JPWO2016114191A1 - Routing and navigation of autonomous vehicles using passenger docking positions - Google Patents

Abstract

A method and apparatus for routing and navigation of autonomous vehicles by passenger docking positions is disclosed. Routing and navigation of autonomous vehicles using passenger docking locations includes identifying traffic network information of autonomous vehicles that indicate a vehicle traffic network including a first destination, and identifying traffic network information includes a plurality of docking locations. Identification of traffic network information including docking position information indicating, wherein each docking position corresponds to each position of the vehicle traffic network. The autonomously traveling vehicle determines a target docking position for the first destination based on the traffic network information, and uses the traffic network information to identify a route from the starting point of the vehicle traffic network to the target docking position. Use the to travel from the starting point to the target docking position.

Description

  The present disclosure relates to routing and navigation of autonomous vehicles.

    An autonomous traveling vehicle can be controlled autonomously so as to pass a traveling route from a departure point to a destination without direct human intervention. The autonomous traveling vehicle includes a control system capable of generating and maintaining a traveling route and controlling the autonomous traveling vehicle to travel along the traveling route. Therefore, autonomous vehicle routing and navigation methods and devices using passenger docking positions are advantageous.

  Here, aspects, features, elements, implementations and examples for autonomous vehicle routing and navigation using passenger docking positions are disclosed.

  One aspect of the disclosed embodiment is an autonomous traveling vehicle in which routing and navigation of the autonomous traveling vehicle by the passenger docking position is executed. The autonomous vehicle includes a processor configured to execute instructions stored in a non-transitory computer readable medium, identifies traffic network information indicating a vehicle traffic network including a first destination, and The identification includes identifying the traffic network information such that the traffic network information includes docking position information indicating a plurality of docking positions, and each docking position of the plurality of docking positions corresponds to each position in the vehicle traffic network. To do. The processor is configured to execute instructions stored in the non-transitory computer readable medium, and identifies one target docking position from a plurality of docking positions for the first destination based on the traffic network information, The network information is used to identify a route from the starting point of the vehicle traffic network to the target docking location. The autonomous traveling vehicle includes a track controller configured to travel the autonomous traveling vehicle from the departure point to the target docking position using this route.

  Variations or other aspects, features, elements, implementations, and examples of the disclosed methods, apparatus, procedures, and algorithms are described in further detail below.

  Various aspects of the methods and apparatus disclosed herein will become more apparent with reference to the following description and the examples illustrated in the drawings.

FIG. 1 illustrates a portion of an autonomous vehicle that can implement the disclosed aspects, features, and elements.

FIG. 2 illustrates a portion of an autonomous vehicle transportation and communication system that can implement the disclosed aspects, features, and elements.

FIG. 3 illustrates a portion of a vehicle traffic network according to the present disclosure.

FIG. 4 illustrates a portion of a vehicle traffic network including candidate docking locations according to the present disclosure.

FIG. 5 illustrates another portion of a vehicle traffic network including candidate docking locations according to the present disclosure.

FIG. 6 illustrates an autonomous vehicle navigation and routing method using a docking position according to the present disclosure.

FIG. 7 illustrates a method for identifying a candidate docking operation according to the present disclosure.

FIG. 8 illustrates a candidate docking operation filtering method according to the present disclosure.

FIG. 9 illustrates a determination method for a dockable position according to the present disclosure.

FIG. 10 shows a portion of a vehicle traffic network including candidate docking locations according to the present disclosure.

FIG. 11 illustrates a portion of a vehicle traffic network including docking location clusters according to the present disclosure.

FIG. 12 illustrates a portion of a vehicle traffic network that includes a docking location based on a docking location cluster in accordance with the present disclosure.

FIG. 13 illustrates a method for identifying a passenger docking position according to the present disclosure.

FIG. 14 illustrates a method for associating a candidate docking position with a set destination according to the present disclosure.

FIG. 15 illustrates a method for associating a candidate docking position with a candidate destination according to the present disclosure.

FIG. 16 illustrates a method for associating a docking location with a destination using a docking location cluster in accordance with the present disclosure.

FIG. 17 illustrates a portion of a vehicle traffic network that includes candidate docking locations in accordance with the present disclosure.

FIG. 18 illustrates a portion of a vehicle traffic network that includes a road segment in accordance with the present disclosure.

FIG. 19 illustrates a portion of a vehicle traffic network including a vehicle traffic network region according to the present disclosure.

FIG. 20 illustrates a method for associating candidate docking positions by division and candidate destinations according to the present disclosure.

FIG. 21 shows a method for specifying an approach position according to the present disclosure.

FIG. 22 illustrates a portion of a vehicle traffic network including candidate docking locations according to the present disclosure.

FIG. 23 illustrates a navigation and routing method for an autonomous vehicle using a plurality of docking positions according to the present disclosure.

FIG. 24 illustrates a method for identifying an expanded passenger docking location according to the present disclosure.

  An autonomous vehicle can travel from a departure point to a destination without human intervention in a vehicle traffic network. The autonomous traveling vehicle includes a controller capable of routing and walking the autonomous traveling vehicle. The controller can generate a travel route from the departure point to the destination based on vehicle information, environmental information, traffic network information indicating a vehicle traffic network, or a combination thereof. The controller uses the generated route to output a travel route to the track controller that enables the vehicle to travel from the departure point to the destination.

  In the traffic network information according to an embodiment, apart from the parking position at the destination, specific information on the docking position where the autonomous vehicle can stop is excluded so that passengers can get on and off.

  Routing and navigation of autonomous vehicles by passenger docking location includes identifying one or more docking locations within the destination vehicle traffic network indicated in the traffic network information. A docking position is specified based on driving information of a plurality of vehicles such as a manual driving vehicle.

  In some embodiments, the docking position specific information may be excluded from the vehicle traffic network information, and the autonomous vehicle stops so as to be able to operate independently of the destination parking position, such as passengers getting on and off. . The identification of the passenger docking position includes the identification of one or more docking positions of the destination vehicle traffic network indicated in the vehicle traffic network information. A docking position is specified based on driving information of a plurality of vehicles such as a manually operated vehicle.

  In some embodiments, the driving information includes a set destination corresponding to the identified candidate docking operation, and the candidate docking operation is associated with the set destination. In one embodiment, the driving information excludes the set destination, and the association between the passenger docking position and the destination includes specifying the destination based on proximity. Determining a destination based on proximity includes using an entrance location associated with the destination, where the entrance location includes a set entrance location or an estimated entrance location. In one embodiment, associating a passenger docking location with a destination generates a docking location cluster based on the candidate docking location and uses the arithmetic average of the destination docking location cluster as the destination docking location. including.

  In one embodiment, the docking position specifying information may be excluded from the vehicle traffic network information, and the autonomously traveling vehicle stops so as to be able to operate independently of the destination parking position, such as passengers getting on and off. The identification of the passenger docking position includes the identification of one or more docking positions of the destination vehicle traffic network indicated in the vehicle traffic network information. A docking position is specified based on driving information of a plurality of vehicles such as a manually operated vehicle. In some embodiments, the driving information includes a set destination corresponding to the identified candidate docking operation, and the candidate docking operation is associated with the set destination. In one embodiment, the set destination is excluded from the driving information, and the association between the passenger docking position and the destination includes specifying the destination based on proximity. Determining a destination based on proximity includes the use of an entrance location associated with the destination, including the use of a set entrance location or an estimated entrance location. In one embodiment, the association between the passenger docking location and the destination is to generate a docking location cluster based on the candidate docking location, and to use an arithmetic average of the destination docking location cluster as the destination docking location. Including.

  In one embodiment, vehicle docking locations are used to associate passenger docking locations with destinations. Based on the vehicle traffic network information, the association between the passenger docking position and the destination using the vehicle traffic network segmentation includes the identification of the road segment based on the vehicle traffic network information, the identification of the docking position cluster for each road segment, and the Voronoi. The median of each docking location cluster is used as a seed for the vehicle traffic network area to set the space associated with the estimated pedestrian travel time.

  In some embodiments, the vehicle traffic network information may exclude docking location specific information, and the autonomous vehicle stops to be able to operate independently of the destination parking location, such as passengers getting on and off. . The identification of the passenger docking position includes the identification of one or more docking positions of the destination vehicle traffic network indicated in the vehicle traffic network information. A docking position is specified based on driving information of a plurality of vehicles such as a manually operated vehicle. In some embodiments, the driving information includes a set destination corresponding to the identified candidate docking operation, and the candidate docking operation is associated with the set destination. In one embodiment, the set destination is excluded from the driving information, and the association between the passenger docking position and the destination includes specifying the destination based on proximity. Determining a destination based on proximity includes the use of an entrance location associated with the destination, including the use of a set entrance location or an estimated entrance location. In one embodiment, associating the passenger docking location with the destination includes generating a docking location cluster based on the candidate docking location and using an arithmetic average of the destination docking location cluster as the destination docking location. In one embodiment, vehicle docking locations are used to associate passenger docking locations with destinations. Based on vehicle traffic network information, the association between passenger docking positions and destinations using vehicle traffic network partitioning is based on the identification of road segments, the identification of docking position clusters for each road segment, and vehicles based on a partitioning scheme such as Voronoi. The median of each docking location cluster is used as a seed for the area of the vehicle traffic network, and a space for the estimated pedestrian travel time is set.

  In one embodiment, an autonomous vehicle generates a vehicle determination model based on a vehicle traffic network to identify a target docking position, and generates a pedestrian determination model based on a pedestrian traffic network close to the destination. , Including a candidate route from the departure point to the target docking position, generating an extended determination model based on the vehicle determination model and the pedestrian determination model, including a candidate route from the target docking position to the destination entrance position, An augmented route having a pedestrian route is included, and an optimum route is specified based on the augmentation determination model.

  Embodiments of the methods disclosed herein, as well as any portion including those aspects, features, or elements may be used by a tangible fixed computer readable storage medium or computer usable for execution by a general purpose or special purpose computer or processor. The present invention can be implemented in a computer program, software, firmware, or a part thereof, which is stored in a storage medium.

  As used herein, “computer” or “computer equipment” includes any device or combination of devices capable of performing any of the disclosed methods or portions thereof.

  As used herein, “processor” refers to one or more general-purpose processors, one or more dedicated processors, one or more conventional processors, one or more digital signal processors, one or more microprocessors, one One or more microcontrollers, one or more application specific integrated circuits, one or more application specific integrated circuits, one or more field programmable gate arrays, any integrated circuit, or combinations thereof, Indicates one or more processors, such as one or more state machines, or any combination of the above.

