • a number of known intersections may be selected to obtain trajectory data and annotation results as training samples. For example, select 100 intersections as known intersections, obtain the left-turn trajectory data of these 100 intersections, and extract the characteristic parameter information, mark whether there is a left-turn waiting area in these 100 intersections, and compare the characteristic parameter information of these 100 intersections with The labeled results are used as training samples and machine learning is performed to obtain the judgment threshold and judgment model corresponding to the feature parameter information.
  • the judgment model may be a decision tree model, including but not limited to classification and regression tree (Classification And Regression Tree, CART), iterative binary tree three generations (Iterative Dichotomiser 3, ID3), C4.5 algorithm, random Forest (Random), Chisquared Automatic Interaction Detection (CHAID), Multiple Adaptive Regression Spline (Multivariate Adaptive Regression Splines, MARS), Gradient Booster (Gradient Boosting Machine, GBM), etc. or any combination thereof .
  • the verification set can be used to verify the model, and the model parameters can be adjusted according to the verification result to achieve the best state of the model.
  • the data in the verification set and the training data of the judgment model are independently and identically distributed, and there is no intersection.
  • the trajectory data of 200 intersections is selected as the sample data, of which the trajectory data of 100 intersections is used as the training sample for building the model, and the trajectory data of the other 100 intersections is the sample data for verification.
  • the sample data for verification is input into the trained judgment model, and the obtained output result is compared with the actual marked result to detect the accuracy of the judgment model.
  • the corresponding judgment threshold may be obtained after processing by sorting, statistics, or other calculation methods.
  • the judgment threshold may also be an empirical value determined manually according to actual requirements.
  • the judgment threshold of the number of stays can be reasonably inferred according to the actual situation. As mentioned above, when the number of stays is less than 2 times, there is a high probability that there will be no left turn waiting area at the intersection, and the number of stays is greater than In the case of 2 times, there is a high probability that the intersection should have a left-turn waiting area. Therefore, the judgment threshold for the number of “stops” can be directly determined by humans as “ ⁇ 2”.
  • FIG. 9 is a flowchart of an exemplary algorithm for applying a decision tree model to determine whether there is a left-turn area at an intersection.
  • step 901 it is possible to obtain left-driving trajectory data of the moving object of the intersection to be measured.
  • Step 901 may be performed by the obtaining module 410.
  • the left-turn driving trajectory data may be collected by a mobile device directly equipped with positioning information.
  • the left-turn driving trajectory data may be left-turn driving trajectory data stored in the storage device for a period of time.
  • the moving object may be a vehicle traveling on a road, an on-board positioning device, or other handheld mobile devices installed with a positioning unit.
  • step 902 feature parameter information such as the number of driving trajectories, the probability of staying twice, the staying time, the stopping distance, the average speed of passing the intersection, and the delay time may be extracted from the trajectory data.
  • Step 902 may be performed by the obtaining module 410.
  • the characteristic parameter information may be statistical values. For example, the statistics of the number of stays, the mean and variance of the stay time, the mean and variance of the stay distance, the mean and variance of the delay time, the mean and variance of the average speed through the intersection, etc. As shown in FIG.
  • the characteristic parameters can be selected as the number of driving trajectories, the probability of staying twice, the stay time, the mean and variance of the stay distance, the mean speed and variance of the average speed through the intersection, and the delay time.
  • Step 903 may be performed by the judgment module 420.
  • the judgment threshold corresponding to the feature parameter information (for example, the first threshold, the second threshold ... the ninth threshold) may be obtained through model training.
  • the determination threshold may also be artificially specified.
  • the first threshold value may be artificially set to 100.
  • step 904 it may be determined whether the probability of staying twice is greater than the second threshold.
  • Step 905 may be performed by the judgment module 420.
  • the second threshold may be obtained through model training.
  • the process may be continued, and if the probability of staying twice is not greater than the second threshold, a result of "no left turn waiting area" is output (step 920).
  • step 906 it may be determined whether the stay time is equal to the third threshold and the fourth threshold.
  • Step 906 may be performed by the judgment module 420.
  • the third threshold and the fourth threshold may be obtained through training.
  • the third threshold may be the time remaining for a straight red light.
  • the fourth threshold may be the time to go straight through the green light.
  • the time for the first stay should be about the remaining time of the straight red light.
  • the time for the second stay should be about the time to go straight to the green light.
  • the judgment process may be continued.
  • the result of "no left turn waiting area" is output (step 920). For example, if a moving object has two stops before and after due to congestion or other unexpected events, the stop time must not conform to the rule of turning left in the waiting area, then even if two stops appear in the driving trajectory, it cannot be considered as The intersection has a left-turn area.
  • step 907 it is determined whether the average value of the stay distance is equal to the fifth threshold. Step 907 may be performed by the judgment module 420.
  • the fifth threshold may be obtained through training. In some embodiments, the fifth threshold may be the distance to turn left in the waiting area.
  • the distance of the left-turning waiting area may be a straight-line distance of the left-turning waiting area. In some embodiments, the distance of the left-turning waiting area may be the actual trajectory distance of the left-turning waiting area. As mentioned above, if the intersection has a left-turning waiting area, the distance between the two stops of the moving object at the intersection should be about the distance of the left-turning waiting area. In some embodiments, if the average value of the stay distance is equal to the fifth threshold, the determination process may be continued. In some embodiments, if the mean value of the stay distance is not equal to the fifth threshold, the result of "no left turn area" is output (step 920).
  • step 908 it may be determined whether the variance of the stay distance is equal to the sixth threshold.
  • Step 908 may be performed by the judgment module 420.
  • the sixth threshold may be obtained through training.
  • the determination process may continue.
  • the result of "no left turn area" is output (step 920).
  • step 909 it can be determined whether the average value of the average speed of the crossing is equal to the seventh threshold.
  • Step 909 may be performed by the judgment module 420.
  • the seventh threshold may be obtained through training.
  • the determination process may continue.
  • the result of "no left turn zone" is output (step 920).
  • the determination process may continue. In some embodiments, if the variance of the average speed through the intersection is equal to the seventh threshold, the determination process may continue. In some embodiments, if the variance of the average speed through the intersection is not equal to the seventh threshold, the result of "no left turn waiting area" is output (step 920). In step 911, it can be determined whether the ratio of the delay time to the signal period is equal to the ninth threshold. Step 911 may be performed by the judgment module 420. In some embodiments, the ninth threshold may be obtained through training. In some embodiments, the delay time is the difference between the time it takes for the moving object to actually pass through the intersection, and the time required for the moving object to pass through the intersection without stopping. The signal period may be the interval between the current straight green light and the next straight green light.
  • the result of "there is a left-turn area” is output (step 921). In some embodiments, if the ratio of the delay time to the signal period is not equal to the ninth threshold, the result of "there is a left-turn area” is output (step 920).
  • FIG. 10 is an exemplary flowchart of determining whether a target road facility exists at an intersection to be determined according to some embodiments of the present invention.
  • the process 1000 may be performed by processing logic, which may include hardware (eg, circuits, dedicated logic, programmable logic, microcode, etc.), software (running on a processing device to perform hardware simulation Instruction) etc. or any combination thereof.
  • processing logic may include hardware (eg, circuits, dedicated logic, programmable logic, microcode, etc.), software (running on a processing device to perform hardware simulation Instruction) etc. or any combination thereof.
  • One or more operations in the process 1000 shown in FIG. 10 for determining whether there is a target road facility at the intersection to be determined can be implemented by the road information system 100 shown in FIG. 1.
  • the process 1000 may be stored in the storage device 130 in the form of instructions and executed and / or executed by the processing engine 112 (for example, the processor 220 of the computing device 200 shown in FIG. 2, the mobile device shown in FIG. 3 300's central processor 340).
  • the trajectory data of the intersection to be measured may be input to the judgment model. Operation 1010 may be performed by the judgment module 420.
  • the trajectory data of the intersection to be measured may be trajectory data of the intersection to be determined within a period of time. A period of time can be one month, one quarter, one year, etc.
  • the acquired driving trajectory of the intersection may also be used as the initial trajectory data, and the initial trajectory data may be extracted in a peak period according to traffic congestion conditions, and the complete trajectory data may be used as the trajectory data. To improve the stability of data and the accuracy of judgment results.
  • the judgment module 430 may access the data stored in the storage device 230 through the network 140 and obtain the archived data in the road information system 200 based on the position information of the intersection to be determined to obtain the information of the intersection to be determined
  • the left-turn driving trajectory of the intersection to be measured may also be obtained as original trajectory data, and the original trajectory data may be filtered.
  • the original trajectory data in the original trajectory data that is in a flat peak period and the feature parameter information is complete may be extracted as the trajectory data.
  • the flat peak period may be a period of time when the vehicle flow is too high and the vehicle flow is too low, and the vehicle flow is relatively stable. For example, in a general city, the peak period is usually between 10 am and 16 pm.
  • a judgment result of whether the target road facility exists at the intersection to be measured may be output. Operation 1020 may be performed in the judgment module 420.
  • the judgment result may be represented by a number "0" or "1". For example, “1" can be set to indicate that the intersection has the target road facility, and "0" indicates that the intersection does not have the target road facility. If the judgment result is that there is a left turn waiting area, the judgment module 420 outputs "1", if When the judgment result is that there is no left-turn area, the judgment module 420 outputs "0".
  • the possible benefits brought by the embodiments of the present application include but are not limited to: (1) can accurately and intelligently determine the road facility configuration of intersections, reduce the loss of human resources and time costs; (2) the present invention provides accurate determination The characteristic parameters required for the left turn waiting area to improve the accuracy of the judgment model; (3) The present invention provides a judgment model, which can accurately determine whether the intersection has a left turn waiting area.
  • different embodiments may have different beneficial effects.
  • the possible beneficial effects may be any one or a combination of the above, or any other beneficial effects that may be obtained.
  • All software or parts of it may sometimes communicate over a network, such as the Internet or other communication networks.
  • Such communications can load software from one computer device or processor to another.
  • another medium that can transmit software elements can also be used as a physical connection between local devices, such as light waves, radio waves, electromagnetic waves, etc., through cables, optical cables, or air.
  • the physical media used for carrier waves such as cables, wireless connections, or optical cables and similar devices can also be considered as software-bearing media.
  • the tangible "storage” medium is limited, other terms that represent a computer or machine "readable medium” all mean a medium that participates in the execution of any instructions by the processor.
  • the computer program code required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C ++, C #, VB.NET, Python Etc., conventional programming languages such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages such as Python, Ruby and Groovy, or other programming languages.
  • the program code may run entirely on the user's computer, or as an independent software package on the user's computer, or partly on the user's computer, partly on a remote computer, or entirely on the remote computer or server.
  • the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (such as through the Internet), or in a cloud computing environment, or as Service usage such as software as a service (SaaS).
  • LAN local area network
  • WAN wide area network
  • SaaS software as a service
  • Some embodiments use numbers describing attributes and numbers. It should be understood that such numbers used for embodiment descriptions use the modifiers "about”, “approximately” or “substantially” to modify in some examples. . Unless otherwise stated, “approximately”, “approximately” or “substantially” indicates that the figures allow a variation of ⁇ 20%.
  • the numerical parameters used in the specification and claims are approximate values, and the approximate value may be changed according to the characteristics required by individual embodiments. In some embodiments, the numerical parameters should consider the specified significant digits and adopt the method of general digit retention. Although the numerical fields and parameters used to confirm the breadth of the ranges in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values is as accurate as possible within the feasible range.

Abstract

一种路口是否存在目标道路设施的判断方法、系统、装置和存储介质。方法包括：获取待测路口的移动物体左转行驶轨迹数据(710)，从轨迹数据中提取与目标道路设施相关联的特征参数信息(720)，基于待测路口的特征参数信息确定待测路口是否存在目标道路设施(730)。能够智能确定路口的道路设施设置状况，减少人力和物力的耗费。

Description

一种路口是否存在目标道路设施的判断方法及系统 技术领域
本发明涉及交通管理信息化领域，更具体的，涉及一种路口是否存在目标道路设施的判断方法、系统、装置及存储介质。
背景技术
随着城市交通的智能化发展，为了实现交通信号灯的智能调控，需要事先清楚交叉路口的道路设施设置状况。例如，交叉路口是否设置有左转待行区、是否有左转交通灯或其他交通控制标识等。以左转待行区为例，交叉路口是否有左转待行区，交通信号灯的配时设置会不同。如果路口具有左转待行区，需要道路上行驶的移动物体直行先走，左转后走。但是有的交叉路口具有左转待行区，有的交叉路口没有左转待行区，需要事先准确知道每个交叉路口是否有左转待行区。因此，交叉路口是否设置有某道路设施，对交通管理的信息化有很重要的影响。目前，还没有智能判断交叉路口是否存在某道路设施的判断方法，如果每个路口都需要人工到现场查看后确定，会耗费大量的人力物力。所以，需要一种准确的智能的判断交叉路口是否存在某道路设施的方法。
发明内容
本发明的在于提供一种路口是否存在目标道路设施的判断方法、系统、装置及存储介质，目的是智能化确定路口的道路设施设置状况，减少人力和物力的耗费。
为了达到上述发明的目的，本发明提供的技术方案如下：
一种路口是否存在目标道路设施的判断方法。所述方法在可以在包括一个处理器和一个存储器的设备上实现。所述方法可以包括以下一个或一个以上操作。可以获取待测路口的移动物体左转行驶轨迹数据，从所述轨迹数据中提取与目标道路设施相关联的特征参数信息，基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施。所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的数据集，所述目标道路设施包括左转待行区，所述特征参数信息包括移动物体进入路口后的行驶参数。
在本发明中，基于所述特征参数信息确定所述待测路口是否存在所述目标道路设施，包括：确定判断阈值；所述判断阈值的个数与所述特征参数的个数一致，并与所述特征参数一一相对应，将所述特征参数信息与对应的所述判断阈值比较，判断所述路口是否存在所述目标道路设施；其中，如果所述特征参数信息处于对应的所述判断阈值范围内，则所述路口存在目标道路设施。
在本发明中，所述特征参数包括以下至少一个：停留次数、停留时间、停留距离、延误时间、通过路口的平均速度和停留两次的概率。
在本发明中，确定判断阈值包括：获取已知路口的移动物体左转行驶轨迹数据，从所述轨迹数据中提取与目标道路设施相关联的特征参数信息，标注已知路口是否存在所述目标道路设施，基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值。
在本发明中，所述方法进一步包括：获取所述已知路口的移动物体 左转行驶轨迹为原始轨迹数据，提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的原始轨迹数据作为所述轨迹数据，所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间
在本发明中，基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值，包括：基于已知路口的所述特征参信息数和所述标注的结果训练判断模型，确定所述判断阈值和所述判断模型。
在本发明中，基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施，包括：将所述待测路口的所述轨迹数据输入所述判断模型，输出所述待测路口是否存在所述目标道路设施的判断结果。
在本发明中，所述判断模型为决策树模型。
在本发明中，所述方法进一步包括：获取所述待测路口的移动物体左转行驶轨迹为原始轨迹数据，提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据，所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间。
一种路口是否存在目标道路设施的判断系统。所述系统包括获取模块，用于获取待测路口的移动物体左转行驶轨迹数据，并从所述轨迹数据中提取与目标道路设施相关联的特征参数信息；判断模块，用于基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施，所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的数据集，所述目标道路设施包括左转待行区，所述特征参数信息包括移动物体进入路 口后的行驶参数。
一种路口是否存在目标道路设施的判断装置，所述装置包括处理器以及存储器；所述存储器用于存储指令，其特征在于，所述指令被所述处理器执行时，导致所述装置实现如上述任一项所述方法对应的操作。
一种计算机可读存储介质，其特征在于，所述存储介质存储计算机指令，当计算机读取存储介质中的计算机指令后，计算机运行如上述任意一项所述路口是否存在目标道路设施的判断方法。
附加的特征将在下面的描述中部分地阐述，并且对于本领域技术人员来说，通过查阅以下内容和附图将变得显而易见，或者可以通过实例的产生或操作来了解。本发明的特征可以通过实践或使用以下详细实例中阐述的方法、工具和组合的各个方面来实现和获得。
附图说明
根据示例性实施例可以进一步描述本申请。参考附图可以详细描述所述示例性实施例。所述实施例并非限制性的示例性实施例，其中相同的附图标记代表附图的几个视图中相似的结构，并且其中：
图1是根据本发明的一些实施例所示的一个示例性道路信息系统的示意图；
图2是根据本发明的一些实施例所示的一个示例性计算设备的示例性硬件组件和/或软件组件的示意图；
图3是根据本发明的一些实施例所示的一个示例性移动设备的示例性硬件组件和/或软件组件的示意图；
图4是根据本发明的一些实施例所示的一个示例性处理引擎的框图；
图5-A是路口有左转待行区的说明性示意图；
图5-B是路口没有左转待行区的说明性示意图；
图6是根据本发明的一些实施例所示的另一个示例性处理引擎的框图；
图7是是根据本发明的一些实施例所示的确定待测路口是否存在目标道路设施的示例性流程图；
图8是根据本发明的一些实施例所示的确定判断模型的示例性流程图；
图9是根据本发明的一些实施例所示的一个判断模型的示意图；
图10是根据本发明的一些实施例所示的利用判断模型确定待测路口是否存在目标道路设施的示例性流程图。
具体实施方式
为了更清楚地说明本申请的实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单的介绍。显而易见地，下面描述中的附图仅仅是本申请的一些示例或实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图将本申请应用于其他类似情景。除非从语言环境中显而易见或另做说明，图中相同标号代表相同结构或操作。
如本申请和权利要求书中所示，除非上下文明确提示例外情形， “一”、“一个”、“一种”和/或“该”等词并非特指单数，也可包括复数。一般说来，术语“包括”与“包含”仅提示包括已明确标识的步骤和元素，而这些步骤和元素不构成一个排它性的罗列，方法或者设备也可能包含其他的步骤或元素。
虽然本申请对根据本申请的实施例的系统中的某些模块做出了各种引用，然而，任何数量的不同模块可以被使用并运行在客户端和/或服务器上。所述模块仅是说明性的，并且所述系统和方法的不同方面可以使用不同模块。
本申请中使用了流程图用来说明根据本申请的实施例的系统所执行的操作。应当理解的是，前面或下面操作不一定按照顺序来精确地执行。相反，可以按照倒序或同时处理各种步骤。同时，也可以将其他操作添加到这些过程中，或从这些过程移除某一步或数步操作。
本申请的实施例可以应用于路面交通系统和/或移动设备，路面交通系统为交通灯、交通摄像头、交通标识、公共道路以及行人、自动车、(例如，小型车、巴士、大型运输车、电动车、人力车、代步工具等)交通工具等移动物体行驶的公共道路交通系统。移动设备为配置有定位系统的可移动设备，包括但不限于车内人员使用的智能手机、智能手表、摄像机、照相机、笔记本、平板电脑、个人数码助理(PDA)、车载电脑、导航、飞行器等可移动设备。应当理解的是，本申请的系统及方法的应用场景仅仅是本申请的一些示例或实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图将本申请应用于其他类似情景。
在本发明内容中所使用的定位技术可基于全球定位系统(GPS)，全球导航卫星系统(GLONASS)，罗盘导航系统(COMPASS)，伽利略定位系统，准天顶卫星系统(QZSS)，无线局域网(WiFi)定位技术等，或其任何组合。一个或多个上述定位系统可以在本发明中互换使用。
图1是根据本发明的一些实施例所示的一种道路信息系统100的示意图。例如，道路信息系统100可以是一个为交通运输服务提供道路信息的平台。所述道路信息包括但不限于道路类型信息、道路路线信息、交通信号灯配置信息、道路标识信息、交通拥堵状况信息等。道路信息系统100可以包括一个服务器110、数据采集端120、一个存储设备130、一个网络140和一个信息源150。服务器110可以包括一个处理引擎112。
在一个实施例中，在一些实施例中，服务器110可以是一个单个的服务器或者一个服务器群组。所述服务器群可以是集中式的或分布式的(例如，服务器110可以是一个分布式的系统)。在一些实施例中，服务器110可以是本地的或远程的。例如，服务器110可以通过网络140访问存储在存储设备130、数据采集端120的信息和/或数据。再例如，服务器110可以直接连接到存储设备130、数据采集端120的信息和/或数据。在一些实施例中，服务器110可以在一个云平台上实现。仅仅举个例子，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、云之间、多重云等或上述举例的任意组合。在一些实施例中，服务器110可以在与本申请图2或图2所示的计算设备上实现。例如，服务器110可以在如图2所示的一个计算设备200上实现，包括计算设备200中的一个或多个部件。再例如，服务器110可以在如图3所示的一个移动设备300上实现， 包括计算设备200中的一个或多个部件。
在一些实施例中，服务器110可以包括一个处理引擎112。处理引擎112可以处理与道路信息相关的信息和/或数据以执行本申请描述的一个或多个功能。例如，处理引擎112可以判断路口是否存在某道路设施，例如，路口是否设置有左转待行区、是否有左转交通灯、是否设置有可变车道或其他交通控制标识。在一些实施例中，处理引擎112可以包括一个或多个处理器(例如，单核处理器或多核处理器)。仅仅举个例子，处理引擎112可以包括一个或多个硬件处理器，例如中央处理器(CPU)、专用集成电路(ASIC)、专用指令集处理器(ASIP)、图像处理器(GPU)、物理运算处理器(PPU)、数字信号处理器(DSP)、现场可编辑门阵列(FPGA)、可编辑逻辑器件(PLD)、控制器、微控制器单元、精简指令集计算机(RISC)、微处理器等或上述举例的任意组合。
数据采集端120可以是视频采集装置或是直接配有定位系统的移动设备等。在一些实施例中，数据采集端120为固定在路口或可移动的摄像头120-1，通过摄像头120-1采集道路和道路上移动物体的行驶视频，经过处理将图像信息转换为数字信息。在一些实施例中，数据采集端120还可以是配置有定位信息的移动设备，包括但不限于内置设备120-2、手持移动设备120-3等或其组合。在一些实施例中，手持移动设备120-3可以包括但不限于智能手机、个人数码助理(Personal Digital Assistance，PDA)、平板电脑、掌上游戏机、智能眼镜、智能手表、可穿戴设备、虚拟显示设备、显示增强设备等或其任意组合。在一些实施例中，车载内置设备120-2可以包括但不限于车载电脑、车载导航等。其中，所述包括但 不限于小型车、巴士、大型运输车、电动车、人力车、代步工具等。在一些实施例中，数据采集端120可以将采集到的道路信息发送至道路信息系统100中的一个或多个设备中。例如，数据采集端120可以将道路信息发送至服务器110进行处理。数据采集端120也可以将道路信息发送至存储设备130中存储。
存储设备130可以存储数据和/或指令。在一些实施例中，存储设备130可以存储从数据采集端120获得的数据。在一些实施例中，存储设备130可以存储供服务器110执行或使用的数据和/或指令，服务器110可以通过执行或使用所述数据和/或指令以实现本申请描述的示例性方法。在一些实施例中，存储设备130可以包括大容量存储器、可移动存储器、挥发性读写存储器、只读存储器(ROM)等或上述举例的任意组合。示例性的大容量存储器可以包括磁盘、光盘、固态硬盘等。示例性的可移动存储器可以包括闪存盘、软盘、光盘、记忆卡、压缩硬盘、磁带等。示例性的挥发性只读存储器可以包括随机存储器(RAM)。示例性的随机存储器可以包括动态随机存储器(DRAM)、双数据率同步动态随机存储器(DDRSDRAM)、静态随机存储器(SRAM)、可控硅随机存储器(T-RAM)和零电容存储器(Z-RAM)等。示例性的只读存储器可以包括掩蔽型只读存储器(MROM)、可编程只读存储器(PROM)、可擦除可编程只读存储器(EPROM)、电可擦除可编程只读存储器(EEPROM)、压缩硬盘只读存储器(CD-ROM)和数字多功能硬盘只读存储器等。在一些实施例中，存储设备130可以在一个云平台上实现。仅仅举个例子，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、云之间、多重 云等或上述举例的任意组合。
在一些实施例中，存储设备130可以与网络140连接以实现与道路信息系统100中的一个或多个部件(例如，服务器110、数据采集端120等)之间的通信。道路信息系统100的一个或多个部件可以通过网络140访问存储在存储设备130中的数据或指令。在一些实施例中，存储设备130可以直接与道路信息系统100的一个或多个部件(例如，服务器110、数据采集端120等)连接或通信。在一些实施例中，存储设备130可以是服务器110的一部分。