  As used herein, “memory” refers to any computer-usable medium, computer that can contain, store, communicate, or transfer any signal or information used or connected by any processor. Means a readable medium or device. For example, one or more read-only memories (ROM), one or more random access memories (RAM), one or more registers, one or more cache memories, one or more semiconductor storage devices, one or more magnetics A medium, one or more optical media, one or more magneto-optical recording media, or any combination thereof may be used as the memory.

  As used herein, “instructions” includes instructions or representations for performing any method disclosed herein, or portions thereof, and may be implemented in hardware, software, or any combination thereof. By way of example, the instructions are stored in a memory executable by a processor to process any of the methods, algorithms, aspects or combinations thereof described herein and can be implemented as information, such as a computer program. In certain embodiments, the instructions or portions thereof may be implemented as a dedicated processor or circuit, including dedicated hardware that implements any of the methods, algorithms, aspects or combinations thereof described herein. In some embodiments, some of the instructions may be distributed across multiple processors on a single or multiple devices that can communicate directly or over a network, such as a local area network, a wide area network, the Internet, or a combination thereof.

  As used herein, "example", "example", "implementation", "aspect", "feature" or "element" means meant for purposes of example, illustration, and illustration. Unless otherwise stated, any example, example, implementation, aspect, feature or element is independent of the other examples, examples, implementations, aspects, features or elements, yet another example, example, Can be used in combination with implementations, aspects, features or elements.

  As used herein, “determined” and “specific” or similar terms may be selected, confirmed, calculated, searched, received, determined, confirmed, acquired, or shown here and described as one Includes any identification or determination using the above devices.

  As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless otherwise specified or apparent from the context, “X includes A or B” is intended to indicate natural inclusive substitution. That is, if X includes A, X includes B, or X includes both A and B, in which case “X includes A or B” is satisfied in any of the above cases. Each component used in this specification and the appended claims means “one or more” components unless specifically stated otherwise or apparent from the context. Should be interpreted.

  For ease of explanation, the drawings and description may include a sequential or series of steps or steps, but the elements of the disclosed method may occur in various orders or simultaneously. In addition, the elements of the disclosed method may appear together with other elements not explicitly presented or described. Moreover, not all elements of the described method may be required to perform the method according to the present disclosure. Although aspects, features, and elements are described in specific combinations, each aspect, feature, or element can be used independently and in various combinations with or in combination with other aspects, features, and elements. It is usable without.

  FIG. 1 illustrates an autonomous vehicle capable of implementing the disclosed aspects, features and elements. In some embodiments, autonomous vehicle 1000 includes a chassis 1100, power train 1200, controller 1300, wheel 1400, or other element or combination of elements of an autonomous vehicle. Although the autonomous vehicle 1000 is simply shown as including four wheels 1400, other propulsion devices such as propellers and treads can be used. In FIG. 1, lines connecting elements such as the power train 1200, the controller 1300, and the wheel 1400 mean information such as data and control signals and / or power such as electric power and torque. For example, the controller 1300 receives power from the power train 1200 and communicates with the power train 1200 and / or the wheel 1400 to control the autonomous traveling vehicle 1000. The control includes acceleration, deceleration, steering or other control of the autonomous traveling vehicle 1000.

  The powertrain 1200 includes other elements of the powertrain, such as a power source 1210, a transmission 1220, a steering portion 1230, an actuator 1240 or suspension, a drive shaft, an axle and an exhaust system, and combinations thereof. Wheel 1400 is shown independently, but may be included in powertrain 1200.

  The power source 1210 may include an engine, a battery, or a combination thereof. The power source 1210 is any device or combination of devices that provides energy, such as power, thermal energy, or kinetic energy. For example, the power source 1210 includes an engine such as an internal combustion engine, an electric motor, or a combination of an internal combustion engine and an electric motor, and operates to supply kinetic energy as power to one or more wheels 1400. In some embodiments, the power source 1210 can be one or more dry cells such as NiCad (NiCd), Nickel Zinc (NiZn), Nickel Metal Hydride (NiMH), Lithium Ion (Li-ion), solar cells, fuel cells or energy supplies. Includes potential energy units, such as other possible devices.

  The transmission 1220 receives energy such as kinetic energy from the power supply 1210 and transmits the energy to the wheel 1400 to supply power. The transmission 1220 can be controlled by the controller 1300 and / or the actuator 1240. The steering unit 1230 can be controlled by the controller 1300 and / or the actuator 1240, and controls the wheel 1400 to steer the autonomously traveling vehicle. The vehicle actuator 1240 receives a signal from the controller 1300 and activates or controls the power source 1210, the transmission 1220, the steering unit 1230, or any combination thereof to operate the autonomous traveling vehicle 1000.

  In some embodiments, the controller 1300 includes a location specifying unit 1310, an electronic communication unit 1320, a processor 1330, a memory 1340, a user interface 1350, a sensor 1360, a communication interface 1370, or any combination thereof. Although shown as a single unit, any one or more elements of the controller 1300 can be integrated into any number of independent physical units. For example, the user interface 1350 and the processor 1330 can be integrated into a first physical unit, and the memory 1340 can be integrated into a second physical unit. Although not shown in FIG. 1, the controller 1300 may include a power source such as a battery. Although shown as independent elements, a location specification unit 1310, an electronic communication unit 1320, a processor 1330, a memory 1340, a user interface 1350, a sensor 1360, a communication interface 1370, or any combination thereof may be combined with one or more electronic It may be integrated into a unit, circuit or chip.

  In one embodiment, processor 1330 may be any existing or future developed device that can operate or process signals or other information, including optical processors, quantum processors, molecular processors, or combinations thereof, or combinations thereof. Including. For example, the processor 1330 may be one or more general-purpose processors, one or more dedicated processors, one or more digital signal processors, one or more microprocessors, one or more controllers, one or more microcontrollers, one or more Integrated circuits, one or more application specific integrated circuits, one or more field programmable gate arrays, one or more programmable logic arrays, one or more programmable logic controllers, one or more state machines, or any of these Includes combinations. The processor 1330 may be operatively coupled to the position location unit 1310, the memory 1340, the communication interface 1370, the electronic communication unit 1320, the user interface 1350, the sensor 1360, the powertrain 1200, or any combination thereof. For example, the processor is operatively coupled to memory 1340 via communication bus 1380.

  Memory 1340 may be any tangible fixed computer-usable medium or computer-readable medium that can contain, store, communicate, or transmit machine-readable instructions and any related information in use or connection of processor 1330. Media can be included. The memory 1340 includes, for example, one or more solid state drives, one or more memory cards, one or more removable media, one or more read-only memories, one or more random access memories, a hard disk, a floppy (registered trademark). ) Discs, including optical discs, can be one or more discs, magnetic or optical cards, or any type of fixed media suitable for storing electronic information, or any combination thereof.

  The communication interface 1370 is any other wired or wireless unit capable of interfacing with a wireless antenna, wired communication port, optical communication port, wired or wireless electronic communication medium 1500 as shown. Although FIG. 1 illustrates a communication interface 1370 that communicates over a single communication link, the communication interface is configured to communicate over multiple communication links. Although FIG. 1 shows a single communication interface 1370, an autonomous vehicle may include any number of communication interfaces.

  The electronic communication unit 1320 can be configured to send and receive signals via a wired or wireless electronic communication medium 1500 such as the communication interface 1370. Although not explicitly shown in FIG. 1, the electronic communication unit 1320 can be connected via any wired or wireless communication medium such as radio frequency (RF), ultraviolet (UV), visible light, optical fiber, wire line, or a combination thereof. And configured to transmit and / or receive. Although FIG. 1 shows a single unit of the electronic communication unit 1320 and the communication interface 1370, any number of electronic communication units and communication interfaces can be used.

  The position specifying unit 1310 can determine positioning information such as longitude, latitude, altitude, traveling direction or speed of the autonomous traveling vehicle 1000. For example, the position specifying unit includes a global positioning system (GPS) unit, wireless triangulation, or a combination thereof. The position specifying unit 1310 includes, for example, information indicating the current traveling direction of the autonomous traveling vehicle 1000, the current position of the autonomous traveling vehicle 1000 in two or three dimensions, the current orientation angle of the autonomous traveling vehicle 1000, or a combination thereof. Can be used to get.

  User interface 1350 includes any unit capable of interfacing with a human, such as a virtual or physical keypad, touchpad, display, touch display, speaker, microphone, video camera, sensor, printer, or any combination thereof. . User interface 1350 may be operatively coupled to the illustrated processor 1330 or any other element of controller 1300. Although the user interface 1350 is shown as a single unit, it may include one or more physical units. For example, the user interface 1350 may include a voice interface that performs voice communication with a person and a touch display that performs communication with a person by visual and touching.

  Sensor 1360 includes one or more sensors, such as a sensor array that operates to provide information that can be used to control an autonomous vehicle. Sensor 1360 provides information regarding characteristics of the current operating state of the vehicle. The sensor 1360 may be, for example, a speed sensor, an acceleration sensor, a rudder angle sensor, a traction-related sensor, a braking-related sensor, an arbitrary sensor, These combinations are included.

  In one embodiment, sensor 1360 includes a sensor that operates to obtain information about the physical surrounding environment of autonomous vehicle 1000. For example, one or more sensors detect road shapes and obstacles such as fixed obstacles, vehicles, and pedestrians. In some embodiments, the sensor 1360 may be a known or future developed one or more video cameras, laser sensing systems, infrared sensing systems, acoustic sensing systems or any suitable on-board environment sensing device, or combinations thereof, Or these are included. In one embodiment, the sensor 1360 and the position specifying unit 1310 are combined.

  Although not shown separately in certain embodiments, autonomous vehicle 1000 includes a track controller. For example, the controller 1300 includes a trajectory controller. The track controller operates to acquire information indicating the current situation of the autonomous traveling vehicle 1000 and the planned route, and based on this information, the track of the autonomous traveling vehicle 1000 is determined and optimized. In an embodiment, the track controller outputs a signal for controlling the autonomous traveling vehicle 1000 so that the autonomous traveling vehicle 1000 follows the track determined by the track controller. For example, the trajectory controller outputs an optimized trajectory that can be supplied to the powertrain 1200 and / or the wheel 1400. In one embodiment, the optimized trajectory is a control input, such as a set of steering angles, with each steering angle corresponding to an operating position or position. In some embodiments, the optimization trajectory is one or more paths, lines, curves, or combinations thereof.

  The one or more wheels 1400 are a steering wheel that rotates at a steering angle controlled by the steering unit 1230, a propeller wheel or an autonomous traveling vehicle to which torque is applied so as to drive the autonomous traveling vehicle 1000 controlled by the transmission 1220. A steering propel wheel for steering and driving 1000 is used.