网络140可以促进信息和/或数据的交换。在一些实施例中，道路信息系统100中的一个或多个部件(例如，服务器110、存储设备130、和数据采集端120等)可以通过网络140向道路信息系统100中的其他部件发送信息和/或数据。例如，服务器110可以通过网络140从数据采集端120获取/得到数据信息。在一些实施例中，网络140可以是有线网络或无线网络中的任意一种，或其组合。例如，网络140可以包括电缆网络、有线网络、光纤网络、远程通信网络、内联网、互联网、局域网(LAN)、广域网(WAN)、无线局域网(WLAN)、城域网(MAN)、公共开关电话网络(PSTN)、蓝牙网络、ZigBee网络、近场通讯(NFC)网络等或上述举例的任意组合。在一些实施例中，网络140可以包括一个或多个网络接入点。例如，网络140可能包括有线或无线网络接入点，如基站和/或互联网交换点140-1、140-2等等。通过接入点，道路信息系统100的一个或多个部件可能连接到网络140以交换数据和/或信息。
信息源150是为道路信息系统100提供其他信息的一个源。信息源 150可以用于为系统提供与道路信息相关的信息，例如，天气情况、交通信息、法律法规信息、新闻信息、生活资讯、生活指南信息等。信息源150可以是一个单独的中央服务器的形式存在，也可以是以多个通过网络连接的服务器的形式存在，还可以是以大量的个人设备形式存在。当信息源150以大量个人设备形式存在时，这些设备可以通过一种用户生成内容(user-generated contents)的方式，例如向云端服务器上传文字、语音、图像、视频等，从而是云端服务器连通与其连接的众多个人设备一起组成信息源150。
图2是根据本发明的一些实施例所示的一种示例性计算设备200的示意图。服务器110和存储设备130可以在计算设备200上实现。例如，处理引擎112可以在计算设备200上实现并被配置为实现本申请中所披露的功能。
计算设备200可以包括用来实现本申请所描述的系统的任意部件。例如，处理引擎112可以在计算设备200上通过其硬件、软件程序、固件或其组合实现。为了方便起见图中仅绘制了一台计算机，但是本申请所描述的与道路信息系统100相关的计算功能可以以分布的方式、由一组相似的平台所实施，以分散系统的处理负荷。
计算设备200可以包括与网络连接的通信端口250，用于实现数据通信。计算设备200可以包括一个处理器(例如，CPU)220，可以以一个或多个处理器的形式执行程序指令。示例性的电脑平台可以包括一个内部总线210、不同形式的程序存储器和数据存储器包括，例如，硬盘270、和只读存储器(ROM)230或随机存储器(RAM)240，用于存储 由计算机处理和/或传输的各种各样的数据文件。示例性的计算设备可以包括存储在只读存储器230、随机存储器240和/或其他类型的非暂时性存储介质中的由处理器220执行的程序指令。本申请的方法和/或流程可以以程序指令的方式实现。计算设备200也包括输入/输出部件260，用于支持电脑与其他部件之间的输入/输出。计算设备200也可以通过网络通讯接收本披露中的程序和数据。
为理解方便，图2中仅示例性绘制了一个处理器。然而，需要注意的是，本申请中的计算设备200可以包括多个处理器，因此本申请中描述的由一个处理器实现的操作和/或方法也可以共同地或独立地由多个处理器实现。例如，如果在本申请中，计算设备200的处理器执行步骤1和步骤2，应当理解的是，步骤1和步骤2也可以由计算设备200的两个不同的处理器共同地或独立地执行(例如，第一处理器执行步骤1，第二处理器执行步骤2，或者第一和第二处理器共同地执行步骤1和步骤2)。
图3是根据本发明的一些实施例所示的一个示例性的移动设备300的示例性硬件和/或软件的示意图。轨迹数据的采集可以在移动设备300上实现。如图3所示，移动设备300可以包括一个定位单元301、一个通信单元310、一个显示单元320、一个图形处理器330、一个处理器340、一个输入/输出单元350、一个内存360和一个存储单元390。移动设备300中还可以包括一个总线或者一个控制器。在一些实施例中，移动操作系统370和一个或多个应用程序380可以从存储单元390加载到内存360中，并由处理器340执行。在一些实施例中，应用程序380可以接收和显示与处理引擎112有关的图像处理或其他信息的信息。输入/输出单元350 可以实现将数据信息与道路信息系统100的交互，并将交互相关信息通过网络140提供给道路信息系统100中的其他部件，如服务器110。
为了实现本申请中描述的各种模块、单元及其功能，计算机硬件平台可以用作这里提到的一个或多个元件的硬件平台。一个拥有用户界面元件的计算机可以用于实现个人计算机(PC)或者其它任何形式的工作站或终端设备。通过合适的编程，一个计算机也可以充当一台服务器。
图4是根据本发明的一些实施例所示的示例性处理引擎112的框图。如果所示，处理引擎112可以包括获取模块410和判断模块420。
获取模块410可以获取数据。在一些实施例中，获取模块410可以从道路信息系统100、数据采集端120、存储设备130、网络140、信息源150或本申请中公开的能够存储数据的任何设备或组件中的一个或一个以上获取数据。所获取的数据可以包括移动物体的行驶轨迹信息、移动物体信息、环境信息、交通拥堵状况信息、算法、模型等中的一种或一种以上的组合。在一些实施例中，获取模块410可以获取路口的移动物体行驶轨迹数据。在一个实施例中，所述移动物体为可在道路上行驶的可移动的物体，包括但不限于车辆、自行车、马车、人力车、可移动的机器人等。在一个实施例中，所述移动物体行驶轨迹数据可以通过固定或移动的视频采集装置采集的图像信息经过处理转换为数字信息后获得。在一些实施例中，所述行驶轨迹数据可以是通过直接配有定位信息的移动设备采集。在一些实施例中，所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的轨迹数据集，包括所有与移动物体行驶相关的数据信息。例如，行驶线路、行驶时间、速度信息、位置信息等。
在一些实施例中，获取模块410可以从所述轨迹数据中提取与目标道路设施相关联的特征参数信息。在一些实施例中，所述目标道路设施包括但不限于交通信号灯、交通标识(包括禁令标识、行进方式指示标识等)、交叉路口左转待行区、可变车道等其他道路交通设施及其任意组合。在一些实施例中，所述目标道路设施为左转待行区。在一些实施例中，所述特征参数信息包括移动物体进入路口后的行驶参数。在一些实施例中，所述特征参数信息包括能够区分具有左转待行区和没有左转待行区的路口的特性的信息。例如，特征参数信息可以是停留次数、停留时间、停留距离、移动物体通过路口的时间、延误时间、通过路口的平均速度、停留次数等于或大于两次的概率等。
图5-A是路口有左转待行区的说明性示意图；图5-B是路口没有左转待行区的说明性示意图。如图5-A所示，当路口具有左转待行区530时，移动物体510在等候左转绿灯时，通常会出现两次停留。第一次是在红灯开始时，在停车线520后的停留位置，第二次是直行绿灯亮后，移动物体510驶入左转待行区530内等候左转绿灯时停留的位置。例如，当左转红灯开始时，如果移动物体510正好停在停车线520上，当前的位置A1就是移动物体510的第一次停留位置，当直行绿灯亮起，移动物体510向前移动驶入左转待行区530，在左转待行区530的停车线531上停止，等待左转绿灯，此时发生第二次停留，位置A2为移动物体510的第二次停留位置。移动物体510两次停留的距离就是由A1到A2的距离，即左转待行区530的长度，所述长度可以是A1到A2的直线距离或是A1到A2的轨迹距离。在A1处移动物体510停留的时间约为直行红灯剩余的时 间，在A2处移动物体510停留的时间约为直行绿灯的时间。如图5-B所示，当路口没有左转待行区时，左转和直行一样，通常只出现一次停留，停留时间为红灯的剩余时间。并且，由于有左转待行区的路口，会出现两次停留，所以通过路口的时间、平均速度也会和通过没有左转待行区的路口不同。因此，没有左转待行区和具有左转待行区的路口，移动物体510的行驶轨迹会有不同，对应的特征参数信息也不同。
在一个实施例中，所述特征参数可以包括停留次数、停留时间、停留距离、延误时间、通过路口的平均速度和停留两次的概率中的至少一个。在一些实施例中，所述路口可以是从左转车道的某一处开始到下一车道的进入端之间，具有一段距离的路面道路，其中，所述左转车道的某一处可以为左转车道上，在位于行驶方向前方路口停车线之前具有一定距离的位置。换句话说，所述路口的长度应包括左转车道上某处到所述路口停车线的距离，及从所述路口停车线到下一车道进入端的转弯长度的总和。例如，所述路口可以为从所述左转车道上某一处到下一车道进入端之间，长度为300m距离的一段路面道路。在一些实施例中，所述停留为所述轨迹数据中，至少两个连续所述轨迹点的速度值均小于一设定值时，认为所述停留一次。例如，所述停留可以是连续三个轨迹点的速度值均小于0.67m/s时，认为发生了一次停留。在一些实施例中，所述停留时间为一次所述停留的时长。例如，停留时间可以为直行红灯的剩余时间，或是直行绿灯的时间。所述停留距离为两次所述停留之间所述移动物体走过的距离，所述距离可以是直线距离或是轨迹距离。例如，所述停留距离可以是左转待行区的直线长度值或轨迹长度值。如前所述，如果移动物体在一路 口发生两次以上的停留，停留距离为左转待行区长度，则认为该路口具有左转待行区。所述延误时间为所述移动物体实际通过所述路口花费的时间，与所述移动物体在没有发生所述停留的情况下通过所述路口需要的时间的差值。如果延误时间在某数值范围内，则认为所述路口具有左转待行区。在一个实施例中，可以将延误时间和信号周期的比值与某一数值范围比较，如果该比值在某一数值范围内则认为该路口具有左转待行区。其中，信号周期可以是交通信号灯变化周期，例如，信号周期可以是当前直行绿灯到下一次直行绿灯出现的间隔时间。在一些实施例中，所述在没有发生所述停留的情况下通过所述路口需要的时间，可以通过以下方式得到：获取一段时间内的路口左转轨迹数据，提取没有发生过所述停留的轨迹数据，并通过提取到的轨迹数据计算通过所述路口的平均时间。在一些实施例中，所述在没有发生所述停留的情况下通过所述路口需要的时间可以得到更新，例如，可以规定每隔一个月的时间进行一次更新。在一个实施例中，所述通过路口的平均速度为所述移动物体通过所述路口的平均速度。如果通过所述路口的平均速度位于某一数值范围内，则认为该路口具有左转待行区。所述停留两次的概率为所述停留次数为两次及大于两次的所述移动物体行驶轨迹的数量占所选取的所述移动物体行驶轨迹的数量总和的比率。如果停留两次的概率大于某一数值时，则认为该路口具有左转待行区。需要理解的是，在一些情况下，即使路口具有左转待行区，也可能不会停留或只需要一次停留就可通过路口。但是路口具有左转待行区时，停留两次的概率会明显比没有左转待行区的高，因此，使用所述停留两次的概率作为特征参数信息判断路口是否有左转待行区，结果会更为准 确。在一些实施例中，所述特征参数信息可以是统计数值。例如，停留次数的个数统计，停留时间的均值和方差，停留距离的均值和方差，延误时间的均值和方差，通过路口的平均速度的均值和方差等。采用均值和方差等统计数据作为特征参数信息，可以减少个别特征参数信息对结果的影响，提高判断的准确性。
在一些实施例中，获取模块410还可以将获取所述待测路口的移动物体左转行驶轨迹为原始轨迹数据，对所述原始轨迹数据进行筛选。在一些实施例中，可以提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据。在一些实施例中，平峰时段可以是排除车流量过高和车流量过低，车流量较稳定的一段时间。例如，一般城市中，平峰时段通常是上午10点到下午16点之间的时段。需要理解的是，如果路口处于拥堵的高峰时段，移动物体的停留次数和停留时间通常会多于平峰时段，行驶速度和通过路口的时间也会有较大的差异，导致数据规律性较差，不利于计算的准确性。另一方面，由于数据采集端220出现工作异常而导致的轨迹数据不连续、有中断等异常状况也会造成干扰，导致计算不准确。因此，通过对所述轨迹数据的筛选，可以提高数据的稳定性和判断结果的准确性。在一些实施例中，获取模块410可以获取一段时间内的所述轨迹数据，所述一段时间可以是一个月、一个季度、一年等，以增加样本的数量。例如，获取模块410可以获取待测路口一个月内的轨迹数据，提取轨迹数据的特征参数信息，并得到特征参数信息的统计数值，根据特征参数信息的统计数值判断该路口是否具有左转待行区。获取的轨迹数据数量越多，判断结果越准确。在一个实施例中，根 据所述待测路口的轨迹数据得到判断结果后，所述待测路口的轨迹数据可以作为已知路口的轨迹数据使用，判断的结果可以作为标注是否有左转待行区的结果使用，将已知路口的轨迹数据和标注的结果作为训练样本使用，用于判断其他待测路口是否具有左转待行区，或是用于判断模型的数据更新。在一个实施例中，所述待测路口的轨迹数据得到判断结果后，处理引擎112可以将所述待测路口的轨迹数据和判断的结果保存在道路信息系统100中，用于训练时的样本数据调用。例如，处理引擎112可以将所述待测路口的轨迹数据和判断的结果作为已知路口的轨迹数据和标注结果，存储在存储设备130中。
判断模块420可以用于基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施。例如，根据所述路口的移动物体左转行驶轨迹数据，提取与左转待行区相关的特征参数信息，确定所述路口是否存在左转待行区。在一些实施例中，判断模块420还可以将所述特征参数信息与对应的判断阈值比较，判断所述路口是否存在所述目标道路设施。如果所述特征参数处于对应的所述判断阈值范围内，则所述路口存在目标道路设施。如果所述特征参数不处于对应的所述判断阈值范围内，则认为所述路口不存在所述目标道路设施。例如，如果某路口的左转的轨迹数据中，移动物体停留次数明显大于等于两次时，判断该路口具有左转待行区。在一些实施例中，所述判断阈值的个数与所述特征参数的个数一致，并与所述特征参数一一相对应。例如，当特征参数为停留次数，对应的判断阈值可以为大于等于2次；当特征参数为停留时间，对应的判断阈值可以是两个时长，一个可以是直行红灯的剩余时间，另一个可以是 直行绿灯的时长。在一些实施例中，所述判断模块420还可以将待测路口移动物体的左转轨迹数据输入到一判断模型中，通过构建的判断模型输出是否存在所述目标道路设施的判断结果。例如，若需要判断某一路口是否存在左转待行区，将获取到的某一段时间内该路口的轨迹数据输入到判断模型中，经过判断模型计算，输出该路口是否具有左转待行区的结果。在一个实施例中，判断模型可以是通过机器学习事先得到的判断模型。
在一个实施例中，如图6所示，处理引擎112还可以包括训练模块430。所述训练模块430可以用于确定判断阈值。在一些实施例中，所述训练模块430还可以用于获取已知路口的移动物体左转行驶轨迹数据，从所述轨迹数据中提取与目标道路设施相关联的特征参数信息，标注已知路口是否存在所述目标道路设施，基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值。在一些实施例中，可以选取若干数量的已知路口作为样本获取轨迹数据和标注结果。例如选取100个路口为已知路口，获取这100个路口的左转轨迹数据，并提取特征参数信息，标注这100个路口是否存在左转待行区。在一些实施例中，标注已知路口是否存在所述目标道路设施可以通过人工现场勘查、交通拍照摄像头、电子地图等方式采集已知路口是否存在目标道路设施的结果。在一些实施例中，已知路口的轨迹数据和标注结果可以从道路信息系统100中获得，例如，已知路口的轨迹数据和标注结果可以是之前保存的由待测路口轨迹数据和对应判断结果转换成的已知数据。在一些实施例中，训练模块430可以根据获取的特征参数信息和已经标注的结果，经过整理、统计、或其他计算方式处理后得到相应的判断阈值。在一些实施例中，判断阈值 也可以是根据实际要求人为确定的经验值。例如，所述停留次数的判断阈值，可根据实际情况合理推断，如前所述，当停留次数小于2次的情况时，该路口很大概率应该是没有左转待行区的，停留次数大于等于2次的情况，该路口很大概率应该具有左转待行区，所以，可以人为直接确定“停留次数”的判断阈值为“≥2”。在一些实施例中，所述训练模块430还可以用于基于已知路口的所述特征参数信息和所述标注的结果训练判断模型，确定所述判断阈值和所述判断模型。例如，获取100个路口的移动物体左转行驶轨迹数据，提取与左转待行区相关的特征参数信息，并标注该100个路口是否存在左转待行区，将100个路口的特征参数信息和标注结果作为训练样本，进行机器学习，得到特征参数信息的判断阈值和判断模型。在一些实施例中，所述判断模型可以为决策树模型，但不限于分类及回归树(Classification And Regression Tree，CART)、迭代二叉树三代(Iterative Dichotomiser 3，ID3)、C4.5算法、随机森林(Random Forest)、卡方自动交互检测(Chisquared Automatic Interaction Detection，CHAID)、多元自适应回归样条(Multivariate Adaptive Regression Splines，MARS)以及梯度推进机(Gradient Boosting Machine，GBM)等或其任意组合。在一些实施例中，在训练过程中，可以利用验证集对模型进行验证，并根据验证结果对模型参数进行调整以使模型达到最佳状态。所述验证集中的数据与所述判断模型的训练数据独立同分布，且没有交集。
应当理解，图4和图5所示的系统及其模块可以利用各种方式来实现。例如，在一些实施例中，系统及其模块可以通过硬件、软件或者软件 和硬件的结合来实现。其中，硬件部分可以利用专用逻辑来实现；软件部分则可以存储在存储器中，由适当的指令执行系统，例如微处理器或者专用设计硬件来执行。本领域技术人员可以理解上述的方法和系统可以使用计算机可执行指令和/或包含在处理器控制代码中来实现，例如在诸如磁盘、CD或DVD-ROM的载体介质、诸如只读存储器(固件)的可编程的存储器或者诸如光学或电子信号载体的数据载体上提供了这样的代码。本申请的系统及其模块不仅可以有诸如超大规模集成电路或门阵列、诸如逻辑芯片、晶体管等的半导体、或者诸如现场可编程门阵列、可编程逻辑设备等的可编程硬件设备的硬件电路实现，也可以用例如由各种类型的处理器所执行的软件实现，还可以由上述硬件电路和软件的结合(例如，固件)来实现。
需要注意的是，以上描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可以在不背离这一原理的情况下，对实施上述方法和系统的应用领域进行形式和细节上的各种修正和改变。例如，获取模块410和训练模块430可以集成在一起成为一个模块，同时实现数据获取以及模型训练的功能。然而，这些变化和修改不脱离本申请的范围。
图7是根据本发明的一些实施例所示的确定路口是否存在目标道路设施的示例性流程图。在一些实施例中，流程700可以通过处理逻辑来执行，该处理逻辑可以包括硬件(例如，电路、专用逻辑、可编程逻辑、微代码等)、软件(运行在处理设备上以执行硬件模拟的指令)等或其任意组合。图7所示的用于判断路口是否存在目标道路设施的流程700中的一 个或多个操作可以通过图1所示的道路信息系统100实现。例如，流程700可以以指令的形式存储在存储设备130中，并由处理引擎112执行调用和/或执行(例如，图2所示的计算设备200的处理器220、图3所示的移动设备300的中央处理器340)。
在710中，获取待测路口的移动物体左转行驶轨迹数据。操作710可以由获取模块410执行。所述行驶轨迹数据可以通过固定或移动的视频采集装置采集的图像信息经过处理转换为数字信息后获得。在一些实施例中，所述行驶轨迹数据可以是通过直接配有定位信息的移动设备采集。在一些实施例中，所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的轨迹数据集，包括所有与移动物体行驶相关的数据信息。例如，行驶线路、行驶时间、速度信息、位置信息等。所述移动物体为可在道路上行驶的可移动的物体，包括但不限于、自行车、马车、人力车、可移动的机器人等。在一个实施例中，所述轨迹数据可以通过安装有定位单元的手持移动设备120-3采集。在一些实施例中，手持移动设备120-3可以包括但不限于智能手机、个人数码助理(Personal Digital Assistance，PDA)、平板电脑、掌上游戏机、智能眼镜、智能手表、可穿戴设备、虚拟显示设备、显示增强设备等或其任意组合。
在720中，可以从所述轨迹数据中提取与目标道路设施相关联的特征参数信息。在一些实施例中，操作720可以由获取模块410执行。在一些实施例中，所述目标道路设施包括但不限于交通信号灯、交通标识(包括禁令标识、行进方式指示标识等)、交叉路口左转待行区、可变车道等其他道路交通设施及其任意组合。在一些实施例中，所述目标道路设施为 左转待行区。在一些实施例中，所述特征参数信息包括进入路口后的行驶参数。在一个实施例中，所述特征参数可以包括停留次数、停留时间、停留距离、延误时间、通过路口的平均速度和停留两次的概率中的至少一个。在一些实施例中，所述路口可以是从左转车道的某一处开始到下一车道的进入端之间，具有一段距离的路面道路，其中，所左转车道的某一处可以为左转车道上，在位于行驶方向前方路口停车线之前具有一定距离的位置。换句话说，所述路口的长度应包括左转车道上某处到所述路口停车线的距离，及从所述路口停车线到下一车道进入端的转弯长度的总和。例如，所述路口可以为从所述左转车道上某一处到下一车道进入端之间，长度为300m距离的一段路面道路。在一些实施例中，所述停留为所述轨迹数据中，至少两个连续所述轨迹点的速度值均小于一设定值时，认为所述停留一次。例如，所述停留可以是的连续三个轨迹点的速度值均小于0.67m/s时，认为发生了一次停留。在一些实施例中，所述停留时间为一次所述停留的时长。例如，停留时间可以为直行红灯的剩余时间，或是直行绿灯的时间。所述停留距离为两次所述停留之间所述移动物体走过的距离。例如，所述停留距离可以是左转待行区的长度值。所述延误时间为所述移动物体实际通过所述路口花费的时间，与所述移动物体在没有发生所述停留的情况下通过所述路口需要的时间的差值。在一些实施例中，所述移动物体在没有发生所述停留的情况下通过所述路口需要的时间，可以通过以下方式得到：获取一段时间内的路口左转轨迹数据，提取没有发生过所述停留的轨迹数据，并通过提取到的轨迹数据计算通过所述路口的平均时间。在一些实施例中，所述移动物体在没有发生所述停留的情况下通过 所述路口需要的时间可以得到更新，例如，可以规定每隔一个月的时间进行一次更新。所述移动物体通过路口的平均速度为所述移动物体通过所述路口的平均速度。所述停留两次的概率为所述停留次数为两次及大于两次的所述行驶轨迹的数量占所选取的所述行驶轨迹的数量总和的比率。在一些实施例中，所述特征参数信息可以是统计数值。例如，停留次数的个数统计，停留时间的均值和方差，停留距离的均值和方差，延误时间的均值和方差，通过路口的平均速度的均值和方差等。
在730中，可以基于所述待测路口的移动物体所述特征参数信息确定所述待测路口是否存在所述目标道路设施。操作730可以由判断模块420执行。在一些实施例中，可以基于所述待测路口的所述特征参数信息确定所述待测路口是否存在左转待行区。在一些实施例中，可以先确定判断阈值，再将所述特征参数信息与对应的判断阈值比较，判断所述路口是否存在所述目标道路设施。如果所述特征参数处于对应的所述判断阈值范围内，则所述路口存在目标道路设施。如果所述特征参数不处于对应的所述判断阈值范围内，则认为所述路口不存在所述目标道路设施。例如，特征参数为停留两次的概率，通过计算，当路口具有左转待行区时，轨迹数据中停留两次的概率为15％-50％之间。当路口没有左转待行区时，轨迹数据中停留两次的概率为<5％。因此，可以判断当轨迹数据中停留两次的概率为3％时，该路口没有左转待行区。当轨迹数据中停留两次的概率为30％时，该路口具有左转待行区。
在一些实施例中，特征参数不止一个时，所述判断阈值的个数与所述特征参数的个数一致，并与所述特征参数一一相对应。例如，当特征参 数为停留次数，对应的判断阈值可以为大于等于2次；当特征参数为停留时间，对应的判断阈值可以是两个时长，一个可以是直行红灯的剩余时间，另一个可以是直行绿灯的时长；当特征参数为停留距离，对应的判断阈值可以是左转待行区的长度；当特征参数为停留两次的概率是，判断阈值可以是大于15％。在一些实施例中，特征参数不止一个时，可以将特征参数进行排序，确定比较判断的先后顺序。例如，先用停留次数进行比较判断，如果停留次数大于等于2次，再进一步比较停留时间是否分别与两个阈值相等。又例如，先用停留两次的概率进行比较，如果停留两次的概率为10％，则认为该路口没有左转待行区，如果停留两次的概率为75％，大于15％的阈值，再进一步比较停留时间是否分别与两个阈值相等。
需要注意的是，以上描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可以在不背离这一原理的情况下，对实施上述方法和系统的应用领域进行形式和细节上的各种修正和改变。
图8是根据本发明的一些实施例所示的确定判断阈值和判断模型的示例性流程图。在一些实施例中，流程800可以通过处理逻辑来执行，该处理逻辑可以包括硬件(例如，电路、专用逻辑、可编程逻辑、微代码等)、软件(运行在处理设备上以执行硬件模拟的指令)等或其任意组合。图8所示的用于确定路口是否存在目标道路设施的判断模型的流程800中的一个或多个操作可以通过图1所示的道路信息系统100实现。例如，流程800可以以指令的形式存储在存储设备130中，并由处理引擎112执行调用和/或执行(例如，图2所示的计算设备200的处理器220、 图3所示的移动设备300的中央处理器340)。
在810中，可以获取已知路口的移动物体左转行驶轨迹数据。操作810可以由训练模块430执行。在一些实施例中，可以选取若干数量的已知路口作为样本获取轨迹数据和标注结果。例如选取100个路口为已知路口，获取这100个路口的左转轨迹数据。在一些实施例中，可以获取一段时间内若干已知路口的左转行驶轨迹数据。所述一段时间可以是一个月、一个季度、一年。在一些实施例中，可以将获取所述待测路口的移动物体左转行驶轨迹为原始轨迹数据，对所述原始轨迹数据进行筛选。在一些实施例中，可以提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据。在一些实施例中，平峰时段可以是排除车流量过高和车流量过低，车流量较稳定的一段时间。例如，一般城市中，平峰时段通常是上午10点到下午16点之间的时段。需要理解的是，如果路口处于拥堵的高峰时段，移动物体的停留次数和停留时间通常会多于平峰时段，行驶速度和通过路口的时间也会有较大的差异，导致数据规律性较差，不利于计算的准确性。另一方面，由于数据采集端220出现工作异常而导致的轨迹数据不连续、有中断等异常状况也会造成干扰，导致计算不准确。因此，通过对所述轨迹数据的筛选，可以提高数据的稳定性和判断结果的准确性。
在820中，可以从所述轨迹数据中提取与目标道路设施相关联的特征参数信息。操作820可以由训练模块430执行。在一些实施例中，所述目标道路设施可以是左转待行区。在一些实施例中，所述特征参数信息可以是统计数值。例如，停留次数的个数统计，停留时间的均值和方差，停 留距离的均值和方差，延误时间的均值和方差，通过路口的平均速度的均值和方差等。
在830中，可以标注已知路口是否存在所述目标道路设施。操作830可以由训练模块430执行。在一些实施例中，标注已知路口是否存在所述目标道路设施可以通过人工现场勘查、交通拍照摄像头、电子地图等方式采集已知路口是否存在目标道路设施的结果。例如，可以通过电子地图中的实景信息标注已知路口是否具有左转待行区。
在840中，可以基于已知路口的所述特征参信息数和所述标注的结果训练判断模型，确定所述判断阈值和所述判断模型。操作840可以由训练模块430执行。在一些实施例中，可以选取若干数量的已知路口获取轨迹数据和标注结果作为训练样本。例如选取100个路口为已知路口，获取这100个路口的左转轨迹数据，并提取特征参数信息，标注这100个路口是否存在左转待行区，将这100个路口的特征参数信息和标注结果作为训练样本，进行机器学习，得到特征参数信息对应的判断阈值和判断模型。在一些实施例中，所述判断模型可以是决策树模型，包括但不限于分类及回归树(Classification And Regression Tree，CART)、迭代二叉树三代(Iterative Dichotomiser 3，ID3)、C4.5算法、随机森林(Random Forest)、卡方自动交互检测(Chisquared Automatic Interaction Detection，CHAID)、多元自适应回归样条(Multivariate Adaptive Regression Splines，MARS)以及梯度推进机(Gradient Boosting Machine，GBM)等或其任意组合。在一些实施例中，可以利用验证集对模型进行验证，并根据验证结果对模型参数进行调整以使模型达到最佳状态。所述验证集中 的数据与所述判断模型的训练数据独立同分布，且没有交集。例如，选取200个交叉路口的轨迹数据作为样本数据，其中100个交叉路口的轨迹数据作为建立模型的训练样本，另100个交叉路口的轨迹数据为验证用样本数据。将验证用样本数据输入至训练好的判断模型，得到的输出结果与实际标注的结果比对，检测判断模型的准确性。在一些实施例中，还可以根据获取的特征参数信息和已经标注的结果，经过整理、统计、或其他计算方式处理后得到相应的判断阈值。在一些实施例中，判断阈值也可以是根据实际要求人为确定的经验值。例如，所述停留次数的判断阈值，可根据实际情况合理推断，如前所述，当停留次数小于2次的情况时，该路口很大概率应该是没有左转待行区的，停留次数大于等于2次的情况，该路口很大概率应该具有左转待行区，所以，可以人为直接确定“停留次数”的判断阈值为“≥2”。