  Autonomous vehicles include a housing not shown in FIG. 1, a Bluetooth (registered trademark) module, a frequency modulation (FM) wireless unit, a near field communication (NFC) module, a liquid crystal display (LCD) display unit, and an organic light emitting diode. (OLED) includes devices and elements such as display units, speakers or combinations thereof.

  FIG. 2 illustrates a portion of an autonomous vehicle traffic communication system in which aspects, features and elements disclosed herein can be implemented. The autonomous traveling vehicle traffic communication system 2000 includes one or more autonomous traveling vehicles capable of traveling via one or more parts of one or more vehicle traffic networks 2200 such as the autonomous traveling vehicle 1000 shown in FIG. The vehicle 2100 can be communicated via one or more communication networks 2300. Although not explicitly shown in FIG. 2, autonomous vehicles may pass through areas that are not shown or included in the vehicle traffic network, such as off-road areas.

  In one embodiment, the communication network 2300 is, for example, a multiple access system and provides voice communication, data communication, video communication, message communication, or a combination thereof between the autonomous vehicle 2100 and one or more communication devices 2400. To do. For example, the autonomous vehicle 2100 receives information indicating the vehicle traffic network 2200 from the communication device 2400 via the communication network 2300.

  In some embodiments, autonomous vehicle 2100 can communicate via a wired communication link (not shown), wireless communication link 2310/2120, or any number of wired or wireless communication link combinations. For example, as shown in the figure, the autonomously traveling vehicle 2100 can communicate via a terrestrial wireless communication link 2310, a non-terrestrial wireless communication link 2320, or a combination thereof. In one embodiment, terrestrial wireless communication link 2310 includes an Ethernet link, a serial link, a Bluetooth link, an infrared (IR) link, an ultraviolet (UV) link, or any link capable of electronic communication.

  In some embodiments, autonomous vehicle 2100 can communicate with communication network 2300 via access location 2330. The access location 2330 includes a computing device and is configured to communicate with the autonomous vehicle 2100, the communication network 2300, one or more communication devices 2400, or a combination thereof via a wired or wireless communication link 2310/2340. For example, the access position 2330 is a base station, a base station device (BTS), a node B, an extended node B (eNode-B), a home node B (HNode-B), a wireless router, a wired router, a hub, a relay, a switch, or these Any wired or wireless device similar to Although the access location is shown alone, any number of interconnect elements may be included.

  In one embodiment, autonomous vehicle 2100 communicates with communication network 2300 via satellite 2350 or other non-terrestrial communication device. The satellite 2350 includes a computing device and is configured to communicate with the autonomous vehicle 2100, the communication network 2300, one or more communication devices 2400, or a combination thereof via one or more communication links 2320/2360. . Although the satellite is illustrated as a single unit, it may comprise any number of interconnected satellites.

Communication network 2300 is any type of network configured to provide voice, data, or other electronic communication. For example, the communication network 2300 includes a local area network (LAN), a wide area network (WAN), a virtual private network (VPN), a mobile or cellular network, the Internet, or any other electronic communication system. The communication network 2300 is a communication protocol such as Transmission Control Protocol (TCP), User Datagram Protocol (UDP), Internet Protocol (IP), Real Time Transport Protocol (RTP), Hypertext Transport Protocol (HTTP), or a combination thereof. Is used.
Although an electronic communication network is illustrated as a single unit, it may comprise any number of interconnected networks.

  In some embodiments, autonomous vehicle 2100 identifies a portion of vehicle traffic network 2200 or connection conditions. For example, an autonomous vehicle includes a speed sensor, a wheel speed sensor, a camera, a gyroscope, an optical sensor, a laser sensor, a radar sensor, an acoustic sensor, another sensor, a device, or a combination thereof, and a part of the vehicle traffic network 2200 or One or more in-vehicle sensors 2110 such as the sensor 1360 shown in FIG.

  In some embodiments, the autonomous vehicle 2100 may use information communicated via the communication network 2300, such as information indicative of the vehicle traffic network 2200, information identified by one or more in-vehicle sensors 2110, or a combination thereof. , Passing through part of one or more vehicle traffic networks 2200 or part of one or more.

  For ease of explanation, FIG. 2 shows one autonomous vehicle 2100, one vehicle traffic network 2200, one communication network 2300, and one communication device 2400, but any number of autonomous vehicles A vehicle, network or computing device may be used. In some embodiments, the autonomous vehicle traffic communication system 2000 includes devices, units or elements not shown in FIG. Although autonomous vehicle 2100 is shown as a single unit, any number of interconnected autonomous vehicles can be included.

  FIG. 3 illustrates a portion of a vehicle traffic network according to the present disclosure. The vehicle traffic network 3000 includes one or more non-navigable areas such as a building 3100, one or more partially walkable areas such as a parking area 3200, one or more walkable areas such as roads 3300/3400, or combinations thereof. Including. In some embodiments, autonomous vehicles such as the autonomous vehicle 1000 shown in FIG. 1 and the autonomous vehicle 2100 shown in FIG. 2 partially pass through the vehicle traffic network 3000.

  The vehicle traffic network includes one or more interchanges 3210 between one or more walkable or partially walkable areas 3200/3300/3400. For example, part of the vehicle traffic network shown in FIG. 3 includes an interchange 3210 between a parking area 3200 and a road 3400. In some embodiments, parking area 3200 includes parking area 3220.

  A portion of the vehicle traffic network, such as road 3300/3400, includes one or more lanes 3320/3340/3360/3420/3440 and is associated with one or more travel directions as indicated by the arrows in FIG. .

  In some embodiments, a vehicle traffic network, or a portion thereof, such as the vehicle traffic network shown in FIG. 3, represents traffic network information. For example, traffic network information is expressed as a hierarchy of elements such as markup language elements that can be stored in a database or file. In short, the figure represents the traffic network information indicating a part of the vehicle traffic network as a diagram or a map. However, the traffic network information can be expressed in any computer-usable format that can indicate a vehicle traffic network or a portion thereof. In some embodiments, the traffic network information includes vehicle traffic network control information such as travel direction information, speed limit information, toll information, slope information such as tilt or angle information, surface material information, aesthetic information, or combinations thereof.

  In some embodiments, a portion of the vehicle traffic network or a combination thereof is identified as a point of interest or destination. For example, the traffic network information specifies the building 3100 and the adjacent partially walkable parking area 3200 as points of interest, and the autonomous vehicle specifies this position as the destination. An autonomous vehicle travels from a departure point to a destination by passing through a vehicle traffic network. Although the parking area 3200 associated with the building 3100 is shown adjacent to the building 3100 in FIG. 3, the destination includes, for example, a building and a parking area that is not physically adjacent to the building. In one embodiment, specifying the destination includes specifying the location of the destination, which is uniquely identifiable geographical location information. For example, the vehicle traffic network includes a destination setting position such as an address, a postal address, a vehicle traffic network address, or a GPS address.

  In some embodiments, the destination is associated with one or more entrances, such as entrance 3500 in FIG. In some embodiments, the traffic network information includes set entrance location information, such as information that identifies the geographic location information of the entrance associated with the destination. In one embodiment, the predicted entrance position information is determined as described above.

  In some embodiments, the vehicle traffic network is associated with or includes a pedestrian traffic network. For example, FIG. 3 includes a portion 3600 of a pedestrian traffic network such as a pedestrian road. In one embodiment, the pedestrian traffic network or a portion thereof, such as the pedestrian traffic network portion 3600 shown in FIG. 3, is shown as pedestrian traffic network information. In one embodiment, the traffic network information includes pedestrian traffic network information. The pedestrian traffic network includes a pedestrian walkable area. Pedestrian walkable areas such as pedestrian roads and sidewalks correspond to non-navigable areas of the vehicle traffic network. Although not individually shown in FIG. 3, a pedestrian walkable area such as a pedestrian crossing corresponds to a walkable area or a partial walkable area of the vehicle traffic network.

  In some embodiments, the destination is associated with one or more docking locations, such as docking location 3700 shown in FIG. The docking position 3700 is a designated or undesignated position or area close to a destination capable of stopping, stopping, or parking an autonomous vehicle that is a docking operation that enables passengers to get on and off.

  In some embodiments, the traffic network information includes docking location information, such as information identifying geographic location information of one or more docking locations 3700 associated with the destination. In one embodiment, the docking location information is docking location information configured to be manually incorporated into the traffic network information. For example, the set docking position information is included in the traffic network information based on the user input. In some embodiments, the docking position information is docking position information that is automatically generated as described herein. Although not shown individually in FIG. 3, the type of docking operation associated with the docking position 3700 is specified by the docking position information. For example, the destination is associated with a first docking position where the passenger gets on and a second docking position where the passenger gets off. The autonomous vehicle can park at the docking position, but the docking position associated with the destination is independent of the parking area associated with the destination and is different.

  In one embodiment, the autonomous vehicle identifies a point of interest including a building 3100, a parking area 3200, and an entrance 3500 as a destination. The autonomous vehicle specifies the building 3100 or the entrance 3500 as the first destination of the point of interest, and specifies the parking area 3200 as the second destination. The autonomous vehicle specifies the docking position 3700 as the docking position of the first destination. The autonomous vehicle generates a route from a departure point (not shown) to the docking position 3700. The autonomously traveling vehicle passes through a vehicle traffic network from the departure point to the docking position 3700 by a route. The autonomous vehicle may be stopped or parked at the docking position 3700 so that passengers can get on and off. The autonomously traveling vehicle generates a subsequent route from the docking position 3700 to the parking area 3200, passes through the vehicle traffic network of the parking area 3200 from the docking position 3700 on the subsequent route, and parks in the parking area 3200.

  FIG. 4 illustrates a portion of a vehicle traffic network that includes a docking location in accordance with the present disclosure. The portion of the vehicle traffic network 4000 shown in FIG. 4 is similar to the portion of the vehicle traffic network 3000 shown in FIG. 3 with the addition or modification of the features described herein.

  The vehicle traffic network 4000 includes one or more non-navigable areas such as a building 4100, one or more partially walkable areas such as a parking area 4200, one or more walkable areas such as roads 4300/4400, or combinations thereof. The vehicle traffic network includes one or more interchanges 4210 between one or more walkable or partially walkable areas 4200/4300/4400. For example, the portion of the vehicle traffic network shown in FIG. 4 includes an interchange 4210 between a parking area 4200 and a road 4400. In some embodiments, parking area 4200 includes parking area 4220. A portion of the vehicle traffic network, such as road 4300/4400, includes one or more lanes 4320/4340/4360/4420/4440 and is associated with one or more travel directions indicated by arrows in FIG. In some embodiments, a portion of the vehicle traffic network or a combination thereof is identified as a point of interest or destination. For example, the traffic network information specifies a building 4100 and a partially walkable parking area 4200 adjacent thereto as a destination. In some embodiments, the destination is associated with one or more entrances, such as entrance 4500. In some embodiments, the vehicle traffic network includes a pedestrian walkable area, such as a pedestrian walkable area 4600 (eg, a pedestrian road).