在一些实施例中，图9是一个应用决策树模型来确定路口是否存在左转待行区的示例性算法流程图。在一些实施例中，在步骤901中，可以获取待测路口的移动物体左转行驶轨迹数据。步骤901可以由获取模块410执行。在一些实施例中，所述左转行驶轨迹数据可以通过直接配有定位信息的移动设备采集。在一些实施例中，所述左转行驶轨迹数据可以是保存在存储设备中的一段时间内的左转行驶轨迹数据。在一些实施例中，移动物体可以是道路上行驶的车辆、车载定位装置或其他安装有定位单元的手持移动设备。步骤902中，可以从轨迹数据中提取行驶轨迹数、停留两次的概率、停留时间、停留距离、通过路口的平均速度和延误时间的特征参数信息。步骤902可以由获取模块410执行。在一些实施例中，所述 特征参数信息可以是统计数值。例如，停留次数的个数统计，停留时间的均值和方差，停留距离的均值和方差，延误时间的均值和方差，通过路口的平均速度的均值和方差等。如图9所示，可以选取特征参数为行驶轨迹数、停留两次的概率、停留时间、停留距离的均值和方差、通过路口的平均速度均值和方差以及延误时间。在步骤903中，可以判断行驶轨迹数是否大于第一阈值。步骤903可以由判断模块420执行。在一些实施例中，所述特征参数信息对应的判断阈值(例如，第一阈值、第二阈值…第九阈值)可以通过模型训练得到。在一些实施例中，判断阈值也可以人为规定。例如，可以人为规定第一阈值为100，当行驶轨迹数大于100时，可以继续流程，否则需要增加样本数量(如步骤904)，继续获取待测路口的左转行驶轨迹数据。在步骤905中，可以判断停留两次的概率是否大于第二阈值。步骤905可以由判断模块420执行。在一些实施例中，第二阈值可以通过模型训练得到。在一些实施例中，如果停留两次的概率大于第二阈值，可以继续流程，如果停留两次的概率不大于第二阈值则输出“没有左转待行区”的结果(步骤920)。在步骤906中，可以判断停留时间是否等于第三阈值和第四阈值。步骤906可以由判断模块420执行。在一些实施例中，第三阈值和第四阈值可以通过训练得到。在一些实施例中，第三阈值可以是直行红灯剩余的时间。在一些实施例中，第四阈值可以是直行绿灯的时间。如前所述(如图5-A)，如果路口具有左转待行区，移动物体在路口停留两次的概率较大，第一次停留的时间应约为直行红灯剩余的时间，第二次停留的时间应约为直行绿灯的时间。在一些实施例中，行驶轨迹中有两次停留并且停留时间分别等于第三阈值、第四阈值时，可 以继续判断流程。在一些实施例中，行驶轨迹中有两次停留并且停留时间不等于第三阈值、第四阈值时，则输出“没有左转待行区”的结果(步骤920)。例如，如果移动物体由于拥堵或是其他突发事情导致前后有两次停留，停留时间肯定不符合通过左转待行区的规律，那么即使该行驶轨迹中出现了两次停留，也不能认为该路口具有左转待行区。在步骤907中，判断停留距离的均值是否等于第五阈值。步骤907可以由判断模块420执行。在一些实施中，第五阈值可以通过训练得到。在一些实施例中，第五阈值可以是左转待行区的距离。在一些实施例中，左转待行区的距离可以是左转待行区的直线距离。在一些实施例中，左转待行区的距离可以是左转待行区的实际轨迹距离。如前所述，如果路口具有左转待行区，移动物体在路口的两次停留的距离应约为左转待行区的距离。在一些实施例中，如果停留距离的均值等于第五阈值，可以继续判断流程。在一些实施例中，如果停留距离的均值不等于第五阈值，则输出“没有左转待行区”的结果(步骤920)。在步骤908中，可以判断停留距离的方差是否等于第六阈值。步骤908可以由判断模块420执行。在一些实施中，第六阈值可以通过训练得到。在一些实施例中，如果停留距离的方差等于第六阈值，可以继续判断流程。在一些实施例中，如果停留距离的方差不等于第六阈值，则输出“没有左转待行区”的结果(步骤920)。在步骤909中，可以判断通过路口的平均速度的均值是否等于第七阈值。步骤909可以由判断模块420执行。在一些实施例中，第七阈值可以通过训练得到。如前所述，具有左转待行区的路口移动物体很大概率会发生两次停留，通过路口的平均速度应会与没有左转待行区的路口不同。在一些实施例中，如果通 过路口的平均速度的均值等于第七阈值，可以继续判断流程。在一些实施例中，如果通过路口的平均速度的均值不等于第七阈值，则输出“没有左转待行区”的结果(步骤920)。在步骤910中，可以判断通过路口的平均速度的方差是否等于第八阈值。步骤910可以由判断模块420执行。在一些实施例中，第八阈值可以通过训练得到。在一些实施例中，如果通过路口的平均速度的方差等于第七阈值，可以继续判断流程。在一些实施例中，如果通过路口的平均速度的方差不等于第七阈值，则输出“没有左转待行区”的结果(步骤920)。在步骤911中，可以判断延误时间和信号周期的比值是否等于第九阈值。步骤911可以由判断模块420执行。在一些实施例中，第九阈值可以是通过训练得到。在一些实施例中，延误时间为移动物体实际通过路口花费的时间，与移动物体在没有发生停留的情况下通过路口需要的时间的差值。信号周期可以是当前直行绿灯到下一次直行绿灯出现的间隔时间。在一些实施例中，如果延误时间和信号周期的比值等于第九阈值，则输出“有左转待行区”的结果(步骤921)。在一些实施例中，如果延误时间和信号周期的比值不等于第九阈值，则输出“有左转待行区”的结果(步骤920)。
需要注意的是，以上描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可以在不背离这一原理的情况下，对实施上述方法和系统的应用领域进行形式和细节上的各种修正和改变。
图10是根据本发明的一些实施例所示的确定待判断路口是否存在目标道路设施的示例性流程图。在一些实施例中，流程1000可以通过处 理逻辑来执行，该处理逻辑可以包括硬件(例如，电路、专用逻辑、可编程逻辑、微代码等)、软件(运行在处理设备上以执行硬件模拟的指令)等或其任意组合。图10所示的用于确定待判断路口是否存在目标道路设施的流程1000中的一个或多个操作可以通过图1所示的道路信息系统100实现。例如，流程1000可以以指令的形式存储在存储设备130中，并由处理引擎112执行调用和/或执行(例如，图2所示的计算设备200的处理器220、图3所示的移动设备300的中央处理器340)。
在1010中，可以将所述待测路口的所述轨迹数据输入所述判断模型。操作1010可以由判断模块420执行。在一些实施例中，所述待测路口的所述轨迹数据可以为一段时间内的待判断路口的轨迹数据。一段时间可以是一个月、一个季度、一年等。在一些实施例中，还可以将获取到的路口的行驶轨迹作为初始轨迹数据，根据交通拥堵状况提取所述初始轨迹数据中处于平峰时段，并且轨迹线完整的数据作为所述轨迹数据。以提高数据的稳定性和判断结果的准确性。在一些实施例中，判断模块430可以通过网络140访问存储在存储设备230中的数据，基于所述待判断路口的位置信息获取道路信息系统200中的存档数据，来获取所述待判断路口的所述轨迹数据。还可以将获取所述待测路口的左转行驶轨迹为原始轨迹数据，对所述原始轨迹数据进行筛选。在一些实施例中，可以提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据。在一些实施例中，平峰时段可以是排除车流量过高和车流量过低，车流量较稳定的一段时间。例如，一般城市中，平峰时段通常是上午10点到下午16点之间的时段。
在1020中，可以输出所述待测路口是否存在所述目标道路设施的判断结果。操作1020可以在判断模块420中执行。在一个实施例中，所述判断结果可以用数字“0”或“1”表示。例如，可以设置“1”代表路口具有所述目标道路设施，“0”代表路口没有所述目标道路设施，如果判断结果是有左转待行区时，判断模块420则输出“1”，如果判断结果是没有左转待行区时，判断模块420则输出“0”。
需要注意的是，以上描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可以在不背离这一原理的情况下，对实施上述方法和系统的应用领域进行形式和细节上的各种修正和改变。
本申请实施例可能带来的有益效果包括但不限于：(1)能够准确智能的判断交叉路口的道路设施配置情况，减少人力资源和时间成本的损耗；(2)本发明提供了能够准确判断左转待行区所需要的特征参数，以提高了判断模型的准确性；(3)本发明提供了一种判断模型，利用该判断模型可以准确判断路口是否具有左转待行区。需要说明的是，不同实施例可能产生的有益效果不同，在不同的实施例里，可能产生的有益效果可以是以上任意一种或几种的组合，也可以是其他任何可能获得的有益效果。
需要注意的是，以上描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可以在不背离这一原理的情况下，对实施上述方法和系统的应用领域进行形式和细节上的各种修正和改变。
以上内容描述了本申请和/或一些其他的示例。根据上述内容，本 申请还可以作出不同的变形。本申请披露的主题能够以不同的形式和例子所实现，并且本申请可以被应用于大量的应用程序中。后文权利要求中所要求保护的所有应用、修饰以及改变都属于本申请的范围。
同时，本申请使用了特定词语来描述本申请的实施例。如“一个实施例”、“一实施例”、和/或“一些实施例”意指与本申请至少一个实施例相关的某一特征、结构或特点。因此，应强调并注意的是，本说明书中在不同位置两次或多次提及的“一实施例”或“一个实施例”或“一替代性实施例”并不一定是指同一实施例。此外，本申请的一个或多个实施例中的某些特征、结构或特点可以进行适当的组合。
本领域技术人员能够理解，本申请所披露的内容可以出现多种变型和改进。例如，以上所描述的不同系统组件都是通过硬件设备所实现的，但是也可能只通过软件的解决方案得以实现。例如：在现有的服务器上安装系统。此外，这里所披露的位置信息的提供可能是通过一个固件、固件/软件的组合、固件/硬件的组合或硬件/固件/软件的组合得以实现。
所有软件或其中的一部分有时可能会通过网络进行通信，如互联网或其他通信网络。此类通信能够将软件从一个计算机设备或处理器加载到另一个。例如：从道路信息系统的一个管理服务器或主机计算机加载至一个计算机环境的硬件平台，或其他实现系统的计算机环境，或与提供订单拼成率预测所需要的信息相关的类似功能的系统。因此，另一种能够传递软件元素的介质也可以被用作局部设备之间的物理连接，例如光波、电波、电磁波等，通过电缆、光缆或者空气实现传播。用来载波的物理介质如电缆、无线连接或光缆等类似设备，也可以被认为是承载软件的介质。 在这里的用法除非限制了有形的“储存”介质，其他表示计算机或机器“可读介质”的术语都表示在处理器执行任何指令的过程中参与的介质。
本申请各部分操作所需的计算机程序编码可以用任意一种或多种程序语言编写，包括面向对象编程语言如Java、Scala、Smalltalk、Eiffel、JADE、Emerald、C++、C#、VB.NET、Python等，常规程序化编程语言如C语言、Visual Basic、Fortran 2003、Perl、COBOL 2002、PHP、ABAP，动态编程语言如Python、Ruby和Groovy，或其他编程语言等。该程序编码可以完全在用户计算机上运行、或作为独立的软件包在用户计算机上运行、或部分在用户计算机上运行部分在远程计算机运行、或完全在远程计算机或服务器上运行。在后种情况下，远程计算机可以通过任何网络形式与用户计算机连接，例如，局域网(LAN)或广域网(WAN)，或连接至外部计算机(例如通过因特网)，或在云计算环境中，或作为服务使用如软件即服务(SaaS)。
此外，除非权利要求中明确说明，本申请所述处理元素和序列的顺序、数字字母的使用、或其他名称的使用，并非用于限定本申请流程和方法的顺序。尽管上述披露中通过各种示例讨论了一些目前认为有用的发明实施例，但应当理解的是，该类细节仅起到说明的目的，附加的权利要求并不仅限于披露的实施例，相反，权利要求旨在覆盖所有符合本申请实施例实质和范围的修正和等价组合。例如，虽然以上所描述的系统组件可以通过硬件设备实现，但是也可以只通过软件的解决方案得以实现，如在现有的服务器或移动设备上安装所描述的系统。
同理，应当注意的是，为了简化本申请披露的表述，从而帮助对一 个或多个发明实施例的理解，前文对本申请实施例的描述中，有时会将多种特征归并至一个实施例、附图或对其的描述中。但是，这种披露方法并不意味着本申请对象所需要的特征比权利要求中提及的特征多。实际上，实施例的特征要少于上述披露的单个实施例的全部特征。
一些实施例中使用了描述属性、数量的数字，应当理解的是，此类用于实施例描述的数字，在一些示例中使用了修饰词“大约”、“近似”或“大体上”来修饰。除非另外说明，“大约”、“近似”或“大体上”表明所述数字允许有±20％的变化。相应地，在一些实施例中，说明书和权利要求中使用的数值参数均为近似值，该近似值根据个别实施例所需特点可以发生改变。在一些实施例中，数值参数应考虑规定的有效数位并采用一般位数保留的方法。尽管本申请一些实施例中用于确认其范围广度的数值域和参数为近似值，在具体实施例中，此类数值的设定在可行范围内尽可能精确。
针对本申请引用的每个专利、专利申请、专利申请公开物和其他材料，如文章、书籍、说明书、出版物、文档、物件等，特将其全部内容并入本申请作为参考。与本申请内容不一致或产生冲突的申请历史文件除外，对本申请权利要求最广范围有限制的文件(当前或之后附加于本申请中的)也除外。需要说明的是，如果本申请附属材料中的描述、定义、和/或术语的使用与本申请所述内容有不一致或冲突的地方，以本申请的描述、定义和/或术语的使用为准。
最后，应当理解的是，本申请中所述实施例仅用以说明本申请实施例的原则。其他的变形也可能属于本申请的范围。因此，作为示例而非限制，本申请实施例的替代配置可视为与本申请的教导一致。相应地，本申 请的实施例不限于本申请明确介绍和描述的实施例。

Claims (20)

  1. 一种路口是否存在目标道路设施的判断方法，其特征在于，包括：
    获取待测路口的移动物体左转行驶轨迹数据，
    从所述轨迹数据中提取与目标道路设施相关联的特征参数信息；
    基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施；
    所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的数据集；
    所述目标道路设施包括左转待行区；
    所述特征参数信息包括移动物体进入路口后的行驶参数。
  2. 根据权利要求1所述的方法，其特征在于，其中，基于所述特征参数信息确定所述待测路口是否存在所述目标道路设施，包括:
    确定判断阈值；所述判断阈值的个数与所述特征参数的个数一致，并与所述特征参数一一相对应；
    将所述特征参数信息与对应的所述判断阈值比较，判断所述路口是否存在所述目标道路设施；其中，如果所述特征参数信息处于对应的所述判断阈值范围内，则所述路口存在目标道路设施。
  3. 根据权利要求1所述的方法，其特征在于，所述特征参数包括以下至少一个：
    停留次数、停留时间、停留距离、延误时间、通过路口的平均速度和停留两次的概率；
    其中，所述停留次数为所述轨迹数据中，至少两个连续所述轨迹点的速度值均小于一设定值时，认为所述移动物体停留一次；
    所述停留时间为一次所述停留的时长；
    所述停留距离为两次所述停留之间所述移动物体走过的距离；
    所述延误时间为所述移动物体实际通过所述路口花费的时间，与所述移动物体在没有发生所述停留的情况下通过所述路口需要的时间的差值；
    所述通过路口的平均速度为所述移动物体通过所述路口的平均速度；
    所述停留两次的概率为所述停留次数为两次及大于两次的所述行驶轨迹的数量占所选取的所述行驶轨迹的数量总和的比率。
  4. 根据权利要求2所述的方法，其特征在于，其中，确定判断阈值包括：
    获取已知路口的移动物体左转行驶轨迹数据；
    从所述轨迹数据中提取与目标道路设施相关联的特征参数信息；
    标注已知路口是否存在所述目标道路设施；
    基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值。
  5. 根据权利要求4所述的方法，其特征在于，所述方法进一步包括：
    获取所述已知路口的移动物体左转行驶轨迹为原始轨迹数据；
    提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的原始轨迹数据作为所述轨迹数据；
    所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间。
  6. 根据权利要求4所述的方法，其特征在于，其中，基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值，包括：
    基于已知路口的所述特征参信息数和所述标注的结果训练判断模型，确定所述判断阈值和所述判断模型。
  7. 根据权利要求6所述的方法，其特征在于，其中，基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施，包括：
    将所述待测路口的所述轨迹数据输入所述判断模型，
    输出所述待测路口是否存在所述目标道路设施的判断结果。
  8. 根据权利要求6所述的方法，其特征在于，
    所述判断模型为决策树模型。
  9. 根据权利要求1所述的方法，其特征在于，所述方法进一步包括：
    获取所述待测路口的移动物体左转行驶轨迹为原始轨迹数据；
    提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据；
    所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间。
  10. 一种路口是否存在目标道路设施的判断系统，其特征在于，所述系统包括：
    获取模块，用于获取待测路口的移动物体左转行驶轨迹数据；并从所述轨迹数据中提取与目标道路设施相关联的特征参数信息；
    判断模块，用于基于所述待测路口的所述特征参数信息确定所述待测路口是否存在所述目标道路设施；
    所述轨迹数据为若干轨迹点信息根据时间先后顺序构成的数据集；
    所述目标道路设施包括左转待行区；
    所述特征参数信息包括移动物体进入路口后的行驶参数。
  11. 根据权利要求10所述的系统，其特征在于，所述系统还包括训练模块， 所述训练模块用于确定判断阈值；
    所述判断模块还用于将所述特征参数信息与对应的所述判断阈值比较，判断所述路口是否存在所述目标道路设施；其中，如果所述特征参数处于对应的所述判断阈值范围内，则所述路口存在目标道路设施。
  12. 根据权利要求11所述的系统，其特征在于，
    所述训练模块还用于获取已知路口的移动物体左转行驶轨迹数据；从所述轨迹数据中提取与目标道路设施相关联的特征参数信息；标注已知路口是否存在所述目标道路设施；基于已知路口的所述特征参数信息和所述标注结果确定所述特征参数信息的判断阈值。
  13. 根据权利要求12所述的系统，其特征在于，
    所述训练模块还用于获取所述已知路口的移动物体左转行驶轨迹为原始轨迹数据；提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的原始轨迹数据作为所述轨迹数据；所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间。
  14. 根据权利要求12所述的系统，其特征在于，
    所述训练模块还用于基于已知路口的所述特征参信息数和所述标注的结果训练判断模型，确定所述判断阈值和所述判断模型。
  15. 根据权利要求14所述的系统，其特征在于，
    所述判断模块还用于将所述待测路口的所述轨迹数据输入所述判断模型，输出所述待测路口是否存在所述目标道路设施的判断结果。
  16. 根据权利要求14所述的系统，其特征在于，
    所述判断模型为决策树模型。
  17. 根据权利要求10所述的系统，其特征在于，
    所述获取模块还用于获取所述待测路口的移动物体左转行驶轨迹为原始轨迹数据；提取所述原始轨迹数据中处于平峰时段且所述特征参数信息完整的所述原始轨迹数据作为所述轨迹数据；所述平峰时段是所述路口上去除车流量过高和车流量过低，车流量稳定的一段时间。
  18. 根据权利要求10所述的系统，其特征在于，所述特征参数包括以下至少一个：
    停留次数、停留时间、停留距离、延误时间、通过路口的平均速度和停留两次的概率；
    其中，所述停留为所述轨迹数据中，至少两个连续所述轨迹点的速度值均小于一设定值时，认为所述移动物体停留一次；
    所述停留时间为一次所述停留的时长；
    所述停留距离为两次所述停留之间所述移动物体走过的距离；
    所述延误时间为所述移动物体实际通过所述路口花费的时间，与所述移动物体在没有发生所述停留的情况下通过所述路口需要的时间的差值；
    所述通过路口的平均速度为所述移动物体通过所述路口的平均速度；
    所述停留两次的概率为所述停留次数为两次及大于两次的所述行驶轨迹的数量占所选取的所述行驶轨迹的数量总和的比率。
  19. 一种路口是否存在目标道路设施的判断装置，所述装置包括处理器以及存储器；所述存储器用于存储指令，其特征在于，所述指令被所述处理器执行时，导致所述装置实现如权利要求1至9中任一项所述方法对应的操作。
  20. 一种计算机可读存储介质，其特征在于，所述存储介质存储计算机指令，当计算机读取存储介质中的计算机指令后，计算机运行如权利要求1至9中任意一项所述方法。
PCT/CN2018/111807 2018-10-25 2018-10-25 一种路口是否存在目标道路设施的判断方法及系统 WO2020082284A1 (zh)