  In some embodiments, an autonomous vehicle such as the autonomous vehicle 1000 shown in FIG. 1 or the autonomous vehicle 2100 shown in FIG. 2 passes through one or more portions of the vehicle traffic network 4000.

  In one embodiment, the traffic network information includes automatically generated docking position information that is generated based on vehicle operation information, excluding set docking position information for one or more destinations.

  Vehicle operation information includes driving information generated for one or more vehicles that are manned driving vehicles, such as vehicle probe data, vehicle location information, vehicle status information, vehicle event information, controller area network (CAN) data, etc. Includes vehicle bus information or any other information generated by vehicle operation. The driving information includes information indicating vehicle operation. The vehicle operation includes an event indicator including a type or event of the vehicle operation such as start, stop, stop, parking, door opening / closing, and getting on / off. Vehicle operation includes date and / or time. Vehicle operations include locations such as GPS locations within the vehicle traffic network. The vehicle operation includes vehicle status information such as the current number of passengers, a change in occupancy, a change in the number of passengers, or a passenger riding state. In one embodiment, the automatic generation of docking position information includes filtering of vehicle operation information.

  In one embodiment, the docking position information is automatically generated based on vehicle operation information including information generated according to the set vehicle type. For example, the driving information includes a vehicle type indicator indicating whether the vehicle is a transport vehicle such as a taxi or a courier vehicle, and the driving information is filtered in order to exclude the driving information of the non-transporting vehicle. In another example, the driving information includes a vehicle driving type indicator that indicates whether the vehicle is a low-occupancy transportation vehicle such as a taxi or a courier vehicle, and the driving information is driven as a transportation vehicle that is not low-occupancy. Filtered to exclude driving information of existing vehicles.

  In one embodiment, the automatic generation of docking position information includes identification of the docking position based on vehicle operation information. For example, the vehicle operation information indicates a vehicle operation including a stationary time such as the time from the stop to the start of the vehicle specified as the candidate docking operation, and the corresponding position is specified as the candidate docking position. For the sake of brevity, each candidate docking position is indicated by the symbol X in FIGS. 4-5, 10-11, 17, 19 and 22. For example, FIG. 4 includes 11 codes X4700 / 4710/4720 indicating 11 candidate docking positions.

  In some embodiments, candidate docking locations are filtered based on one or more metrics, such as rest time, vehicle status or location information. For example, candidate docking positions 4710 close to the intersection of roads 4300/4400 are filtered or excluded from candidate docking positions indicated by the traffic network information.

  FIG. 5 illustrates another portion of a vehicle traffic network that includes candidate docking locations according to the present disclosure. 5 includes one or more non-navigable areas such as a building 5100, one or more walkable areas such as a road 5200, and one or more parts such as a parking area 5300. It includes a walkable area, one or more pedestrian walkable areas 5400, one or more candidate docking positions 5500, or combinations thereof. In one embodiment, the traffic network information includes scheduled entrance information shown as white triangles 5110 shown in FIG. In one embodiment, the traffic network information includes set entry information shown as black diamonds 5120 in FIG.

  FIG. 6 illustrates an autonomous vehicle navigation and routing method according to docking positions according to the present disclosure. Navigation and routing of the autonomous vehicle using the docking position is performed by an autonomous vehicle such as the autonomous vehicle 1000 shown in FIG. 1 or the autonomous vehicle 2100 shown in FIG. For example, the processor 1330 of the controller 1300 of the autonomous traveling vehicle 1000 shown in FIG. 1 uses the docking position in accordance with instructions stored in the memory 1340 of the controller 1300 of the autonomous traveling vehicle 1000 shown in FIG. Perform routing. Navigation and routing of autonomous vehicles using docking locations can be done by identifying traffic network information at 6100, determining target docking locations at 6200, identifying routes at 6300, traveling at 6400, or a combination thereof including.

  In one embodiment, traffic network information as shown in FIG. For example, an autonomous vehicle control unit such as the controller 1300 shown in FIG. 1 receives traffic network information from a data storage unit such as the memory 1340 shown in FIG. 1 via a communication system such as the communication network 2300 shown in FIG. Read or receive traffic network information from an external data source such as communication device 2400 shown in FIG. In some embodiments, the traffic network information includes docking location information indicating one or more docking locations within the network. In one embodiment, the autonomous vehicle identifies the traffic network information such that the traffic network information includes set docking position information and / or automatically generated docking position information.

  In certain embodiments, identifying traffic network information includes transcoding traffic network information, reformatting and / or storing reformatted traffic network information.

  In one embodiment, the destination is identified at 6200. The destination is specified by specifying a destination such as the building 3100 shown in FIG. 3, the building 4100 shown in FIG. 4 and the building 5100 shown in FIG. 5 as the first destination, and the parking area associated with the point of interest as the second destination. This includes specifying as a place or specifying first and second destinations. In one embodiment, the target docking location of the first destination in the vehicle traffic network is identified at 6200 based on the traffic network information. For example, a building such as the building 3100 shown in FIG. 3 is specified as the first destination, and a docking position such as the docking position 3700 shown in FIG. 3 is specified as the target docking position.

  In one embodiment, the automatic generation of docking position information includes the identification of a series of events indicated by vehicle driving information. For example, the driving information includes an event indicator that represents a series of events, including a stop and a subsequent start, and identifying a series of actions includes determining a rest time as the time difference between the stop and start. In one embodiment, identifying the candidate docking location includes identifying traffic network information corresponding to the vehicle traffic network location indicated in the driving information.

  At 6300, a route is generated. In one embodiment, generating the route includes specifying a starting point. For example, the departure point indicates a target start point such as the current position of the autonomous vehicle. In one embodiment, the departure point identification includes control of a position identification unit, such as 1310 shown in FIG. 1, to determine the current geographic position of the autonomous vehicle. In one embodiment, identifying the departure point at 6300 includes identifying traffic network information corresponding to the departure point. For example, specifying the departure point includes specifying a road, a road segment, a lane, an intermediate position, or a combination thereof. In some embodiments, the current position of the autonomous vehicle is in a navigable non-road area, such as an off-road area, or an area that is not displayed or not fully contained in the vehicle traffic network, and The identification includes identification of a road, a road segment, a lane, an intermediate position, or a combination thereof close to or close to the current position of the autonomous vehicle. The generation of the route includes the determination of the route from the departure point to the target docking location identified at 6200.

  In one embodiment, generating the route includes generating a candidate route from the starting point to the target docking location. In one embodiment, the candidate route indicates a unique or individual route from the departure point to the target docking location. For example, candidate routes include unique or individual combinations of roads, road segments, lanes, intermediate locations, and interchanges.

  In one embodiment, generating the route includes specifying the routing state. In one embodiment, specifying the routing state includes specifying a routing state corresponding to each intermediate location for each candidate route. For example, the first routing state indicates roads, road segments, lanes, intermediate positions or combinations of the first candidate routes, and the second routing state indicates roads, road segments, lanes, intermediate positions or combinations of the second candidate routes. The combination of is shown.

  In one embodiment, route generation includes a cost estimate that is expected to transition from one routing state to another. This corresponds to a movement from one intermediate position to another intermediate position during routing, and is assumed to be a cost of traveling an autonomous vehicle from the position of the first intermediate position to another position of the second intermediate position. Indicates. In one embodiment, an operation is shown that indicates a move from a routing state to the most recent routing state. This corresponds to the movement from one intermediate position to the nearest intermediate position without going through another relay position, and the autonomous vehicle travels from the position of the first intermediate position to another position that is the nearest intermediate position. It means to do.

  In one embodiment, the operating cost is determined based on traffic network information. For example, within the candidate route, the first routing state corresponds to the first relay position and corresponds to the first position of the vehicle traffic network. The second routing state corresponds to the second relay position and corresponds to the second position of the vehicle traffic network. The operation cost indicates an estimated cost, a predicted cost, or an assumed cost associated with the traveling of the autonomous vehicle from the first position to the second position. In some embodiments, the operating cost depends on the situation. For example, the operation cost for movement between two relay positions at a certain time of the day may be considerably higher than the operation cost for movement between two relay positions at another time of the day.

  In some embodiments, route generation includes generation of a probability distribution. In one embodiment, generating the probability distribution includes generating a reasonable cost allocation for operations such as transitioning from one routing state to another. The generation of a reasonable cost allocation includes determining the success probability of the operation, the failure probability of the operation, a plurality of costs assumed by the execution of the operation, an estimated cost associated with the probability of a possible cost, or a combination thereof.

  In one embodiment, the generation of a probability distribution includes the use of a normal distribution, a Gaussian distribution, N (μ, σ), where μ is the arithmetic mean of the normal distribution and σ is the standard deviation. The arithmetic mean and standard deviation of the normal distribution vary depending on the operation. In one embodiment, the standard deviation is increased based on an operational cost uncertainty deviation modifier. This means a variation in the uncertainty of operating costs.

  In one embodiment, generating the probability distribution includes generating a combination of individual cost probabilities of actions. For example, for an action in a route, the generation of a probability distribution is a first estimated cost as a combination of first operation costs such as 45, a first probability generation such as 0.05, and a second action such as 50. Including a second estimated cost as a combination of costs and generation of a second probability such as 0.08.

  In one embodiment, generating a probability distribution includes the use of a linear model of resources and costs. For example, the probability distribution of travel time associated with the motion is represented by a piecewise constant function, and the cost for the motion is represented by a piecewise linear function.

  In some embodiments, determining operating costs includes calculating cost metrics such as distance cost metrics, time cost metrics, fuel cost metrics, acceptable cost metrics, or combinations thereof. In certain embodiments, cost metrics are dynamically determined or generated, stored in a memory, such as a database, and accessed from the memory. In some embodiments, determining the operating cost includes calculating a cost function based on one or more metrics. For example, the cost function minimizes the distance cost metric, the time cost metric, the fuel cost metric, etc., and maximizes the allowable cost metric.

  The distance cost metric indicates the distance from the first position of the first relay position corresponding to the first routing state to the second position of the second relay position corresponding to the second routing state.