Priority Applications (9)

Application Number Priority Date Filing Date Title
CN201880002448.3A CN111386559B (zh) 2018-10-25 2018-10-25 一种路口是否存在目标道路设施的判断方法及系统
EP18812027.3A EP3678108A1 (en) 2018-10-25 2018-10-25 Method and system for determining whether target road facility is present at intersection
JP2018565762A JP2021503106A (ja) 2018-10-25 2018-10-25 交差点における対象道路設備の有無を確定する方法及びシステム
AU2018279045A AU2018279045B2 (en) 2018-10-25 2018-10-25 A method and system for determining whether there is target road facility at intersection
SG11201811243UA SG11201811243UA (en) 2018-10-25 2018-10-25 A method and system for determining whether there is target road facility at intersection
CA3027615A CA3027615A1 (en) 2018-10-25 2018-10-25 A method and system for determining whether there is target road facility at intersection
PCT/CN2018/111807 WO2020082284A1 (zh) 2018-10-25 2018-10-25 一种路口是否存在目标道路设施的判断方法及系统
TW107145159A TWI715898B (zh) 2018-10-25 2018-12-14 路口是否存在目標道路設施的判斷方法、系統、裝置及儲存媒體
US16/221,576 US20200134325A1 (en) 2018-10-25 2018-12-16 Method and system for determining whether there is target road facility at intersection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/111807 WO2020082284A1 (zh) 2018-10-25 2018-10-25 一种路口是否存在目标道路设施的判断方法及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/221,576 Continuation US20200134325A1 (en) 2018-10-25 2018-12-16 Method and system for determining whether there is target road facility at intersection