  The time cost metric indicates the predicted travel time from the first position of the first relay position corresponding to the first routing state to the second position of the second relay position corresponding to the second routing situation, and the autonomous traveling vehicle and Based on vehicle traffic network status information, the information includes fuel efficiency information, assumed initial speed information, assumed average speed information, assumed final speed information, road surface information or any other information relating to travel time. .

  The fuel cost metric indicates the predicted fuel utilization rate for the transition from the first routing state to the second routing state, and based on the state information of the autonomous vehicle and the vehicle traffic network, the information includes fuel consumption information, assumed initial Includes speed information, assumed average speed information, assumed final speed information, road surface information, or any other information regarding fuel cost.

  The allowable cost metric indicates the predicted travel tolerance from the first position of the first relay position corresponding to the first routing state to the second position of the second relay position corresponding to the second routing state, and is autonomous. Based on the status information of the traveling vehicle and the vehicle traffic network, the information may include expected initial speed information, predicted average speed information, predicted final speed information, road surface information, aesthetic information, toll information, or any other information relating to travel tolerance. Including. In some embodiments, the acceptable cost metric is based on an admissibility factor. In one embodiment, the acceptability factor indicates a location and includes a specified road or area, such as an unacceptable or negative industrial area, or a road type, such as an unpaved road or a toll road, This indicates that the location tolerance factor is high, that is, positive, such as a road with good scenery.

  In one embodiment, the cost metric calculation includes a cost metric weighting and an operating cost calculation based on the weighted cost metric. The weighting of the cost metric includes the calculation of a weighting factor associated with the cost metric. For example, calculating the weighting factor includes accessing a record that indicates the weighting factor and associating the weighting factor with a cost metric. In one embodiment, the cost metric weighting includes generating a weighted cost metric based on the weighting factor and the cost metric. For example, the weighted cost metric is the product of the weighting factor and the cost metric. In one embodiment, the operating cost estimate includes a calculation that determines a total cost metric or a weighted cost metric total.

  In some embodiments, generating a route includes identifying an optimal route. The identification of the optimum route includes selection of candidate routes from a group of candidate routes based on the probability distribution. For example, the candidate route with the smallest route cost estimate is identified as the optimum route. In some embodiments, identifying the optimal route includes the use of a constant time probability control process, such as a hybrid Markov decision process.

  In one embodiment, the optimal route is identified by an operating cost probability distribution during the transition from the first routing state to the second routing state and an operating cost probability distribution during the transition from the first routing state to the third routing state. Includes the selection of the lowest assumed operating cost.

  In one embodiment, identifying the optimal route includes generating a route cost probability distribution for the candidate route based on an operational cost probability distribution for each operation in the route. In one embodiment, identifying the optimal route includes generating a route cost probability distribution for each candidate route and selecting the candidate route with the lowest or lowest route cost estimate as the optimal route.

  In some embodiments, the controller stores candidate routes and / or optimal routes in output or memory. For example, the controller stores the candidate route and the optimum route in the memory, outputs the optimum route to the track controller, the vehicle actuator, or a combination thereof, and causes the autonomously traveling vehicle to travel from the departure point to the target docking position by the optimum route.

  In one embodiment, at 6400, the autonomous vehicle travels from the departure point to the target docking position by an optimal route. For example, the autonomously traveling vehicle includes a vehicle actuator such as the actuator 1240 shown in FIG. 1, and the vehicle actuator operates the autonomously traveling vehicle and starts traveling from the departure point to the target docking position by the optimum route. In one embodiment, the autonomous vehicle includes a track controller that controls the autonomous vehicle to start traveling based on the optimal route, the current operating characteristics of the autonomous vehicle, and the physical environment around the vehicle. To do.

  In some embodiments, the optimal route is updated. In one embodiment, updating the optimal route includes updating or regenerating the candidate route and probability distribution, and identifying the updated optimal route from the updated or regenerated candidate route and probability distribution.

  In one embodiment, the optimal route is the updated traffic network information, the difference between the actual travel cost and the cost estimate of the selected route, and the updated traffic network information, the actual travel cost and the cost of the selected route. It is updated based on the combination with the difference from the estimated value.

  In one embodiment, the autonomous vehicle receives current vehicle traffic network state information before or during travel. In one embodiment, the autonomous vehicle receives current vehicle traffic network status information, such as sensor information outside the vehicle, either directly from the sensors outside the vehicle or via a network, such as the communication network 2300 shown in FIG. In one embodiment, the optimal route is updated according to current vehicle traffic network status information. For example, the current vehicular traffic network status information indicates a change in the status of a part of the vehicular traffic network included in the optimal route, such as a change from an open state to a closed state. It includes the exclusion of candidate routes including obstructions, generation of new candidate routes and probability distributions using the current position of the autonomous vehicle as the departure point, and the update of the optimum route is a new optimum route from the new candidate route. Including identification.

  In one embodiment, the autonomous vehicle completes traveling from the current position of the autonomous vehicle to the target docking position using the updated optimal route.

  In one implementation, the docking position in the vehicle traffic network is identified by the traffic network information identified at 6100. For example, the traffic network information includes set docking position information and / or automatically generated docking position information. The automatically generated docking position information is generated based on the vehicle operation information of a plurality of vehicles. Generation of docking position information based on vehicle operation information includes filtering or specifying driving information for specifying candidate docking operations and corresponding candidate docking positions shown in FIGS.

  FIG. 7 illustrates a method for identifying a candidate docking operation according to the present disclosure. In one embodiment, identifying candidate docking operations includes filtering or identifying driving information for multiple vehicles. Implementations for identifying candidate docking actions include: identifying a vehicle at 7100, identifying a vehicle operation at 7200, identifying a stop time at 7300, identifying a subsequent start time at 7400, candidate docking action at 7500 Specific or combinations thereof. In one embodiment, the candidate docking operation is a docking operation performed at the target docking position.

  The vehicle is identified at 7100. The vehicle operation information includes driving information of a plurality of vehicles, and the specification of the candidate docking operation includes the specification of the vehicle from the vehicle operation information. For example, the vehicle operation information includes records, each record is associated with a vehicle identifier, and each vehicle is uniquely identified based on the vehicle identifier.

  Vehicle events are identified at 7200, just as vehicles are identified at 7100. For example, the vehicle operation information includes information indicating vehicle events such as stop and start. In one embodiment, event identification includes event sequencing. For example, ordering of events by time is performed.

  Based on the event identified at 7200, etc., the downtime is calculated at 7300. Subsequent start-up times are calculated at 7400 based on the calculated time, such as the event and stop time calculated at 7300. For example, the subsequent start time corresponds to the most recent start operation after the stop time calculated at 7300.

  A candidate docking operation is identified at 7500. For example, the identification of the candidate docking operation includes the identification of the vehicle operation information for the vehicle specified by 7100 corresponding to the stationary time from the stop time calculated by 7300 to the start time calculated by 7400. Includes identifying the vehicle position during the calculated rest time. For example, the stationary time is calculated as a time difference between the stop time calculated in 7300 and the start time calculated in 7400. Candidate docking operations and corresponding candidate docking positions are filtered as shown in FIGS.

  FIG. 8 illustrates a candidate docking operation filtering method according to the present disclosure. Candidate docking motion filtering includes: calculation of rest time at 8100, specification of vehicle type and / or vehicle operation type at 8200, calculation of vehicle occupancy at 8300, specification of vehicle status information at 8400, Includes identifying dockable locations, including or excluding certain operations from candidate docking operations at 8600, or combinations thereof. In some embodiments, the candidate docking operation is a docking operation that is performed at the target docking position.

  The rest time of the candidate docking operation is calculated at 8100. In some embodiments, calculating the quiesce time includes determining the quiesce time. For example, the stationary time is calculated based on the driving information, and includes a stop time and a subsequent start time as shown in FIG. The calculation of the stationary time includes a determination as to whether the stationary time is longer than the set minimum docking time, and filtering operation information for excluding candidate docking operations whose stationary time is shorter than the set minimum docking time. The calculation of the stationary time includes a determination as to whether the stationary time is smaller than the set maximum docking time, and filtering of driving information for excluding candidate docking operations whose stationary time exceeds the set maximum docking time.

  For example, the stationary time is larger than the minimum docking time indicating the minimum time of the docking operation, and the position corresponding to the vehicle operation is specified as the candidate docking position. Positions corresponding to rest times shorter than the minimum docking time are excluded from candidate docking positions.

  In another example, the stationary time is within a maximum docking time that indicates the maximum time of the docking operation, and the position corresponding to the vehicle operation is identified as a candidate docking position. A position corresponding to a stationary time longer than the maximum docking time is excluded from the candidate docking positions. For example, the driving information of the candidate docking position 4720 of the parking area 4200 shown in FIG. 4 indicates a parking operation and / or a stationary time exceeding the maximum docking time, and the position is excluded from the candidate docking positions.

  The vehicle type and / or the vehicle operation type is specified by 8200. The vehicle type is specified by determining whether the type of the vehicle represents a set vehicle such as a transport vehicle, and filtering of the drive information to exclude candidate docking operations not related to the transport vehicle. Including. The identification of the vehicle operation type is related to the determination whether the driving information represents a set vehicle operation type such as a low-occupancy transport vehicle and the vehicle operation as a low-occupancy transport vehicle. And filtering of driving information to exclude candidate docking operations that are not performed. In one embodiment, identifying the vehicle operation type includes filtering driving information to exclude candidate docking operations that are not associated with a transport vehicle as a low occupancy transport vehicle.

  Vehicle occupancy is identified by 8300. For example, the driving information includes passenger detection information such as the number of vehicle passengers, driver seat occupancy information or passenger seat occupancy information, or occupancy information indicating a combination of vehicle and seat occupancy information, and the candidate docking position is based on a change in occupancy Identified. For example, a position related to a vehicle operation in which the stationary time is larger than the minimum docking time and within the maximum docking time and the occupancy does not change is excluded from the candidate docking positions. In some embodiments, the change in occupancy is indicated based on a signal from a vehicle sensor, such as a passenger detection sensor. For example, the calculation of vehicle occupancy includes filtering of driving information to exclude candidate docking operations that are not associated with changes in passenger occupancy signals.

  Vehicle status information is used at 8400. For example, the driving information includes half-door information such as half-open information of the vehicle door on the passenger seat side, and the candidate docking position is specified based on the half-door information. For example, identification of vehicle state information includes filtering of driving information to exclude candidate docking operations that are not associated with half-door information.

  The dockable position is specified by 8500. For example, as shown in FIG. 9, the dockable position is specified based on adjacency with a pedestrian road, tolerance of stopping or getting on, distance from an intersection, or a combination thereof.

  Vehicle operation is included in the candidate docking operation at 8600 based on one or more metrics described in 8100, 8200, 8300, 8400, or 8500. In some embodiments, one or more metrics are excluded. For example, the metrics 8200, 8300, 8400, and 8500 are shown with dashed lines indicating that one or more metrics can be excluded.

  FIG. 9 illustrates a method for determining a dockable position according to the present disclosure. The determination of the dockable position is performed by specifying the position at 9100, specifying the traffic network information corresponding to the position at 9200, determining whether the position at 9300 is adjacent to the pedestrian walking area, This includes determining whether the position is a stop or no-riding area, determining whether the position is close to an intersection at 9500, including or excluding dockable positions at 9600, or a combination thereof. In some embodiments, the candidate docking operation is a docking operation that is performed at the target docking position.

  As shown in FIGS. 7-8, the corresponding position of the candidate docking operation, such as the identified candidate docking operation, is identified at 9100. For example, the vehicle operation information includes position information corresponding to the candidate docking operation, such as a GPS position in the vehicle traffic network.

  The traffic network information corresponding to the location specified at 9100 is used at 9200. In some embodiments, identifying traffic network information corresponding to the location identified at 9100 includes identifying traffic network information for a portion of the vehicle traffic network proximate to the location. For example, the identified traffic network information includes attributes such as roads, road segments, lanes, intersections, or other traffic control regulations corresponding to speed limits and identified locations.

  The traffic network information identified at 9100 is used at 9300 to determine whether the location is adjacent to a pedestrian walkable area. For example, the traffic network information of the candidate docking operation position shown in the lane 4360 of FIG. 4 indicates that the position is adjacent to the lanes 4320 and 4340 and not adjacent to the pedestrian walking area. , Including filtering of driving information to exclude candidate docking operations corresponding to the position. As another example, the traffic network information of the candidate docking operation position shown in the lane 4320 of FIG. 4 indicates that the position is adjacent to the pedestrian walkable area 4600, and the traffic network information is specified in the lane 4320. It includes filtering of driving information including a candidate docking operation corresponding to the indicated position.

  The traffic network information identified at 9100 is used at 9400 to determine whether the location is a stop or no boarding area. For example, the traffic network information indicates that the position related to the candidate docking operation is within a portion where the vehicle traffic network is specified as a stop, stop, or no-boarding area, and the use of the vehicle state information indicates stop, stop, Alternatively, it includes filtering of driving information to exclude candidate docking operations associated with positions identified as no-boarding zones.

  The traffic network information identified at 9100 is used at 9500 to determine whether the location is near an intersection. For example, the traffic network information displays a spatial difference such as the distance between the position associated with the candidate docking operation and the intersection exceeding the set intersection docking buffer distance, and the evaluation of the vehicle state information is based on the intersection from the intersection. Including driving information filtering to exclude candidate docking operations associated with locations within the docking buffer distance.

  The location associated with the candidate docking operation is identified at 9600 as to whether docking is possible. For example, the position corresponding to the candidate docking operation is specified based on one or more metrics at 9300, 9400, or 9500 to determine whether docking is possible. In some embodiments, one or more metrics are excluded. For example, the metrics 9300, 9400, and 9500 are shown with dashed lines indicating that one or more metrics are excluded. In some embodiments, one or more additional metrics are added. For example, a position in a parking area such as the parking area 3200 is excluded from the candidate docking positions.

  10-12 illustrate a portion of a vehicle traffic network 10000 according to the present disclosure. A part of the vehicle traffic network 10000 shown in FIGS. 10 to 12 corresponds to a part of the vehicle traffic network shown in FIG.

  FIG. 10 illustrates a portion of a vehicle traffic network 10000 that includes candidate docking locations according to the present disclosure. In one embodiment, as shown in FIG. 10, the traffic network information indicating a part of the vehicle traffic network 10000 is a leftward traveling direction, an upper road 10100 corresponding to a westbound road, an upward traveling direction, The right road 10110 corresponding to the southbound road, the leftward road 10120 corresponding to the northbound road in the downward driving direction, the central road 10130 corresponding to the eastbound road in the rightward driving direction, and the rightward direction Information indicating a walkable area, such as a lower road 10140 corresponding to an eastbound road. The traffic network information includes information indicating a non-navigable area such as the building 10200 in the upper right corner and the building 10210 in the upper left corner. The code | symbol X of FIG. 10 shows the candidate docking position identified automatically based on the driving information of multiple vehicles.

  In one embodiment, the traffic network information includes information indicating a set entrance location, such as 10202 of the building 10200 in the upper right corner. In one embodiment, the traffic network information includes information indicating the location of the set destination, such as 10204 of the building 10200 in the upper right corner, shown as a black square in FIG. In one embodiment, the traffic network information includes information indicating the edge location of the destination. In one embodiment, the traffic network information includes information indicating a predicted entrance location, such as 10212 of the building 10210 in the upper left corner.

  In some embodiments, the automatic generation of docking location information includes associating one or more candidate docking locations with points of interest based on traffic network information and driving information. For example, the candidate docking position in the upper left corner of FIG. 10 is associated with the building 10210 in the upper left corner, and the candidate docking position in the upper right corner is associated with the building 10200 in the upper right corner. The candidate docking position shown in the parking area 5300 is associated with the parking area 5300 and is associated with the building based on proximity, or the candidate docking position is the candidate docking position shown in the parking area 5300 of FIGS. Filtered to exclude.

  FIG. 11 illustrates a portion of a vehicle traffic network 10000 that includes docking location clusters in accordance with the present disclosure. In some embodiments, the automatic generation of docking location information includes one or more docking location clusters 11000/11002/11004/11006/11008/11010 of points of interest based on candidate docking locations associated with the points of interest. Each docking position cluster 11000/11002/11004/11006/11008/11010 includes a plurality of candidate docking positions. For example, the upper right docking position cluster 11002 includes four candidate docking positions.

  In one embodiment, docking position clusters are generated using the k-means method. Generating the docking location cluster includes excluding one or more candidate docking locations associated with the point of interest from the docking location cluster. For example, candidate docking positions along the right side are associated with the building 10200, and generation of the docking position cluster 11000 includes identifying the candidate docking position 11100 as an abnormal value and excluding the candidate docking position 11100 from the docking position cluster 11000. including.

  In some embodiments, multiple docking location clusters are associated with the point of interest. For example, docking location cluster 11000/11002/11004/11006 is associated with building 10200 and docking location cluster 11008/11010 is associated with building 10210.

  FIG. 12 illustrates a portion of a vehicle traffic network 10000 that includes a docking location based on a docking location cluster according to the present disclosure. In one embodiment, the docking location of the point of interest is identified based on a docking location cluster associated with the point of interest. For example, the docking position identified based on the docking position cluster corresponds to the docking position cluster arithmetic average of the docking position cluster. FIG. 12 shows the docking position cluster arithmetic average of each docking position cluster shown in FIG. However, the determination of the arithmetic average of docking position clusters of one or more docking position clusters is excluded. For example, the docking position cluster arithmetic average is specified for the maximum docking position cluster at each point of interest. For example, docking location cluster 11004 includes the maximum cardinality of candidate docking locations among the docking location clusters associated with building 10200, and docking location cluster arithmetic average 12000 is identified for docking location cluster 11004 and other associated with building 10200. Acquisition of the docking position cluster arithmetic average of the docking position cluster is excluded.

  FIG. 13 illustrates a method for identifying a passenger docking position according to the present disclosure. The identification of the passenger docking position is performed by specifying the candidate docking position at 13100, specifying the candidate destination at 13200, associating the candidate docking position with the candidate destination at 13300, specifying the selected destination at 13400, and at 13500. Including the identification of the docking position or a combination thereof. In one embodiment, the candidate docking position corresponds to a docking operation at the target docking position, and the selected destination is the first destination.

  Based on the driving information, a candidate docking position is identified at 13100. For example, candidate docking positions as shown in FIG. 10 are specified as shown in FIGS.

  One or more candidate destinations are identified at 13200 based on the traffic network information. Each point of interest in the vehicle traffic network corresponds to a candidate destination. For example, each building shown in FIG. 10 is specified as a candidate destination.

  Each candidate docking location is associated with a candidate destination at 13300 based on traffic network information and driving information. In one embodiment, the candidate docking position is associated with a candidate destination based on proximity to the set destination shown in the driving information of FIG. In one implementation, the candidate docking location is associated with a candidate destination based on proximity shown in FIG.

  In some embodiments, one or more docking locations are identified for the selected destination based on candidate docking locations associated with the selected destination. For example, the candidate destination is identified as the destination selected at 13400, and the docking position of the selected destination is identified at 13500 based on the candidate docking position associated with the destination selected at 13300. In one implementation, identifying the docking location at 13500 includes the use of docking location clustering as shown in FIG.

  FIG. 14 illustrates a method for associating a candidate docking position with a set destination according to the present disclosure. The implementation of the association between the candidate docking position and the set destination includes identification of the candidate docking position at 14100, determination of the association between the candidate docking position and the destination set in the driving information at 14200, and 14300. Including determination as to whether the set destination in the vicinity is close to the candidate docking position, determination regarding association between the candidate docking position and the set destination in 14400, or a combination thereof. In one embodiment, the candidate docking position corresponds to a docking operation at the target docking position and the candidate destination is the first destination.

  The candidate docking position is identified at 14100. For example, the candidate docking position is specified based on the driving information as shown in FIGS.

  Whether the driving information corresponding to the candidate docking position indicates the set destination is determined in 14200. In one embodiment, the driving information corresponding to the candidate docking location includes a set destination that identifies a point of interest that is clearly identified as the destination. For example, the passenger of the vehicle indicated in the driving information inputs the point of interest as a destination to the vehicle navigation system, and the driving information indicates that the vehicle travels to the set destination within the set time. The candidate docking position identified at 14100 corresponds to the set time.

  The determination as to whether the driving information corresponding to the candidate docking position indicates the set destination in 14200 means that the driving information corresponding to the candidate docking position specified in 14100 indicates the set destination. Whether the candidate docking position is close to the set destination is determined at 14300. For example, the distance between the candidate docking position and the set destination is within the set threshold, and this determination indicates that the candidate docking position is close to the set destination. Although not shown in FIG. 14, in one embodiment, determining the distance between the candidate docking position and the set destination includes identifying a destination position that can be used for the set destination shown in FIG. 15.

  The determination as to whether the candidate docking position is close to the set destination at 14300 indicates that the candidate docking position is close to the set destination, and at 14400, the candidate docking position is associated with the set destination. In one embodiment, the determination of whether the candidate docking position at 14300 is close to the set destination indicates that the candidate docking position is not close to the set destination, and the candidate docking position is as shown in FIG. Associated with different destinations.

  FIG. 15 illustrates a method for associating a candidate docking position with a candidate destination according to the present disclosure. The association between the candidate docking position and the candidate destination is performed by specifying the candidate docking position at 15100, determining the association between the candidate docking position and the destination specified by the driving information in 15200, and approaching at 15300. This includes identifying candidate destinations, associating candidate docking locations with proximity candidate destinations at 15400, or combinations thereof. In one embodiment, the candidate docking position corresponds to a docking operation at the target docking position and the candidate destination is the first destination.

  The candidate docking position is identified at 15100. For example, the candidate docking position is specified based on the driving information shown in FIGS. 6-9.

  Whether the driving information corresponding to the candidate docking position indicates the set destination is determined in 15200. The determination at 15200 is the same as the determination at 14200 shown in FIG.

  In 15200, the determination as to whether the driving information corresponding to the candidate docking position indicates the set destination indicates that the set destination is excluded from the driving information corresponding to the candidate docking position specified in 15100. The proximity candidate destination is identified at 15300.

  In some embodiments, identifying the proximity candidate destination at 15300 includes identifying one or more candidate destinations to identify a candidate destination that is spatially closest to the candidate docking location. Each of the points of interest included in a part of the vehicle traffic network corresponds to a candidate destination. For example, each building shown in FIG. 10 is specified as a candidate destination.

  In one embodiment, the candidate docking location identified at 15100 is close to the candidate destination, such as within a set associated proximity threshold. If the distance between the candidate docking position and another candidate destination exceeds the setting-related proximity, the candidate docking position is associated with the proximity candidate destination.

  In one embodiment, the candidate docking position is associated with the plurality of candidate destinations, and the candidate docking position is associated with the candidate docking position such that the distance between the candidate docking position and the candidate destination is minimized. Is done. For example, the distance between the candidate docking position and the candidate destination is minimized by determining the distance between the candidate docking position and each candidate destination based on the traffic network information, and associating the candidate destination corresponding to the minimum distance with the candidate docking position. To be executed.

  In one embodiment, the determination of proximity candidate destinations at 15300 includes an evaluation of traffic network information to identify the location of the destination for each candidate destination at 15310. In one embodiment, the evaluation for identifying the location of the destination as a candidate destination at 15310 includes identifying locations that can be used as candidate destinations.

  In some embodiments, identifying the location of a destination that can be used as a candidate destination includes determining at 1531 whether the first location can be used as a candidate destination. For example, the traffic network information indicates a first position of a candidate destination, such as a set entrance position, and the set entrance position is used as a usable destination position.

  In some embodiments, the first location of the candidate destination cannot be used, and identifying the location of the destination that can be used as the candidate destination includes a determination as to whether the predicted location at 15314 is available as the candidate destination. . For example, the predicted position is specified based on the traffic network information. In one embodiment, the predicted position is specified based on the traffic network information shown in FIG.

  In one embodiment, the first location and the predicted location of the candidate destination are not usable, and the identification of the usable destination location for the candidate destination is a candidate at 15316 based on traffic network information. Includes a determination as to whether the destination edge location is available. In one embodiment, the traffic network information explicitly specifies edge information. For example, the traffic network information indicates the edge position of a building. In one embodiment, the traffic network information excludes explicitly specified edge positions and a predicted edge position is selected. For example, the candidate destination is a building, and the traffic network information identifies the position of the building, indicates a road adjacent to the building, identifies the position of at least a part of the road, and the edge position of the building is the position of the building And based on the position of the road.

  In one embodiment, the first location, predicted location, and edge location cannot be used for a candidate destination, and the location of a destination that can be used as a candidate destination is identified in 15318 based on traffic network information. As including the use of points of interest. For example, the traffic network information indicates the position of a candidate destination such as an address, a postal address, a vehicle traffic network address, or a GPS address.

  The proximity candidate destination is identified at 15300 and the candidate docking location is associated with the proximity candidate destination at 15400.

  FIG. 16 illustrates a method for associating a docking location with a destination by using a docking location cluster in accordance with the present disclosure. Associating candidate docking positions and destinations using docking position clusters includes identifying a set of candidate docking positions at 16100, identifying one or more docking position clusters at 16200, and docking position clusters at 16300. Selection of the docking position cluster at 16400, the association of the docking position with the destination at 16500, or a combination thereof. In one embodiment, the docking position corresponds to a docking operation at the target docking position and the destination is the first destination.

  A set of candidate docking locations is identified at 16100. In some embodiments, the candidate docking location cluster includes candidate docking locations associated with the destination. For example, FIG. 10 illustrates candidate docking positions associated with destinations, and candidate docking positions are associated with destinations as shown in FIGS.

  In one embodiment, one or more docking location clusters are identified at 16200. The docking position cluster includes a plurality of candidate docking positions of a set of candidate docking positions. For example, FIG. 11 shows an example of a docking position cluster. In one embodiment, docking position clusters are generated using the k-means method. Generating the docking location cluster includes excluding one or more candidate docking locations associated with the point of interest from the docking location cluster. For example, the candidate docking position shown on the right side of FIG. 11 is associated with the building 10200 shown in FIG. The generation of the docking position cluster 11000 shown in FIG. 11 is performed by specifying the candidate docking position 11100 shown in FIG. 11 as an abnormal value and excluding the candidate docking position 11100 shown in FIG. 11 from the docking position cluster 11000 shown in FIG. Including. In some embodiments, multiple docking location clusters are associated with a destination. For example, the docking position cluster 11000/11002/11004/11006 shown in FIG. 11 is associated with the building 10200 shown in FIG. 11, and the docking position cluster 11008/11010 shown in FIG. 11 is associated with the building 10210 shown in FIG.

  In one embodiment, the docking location cluster is selected at 16300. For example, the destination is associated with a docking position cluster that includes a plurality of docking position clusters and a maximum cardinality of a candidate complementary docking position that is a maximum size docking position cluster. In one embodiment, the selection of docking location clusters includes a size determination indicating a count or cardinality of candidate docking locations for each docking location cluster associated with the destination. For example, the docking position cluster 11000 illustrated in FIG. 11 includes eight candidate docking positions, and the docking position cluster 11004 illustrated in FIG. 11 includes ten candidate docking positions. In one embodiment, selecting a destination docking location cluster includes selecting a maximum cardinality docking location cluster from a docking location cluster associated with the destination. In some embodiments, docking position clusters are ordered or ranked by size.

  In one embodiment, the arithmetic average, median, or average value of the docking location cluster is identified at 16400. For example, the docking position cluster 11004 shown in FIG. 11 is identified as the selected docking position cluster in 16300, and the arithmetic average, median value, or average value of the candidate docking positions included in the docking position cluster is calculated by 12000 of FIG. As shown, it is specified as the docking position cluster arithmetic average. In one embodiment, an arithmetic average of docking location clusters is identified for each docking location cluster associated with the destination.

  The docking location is associated with the destination at 16500. For example, the docking position cluster 11004 shown in FIG. 11 is identified as the selected docking position cluster at 16300, and the docking position cluster arithmetic average 12000 shown in FIG. 12 is identified as the docking position cluster arithmetic average at 16400 and at 16500. The docking position corresponding to the docking position cluster arithmetic average is associated with the destination. In one embodiment, a docking position cluster arithmetic average is identified for each docking position cluster associated with the destination, and a docking position corresponding to each docking position cluster arithmetic average is associated with the destination. In some embodiments, docking positions are ordered or ranked based on the cardinality of the corresponding docking position cluster.

  17-19 illustrate another portion 17000 of a vehicle traffic network according to the present disclosure. A portion 17000 of the vehicle traffic network shown in FIGS. 17 to 19 corresponds to a portion similar to a part of the vehicle traffic network shown in FIG.

  FIG. 17 illustrates a portion 17000 of a vehicle traffic network that includes candidate docking locations in accordance with the present disclosure. As shown, the vehicle traffic network portion 17000 includes a building 17100 at the top, a building 17200 at the bottom right, a building 17300 at the bottom left, a parking area 17400, a road 17500 on the left in the upward direction, and a right side in the downward direction. Road 17510 includes a center road 17520 having a lane 17522 in the left-hand travel direction and a lane 17524 in the right-hand travel direction. FIG. 17 includes a symbol X indicating a candidate docking position. The lower right building 17200 includes a set destination location 17210 indicated by a black square, and the lower left building 17300 includes a set entrance 17310 indicated by a black diamond. Portion 17000 of the vehicle traffic network includes pedestrian pedestrian areas shown in diagonal lines, such as pedestrian road 17600 adjacent to upper buildings 17100-17110 and pedestrian crossing 17610 intersecting central road 17520.

  FIG. 18 illustrates a portion 17000 of a vehicle traffic network that includes a road segment in accordance with the present disclosure. In some embodiments, associating a candidate docking location with a destination includes identifying one or more road segments. In one embodiment, the traffic network information includes road segment information indicating a plurality of road segments of the vehicle traffic network. For example, the right road 17510 includes an upper first road segment (segment 1) and a lower second road segment (segment 2), and the central road 17520 includes a third road segment (segment 3). The first road segment (Segment 1) includes a first docking position cluster 18000, which includes a first docking position cluster median 18010. The second road segment (segment 2) includes a second docking position cluster 18100 that includes a second docking position cluster median 18110. The third road segment (Segment 3) includes a third docking position cluster 18200 including a third docking position cluster median 18210, a fourth docking position cluster including a fourth docking position cluster median, and a fifth docking position cluster center. Contains a fifth docking position cluster containing values. For the sake of brevity, FIG. 18 does not show the candidate docking position, segment, docking position cluster, and docking position cluster median for the left road.

  In one embodiment, as shown in FIG. 21, a predicted entrance position is generated as a destination. FIG. 18 shows a line 18300 from the set destination position 17210 perpendicular to the third segment (segment 3) and a line 18310 from the set destination position 17210 perpendicular to the second segment (segment 2); From the predicted entry position 18400 corresponding to the intersection of the line 18300 from the set destination position 17210 perpendicular to the third segment (segment 3) and the set destination position 17210 perpendicular to the second segment (segment 2) The predicted entrance position 18410 corresponding to the intersection of the line 18310 is shown.

  FIG. 19 shows a portion 17000 of a vehicle traffic network that includes a vehicle traffic network region according to the present disclosure. In FIG. 19, for the sake of brevity, the vehicle traffic network area includes an upper right vehicle traffic network area 19000, a lower right vehicle traffic network area 19100, a center right vehicle traffic network area 19200, and an upper left vehicle traffic network area 19300. Shown in bold lines. A candidate docking position 19500 is shown on the left side of the upper right vehicle traffic network area 19000.

  The vehicle traffic network area includes a walkable area such as a road, a partially walkable area such as a parking area 17400, a non-navigable area such as a building 17100-17110 / 17200, or other part of the vehicle traffic network. The vehicle traffic network area has any regular or irregular shape. For the sake of brevity, FIG. 19 omits the illustration of one or more vehicle traffic network regions of the vehicle traffic network portion 17000.

  In one embodiment, the vehicle traffic network region includes a destination or a portion of the destination, and the median docking location cluster of the vehicle traffic network region is associated with the destination as the docking location of the destination. In some embodiments, destinations such as the lower right building 17200 or a portion thereof may be included in multiple vehicle traffic network areas, such as the lower right vehicle traffic network area 19100, the center right vehicle traffic network area 19200, and the like. It is. In some embodiments, a plurality of destinations, such as upper buildings 17102-17110 or portions thereof, are included in a vehicle traffic network area, such as central vehicle traffic network area 19400.

  FIG. 20 is a diagram illustrating a method of associating a candidate docking position with a candidate destination by network division according to the present disclosure. The association between the candidate docking position and the candidate destination by network division is as follows: the association of the candidate docking position with the road segment in 20100, generation of the docking position cluster of the segment in 20200, and Identification, partitioning of the vehicle traffic network at 20400, associating docking location cluster medians with destinations based on the partitioning at 20500, ordering of docking locations 20600 or combinations thereof. In one embodiment, the candidate docking position corresponds to a docking operation at a docking position cluster corresponding to the target docking position, and the destination associated with the docking position cluster median is the first destination.

  In one embodiment, candidate docking locations are associated with road segments at 20100. For example, the traffic network information includes road segment information for identifying a road segment as shown in FIG. 18 and candidate docking positions as shown in FIG. 17, and is associated with the corresponding road segment based on the traffic network information and driving information. It is done.

  In one embodiment, at 20200, the docking location cluster is determined as the road segment identified at 20100. For example, docking position clusters as shown in FIGS. 18 to 19 are determined for each segment. The determination of the docking position cluster of the road segment is similar to the identification of the docking position cluster shown at 16200 in FIG. 16 except that the docking position cluster generated for the segment excludes the candidate docking position from the segment.

  In one embodiment, the docking position cluster median is calculated at 20300. For example, the first docking position cluster 18000 shown in FIGS. 18 to 19 is identified by 20200 for the first segment (segment 1), and the arithmetic average, median, and average of the candidate docking positions included in the first docking position cluster 18000 are The second docking position cluster 18100 calculated as the first docking position cluster median 18010 and shown in FIGS. 18-19 is identified by 20200 for the second segment (segment 2) and is included in the second docking position cluster 18100. The arithmetic average, median, and average of the docking positions are calculated as the second docking position cluster median 18110, and the third docking position cluster 18200 shown in FIGS. 18-19 is identified at 20200 for the third segment (segment 3). The second The arithmetic mean of the candidate docking position included in the docking position cluster 18200, median or average value is computed as a third docking position cluster median 18210.

  In one embodiment, the vehicle traffic network is divided at 20400 to determine a vehicle traffic network region, such as the vehicle traffic network region 19000, 19100, or 19200 shown in FIG. In one embodiment, the vehicle traffic network is split at 20300 based on the docking location cluster median. For example, the vehicle traffic network area 19000 is divided based on the first docking position cluster median value 18010, the vehicle traffic network area 19100 is divided based on the second docking position cluster median value 18110, and the vehicle traffic network area 19200 is It is divided based on the docking position cluster median 18210. In one embodiment, each vehicle traffic network region includes one docking location cluster median. Each vehicle traffic network region includes a plurality of locations from the vehicle traffic network. The position indicates any identifiable position within the vehicle traffic network.

  In one embodiment, the partitioning of the vehicle traffic network includes identifying a vehicle traffic network region based on proximity to the corresponding docking location cluster median. For example, traffic network information indicating a vehicle traffic network area is represented as a Voronoi diagram. The docking location cluster median is the seed for each vehicle traffic network region. In one embodiment, the distance between the position of one vehicle traffic network region and the median docking position cluster of the vehicle traffic network region is within the distance between the location and the median docking position cluster. For example, the distance between an arbitrary position in the lower right vehicle traffic network area 19100 shown in FIG. 19 and the second docking position cluster median value 18110 in the vehicle traffic network area 19100 is the center of each position and any other arbitrary docking position cluster. Within the distance from the value.

  In one embodiment, the distance between the position and the median docking position cluster is determined based on an estimate of pedestrian travel time. In one embodiment, the estimation of pedestrian travel time is determined based on pedestrian traffic network information. For example, pedestrian traffic network information shows the pedestrian walkable area shown in FIGS.

  In the first example, the pedestrian walks from the position of the pedestrian walkable area adjacent to the candidate docking position 19500 on the left side of the upper right vehicle traffic network area 19000 via the pedestrian walkable area. It moves to the position of the pedestrian walking area adjacent to the first docking position cluster median value 18010 of the traffic network area 19000. For example, a pedestrian moves via a sidewalk adjacent to buildings 17100 to 17110.

  In the second example, the pedestrian walks from the position of the pedestrian pedestrian area adjacent to the candidate docking position 19500 on the left side of the vehicle traffic network area 19000 on the upper right side via the pedestrian pedestrian area. It moves to the position of the pedestrian walkable area adjacent to the third docking position cluster median 18210 of the traffic network area 19200. For example, the pedestrian moves via a sidewalk adjacent to the upper buildings 17100 to 17110, a right pedestrian crossing, and a sidewalk adjacent to the lower right building 17200.

  The geographical distance between the candidate docking position 19500 on the left side of the upper right vehicle traffic network area 19000 and the position of the pedestrian walkable area adjacent to the median third docking position cluster 18210 of the center right vehicle traffic network area 1920 is The candidate docking position 19500 on the left side of the upper right vehicle traffic network area 19000 is within the distance between the position of the pedestrian walkable area adjacent to the first docking position cluster median value 18010 of the upper right vehicle traffic network area 19000.

  Estimation of the pedestrian travel time between the candidate docking position 19500 on the left side of the upper right vehicle traffic network area 19000 and the pedestrian walkable area adjacent to the first docking position cluster median value 18010 of the upper right vehicle traffic network area 19000 The value is a pedestrian between the candidate docking position 19500 on the left side of the upper right vehicle traffic network area 19000 and the pedestrian walkable area adjacent to the third docking position cluster median value 18210 of the center right vehicle traffic network area 1920. Within estimated travel time.

  In one embodiment, at 20500, a docking location is associated with the destination. In one embodiment, the docking location corresponding to the median docking location cluster value of the vehicle traffic network region that includes the destination is associated with the destination. For example, the vehicle traffic network area 19300 in the upper left includes the building 17100 shown in the upper left of FIG. 19, and the docking position cluster median of the vehicle traffic network area 19300 is calculated as the docking position of the building 17100 shown in the upper left of FIG.

  In one embodiment, the docking location is associated with the destination based on available location information. For example, the entrance position of the building is determined based on the usable position information, and the docking position is associated with the destination based on the determined entrance position. In one embodiment, the available location information includes the generation of a predicted entry location, as shown in FIG. 21, and is identified as in FIG.

  In some embodiments, the plurality of vehicle traffic network regions includes a portion of the destination, and the one or more docking locations are associated with the destination based on the entrance location information.

  For example, as in FIG. 19, the lower right vehicle traffic network region 19100 includes a portion of the lower right building 17200, which includes a predicted entrance location 18410 at the right edge of the building. The docking position corresponding to the second docking position cluster median 18110 of the vehicle traffic network area 19100 is associated with the predicted entrance position 18410. A central right vehicle traffic network area 19200 that includes a portion of the building 17200 includes a predicted entrance location 18400 at the upper edge of the building. The docking position corresponding to the third docking position cluster median 18210 of the vehicle traffic network area 19200 is associated with the predicted entrance position 18400.

  In some embodiments, docking locations associated with destinations are ordered or ranked at 20600. In one embodiment, the docking location associated with the destination is ordered based on an estimate of the pedestrian travel time. For example, as shown in FIG. 19, the estimated pedestrian travel time between the docking position corresponding to the second docking position cluster median 18110 and the associated predicted entrance position 18410 is the third docking position cluster median 18210. Greater than the estimated pedestrian travel time between the corresponding docking location and the associated predicted entrance location 18400. The docking position corresponding to the third docking position cluster median 18210 has a higher rank or order than the docking position corresponding to the second docking position cluster median 18110.

  FIG. 21 shows a method for specifying an entrance position according to the present invention. The entry position is specified by specifying the destination at 21100, determining whether the set entrance position at 21200 can be used as the destination, generating the predicted entrance position at 21300, ordering the predicted entrance position at 21400, 21500 Including the identification of the entrance position as the destination location or a combination thereof. In one embodiment, the median docking position cluster corresponds to the target docking position and the destination is the first destination.

  In one embodiment, the destination is identified at 21100. The specification of the destination includes the specification of the position of the set destination indicated in the traffic network information of the destination such as an address, a postal address, a vehicle traffic network address or a GPS address. For example, the setting destination position 17210 shown in FIGS. 17 to 19 is specified as the setting destination position of the building 17200 shown in FIGS.

  In one embodiment, it is determined at 21200 whether the set entrance location can be used as a destination. For example, the traffic network information includes destination entry position information.

  In some embodiments, the traffic network information excludes destination entry location information, and one or more predicted entrance locations are generated for the destination at 21300 based on the traffic network information. In some embodiments, generating a predicted entrance location for a destination based on traffic network information includes identifying a road segment at 21310, drawing a line at 21320, identifying an intersection at 21330, or a combination thereof.

  In one embodiment, one or more road segments proximate to the set destination location identified at 21100 are identified at 21310. FIG. 18 illustrates a portion of a vehicle traffic network that includes road segments. In one embodiment, the road segment is indicated in the traffic network information. For example, the linear distance between the road segment and the position of the set destination is within the set threshold, and the road segment is close to the destination. For example, the second segment (segment 2) and the third segment (segment 3) shown in FIG. 18 are close to the set destination position 17210 shown in FIG.

  In one embodiment, the line is drawn at 21320. For example, as each road segment specified in 21310, a line perpendicular to each road segment is drawn from the position of the set destination specified in 21100. 18 to 19 show the drawn lines as bold lines. 18-19 intersect the third segment (Segment 3) and intersect the second segment (Segment 2) with a line 18300 drawn perpendicularly from the set destination location 17210 to the third segment (Segment 3) And a line 18310 drawn perpendicularly from the set destination position 17