Publications (1)

Publication Number Publication Date
WO2020082284A1 true WO2020082284A1 (zh) 2020-04-30

Family

ID=70325319

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/111807 WO2020082284A1 (zh) 2018-10-25 2018-10-25 一种路口是否存在目标道路设施的判断方法及系统

Country Status (9)

Country Link
US (1) US20200134325A1 (zh)
EP (1) EP3678108A1 (zh)
JP (1) JP2021503106A (zh)
CN (1) CN111386559B (zh)
AU (1) AU2018279045B2 (zh)
CA (1) CA3027615A1 (zh)
SG (1) SG11201811243UA (zh)
TW (1) TWI715898B (zh)
WO (1) WO2020082284A1 (zh)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019205069A1 (en) * 2018-04-27 2019-10-31 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for updating 3d model of building
KR102155055B1 (ko) * 2019-10-28 2020-09-11 라온피플 주식회사 강화학습 기반 신호 제어 장치 및 신호 제어 방법
CN111597285B (zh) * 2020-05-13 2023-09-15 汉海信息技术(上海)有限公司 路网拼接方法、装置、电子设备及存储介质
CN111882906B (zh) * 2020-07-31 2022-08-12 北京航迹科技有限公司 确定车辆的停车位置的方法、装置、设备和介质
CN112115890B (zh) * 2020-09-23 2024-01-23 平安国际智慧城市科技股份有限公司 基于人工智能的酒驾识别方法、装置、设备及介质
CN112700643A (zh) * 2020-12-21 2021-04-23 北京百度网讯科技有限公司 输出车辆流向的方法、装置、路侧设备以及云控平台
CN113129596B (zh) * 2021-04-28 2022-11-29 北京百度网讯科技有限公司 行驶数据处理方法、装置、设备、存储介质及程序产品
CN113920722B (zh) * 2021-09-23 2023-04-14 摩拜(北京)信息技术有限公司 路口通行状态获取方法、装置、电子设备及存储介质
WO2023084890A1 (ja) * 2021-11-10 2023-05-19 住友電気工業株式会社 情報生成システム、情報生成方法及びコンピュータプログラム
CN114463969B (zh) * 2021-12-22 2023-05-16 高德软件有限公司 红绿灯周期时长的挖掘方法、电子设备及计算机程序产品
CN115240411B (zh) * 2022-06-29 2023-05-09 合肥工业大学 一种城市道路交叉口右转冲突警示线测画方法
CN115311759B (zh) * 2022-07-08 2023-09-05 东风汽车集团股份有限公司 一种车辆耐久目标获取方法、装置、设备及存储介质
CN116777703B (zh) * 2023-04-24 2024-02-02 深圳市普拉图科技发展有限公司 一种基于大数据的智慧城市管理方法和系统
CN117077042B (zh) * 2023-10-17 2024-01-09 北京鑫贝诚科技有限公司 一种农村平交路口安全预警方法及系统
CN117253365B (zh) * 2023-11-17 2024-02-02 上海伯镭智能科技有限公司 一种车辆交通状况自动检测方法和相关装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103413437A (zh) * 2013-07-31 2013-11-27 福建工程学院 一种基于车辆数据采集的道路交叉口转向识别方法及系统
CN104205186A (zh) * 2012-03-16 2014-12-10 日产自动车株式会社 意外情况预测灵敏度判断装置
US20150168157A1 (en) * 2013-12-17 2015-06-18 Volkswagen Ag Method and system for determining parameters of a model for the longitudinal guidance and for the determination of a longitudinal guide for a vehicle
CN105547304A (zh) * 2015-12-07 2016-05-04 北京百度网讯科技有限公司 一种道路识别方法及装置
CN106205120A (zh) * 2015-05-08 2016-12-07 北京四维图新科技股份有限公司 一种提取道路路口交通限制的方法及装置
CN106530708A (zh) * 2016-12-14 2017-03-22 北京世纪高通科技有限公司 一种获取交通限制信息的方法及装置
CN108242167A (zh) * 2016-12-24 2018-07-03 钱浙滨 一种道路交通安全设施信息获取方法、使用方法及装置

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3568768B2 (ja) * 1998-01-20 2004-09-22 三菱電機株式会社 車両位置同定装置
CN1162797C (zh) * 2001-06-05 2004-08-18 郑肖惺 智能化城市交通管理网络系统
US7986339B2 (en) * 2003-06-12 2011-07-26 Redflex Traffic Systems Pty Ltd Automated traffic violation monitoring and reporting system with combined video and still-image data
JP2006162409A (ja) * 2004-12-07 2006-06-22 Aisin Aw Co Ltd 交差点進出道路のレーン判定装置
JP4983335B2 (ja) * 2007-03-28 2012-07-25 アイシン・エィ・ダブリュ株式会社 信号機データ作成方法及び信号機データ作成システム
US9131167B2 (en) * 2011-12-19 2015-09-08 International Business Machines Corporation Broker service system to acquire location based image data
US9471838B2 (en) * 2012-09-05 2016-10-18 Motorola Solutions, Inc. Method, apparatus and system for performing facial recognition
CN104123833B (zh) * 2013-04-25 2017-07-28 北京搜狗信息服务有限公司 一种道路状况的规划方法和装置
US20160055744A1 (en) * 2014-08-19 2016-02-25 Qualcomm Incorporated Systems and methods for traffic efficiency and flow control
US10013508B2 (en) * 2014-10-07 2018-07-03 Toyota Motor Engineering & Manufacturing North America, Inc. Joint probabilistic modeling and inference of intersection structure
KR102289142B1 (ko) * 2014-10-28 2021-08-12 현대엠엔소프트 주식회사 교통정보 제공 방법 및 그 장치
CN105806349B (zh) * 2014-12-31 2019-04-30 易图通科技（北京）有限公司 一种真三维导航转向诱导方法和转向诱导导航设备
CN105788273B (zh) * 2016-05-18 2018-03-27 武汉大学 基于低精度时空轨迹数据的城市交叉口自动识别的方法
CN105788274B (zh) * 2016-05-18 2018-03-27 武汉大学 基于时空轨迹大数据的城市交叉口车道级结构提取方法
CN107990905B (zh) * 2016-10-27 2020-04-10 高德软件有限公司 一种掉头路口的确定方法及装置
CN107742418B (zh) * 2017-09-29 2020-04-24 东南大学 一种城市快速路交通拥堵状态及堵点位置自动识别方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104205186A (zh) * 2012-03-16 2014-12-10 日产自动车株式会社 意外情况预测灵敏度判断装置
CN103413437A (zh) * 2013-07-31 2013-11-27 福建工程学院 一种基于车辆数据采集的道路交叉口转向识别方法及系统
US20150168157A1 (en) * 2013-12-17 2015-06-18 Volkswagen Ag Method and system for determining parameters of a model for the longitudinal guidance and for the determination of a longitudinal guide for a vehicle
CN106205120A (zh) * 2015-05-08 2016-12-07 北京四维图新科技股份有限公司 一种提取道路路口交通限制的方法及装置
CN105547304A (zh) * 2015-12-07 2016-05-04 北京百度网讯科技有限公司 一种道路识别方法及装置
CN106530708A (zh) * 2016-12-14 2017-03-22 北京世纪高通科技有限公司 一种获取交通限制信息的方法及装置
CN108242167A (zh) * 2016-12-24 2018-07-03 钱浙滨 一种道路交通安全设施信息获取方法、使用方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3678108A4 *

Also Published As

Publication number Publication date
CN111386559A (zh) 2020-07-07
JP2021503106A (ja) 2021-02-04
SG11201811243UA (en) 2020-05-28
EP3678108A4 (en) 2020-07-08
TWI715898B (zh) 2021-01-11
AU2018279045A1 (en) 2020-05-14
AU2018279045B2 (en) 2021-01-21
CA3027615A1 (en) 2020-04-25
US20200134325A1 (en) 2020-04-30
TW202016728A (zh) 2020-05-01
CN111386559B (zh) 2022-07-19
EP3678108A1 (en) 2020-07-08

Similar Documents

Publication Publication Date Title
WO2020082284A1 (zh) 一种路口是否存在目标道路设施的判断方法及系统
KR102338270B1 (ko) 전자 지도를 업데이트하기 위한 방법, 장치 및 컴퓨터 판독 가능한 저장 매체
US9697730B2 (en) Spatial clustering of vehicle probe data
CN104200649B (zh) 基于预先申请的交通高峰期路线资源调度分配系统与方法
CN108734955B (zh) 预测路况状态的方法及装置
WO2019205020A1 (zh) 一种路况识别方法、装置及设备
TWI715904B (zh) 用於確定交通物件的處理優先順序的系統、方法和儲存媒體
WO2020132990A1 (zh) 一种识别交叉路口的方法和系统
Jiang et al. SunChase: Energy-efficient route planning for solar-powered EVs
WO2021082464A1 (zh) 预测车辆的目的地的方法和装置
US20200193808A1 (en) Systems and methods for processing traffic objects
CN111489555A (zh) 一种交通运行状态预测方法、装置和系统
CN110765224A (zh) 电子地图的处理方法、车辆视觉重定位的方法和车载设备
CN104299445B (zh) 一种停车空位实时信息采集、处理、发布系统及方法
CN110879862B (zh) 基于物联网的车辆信息查询系统
CN115512336B (zh) 基于路灯光源的车辆定位方法、装置和电子设备
WO2022148166A1 (zh) 电子地图校正方法、导航信息设置方法、导航方法及装置
CN115691165A (zh) 一种交通信号灯调度方法、装置、设备及可读存储介质
WO2019127261A1 (zh) 一种智能轮椅自动驾驶方法、系统及计算机可读介质
CN113344277B (zh) 预测模型训练方法、状态更新方法、装置、设备和介质
CN115285148A (zh) 自动驾驶车速规划方法、装置、电子设备和存储介质
CN114550443A (zh) 路网数据处理方法、设备和可读介质
CN114036411A (zh) 一种路线规划方法、装置、设备及介质
Liu Bi-level optimization algorithm for dynamic reversible lane control based on short-term traffic flow prediction
KR102592628B1 (ko) 분기점 판별 장치 및 그 방법

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2018565762

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2018812027

Country of ref document: EP

Effective date: 20181214

ENP Entry into the national phase

Ref document number: 2018812027

Country of ref document: EP

Effective date: 20181214

ENP Entry into the national phase

Ref document number: 2018279045

Country of ref document: AU

Date of ref document: 20181025

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE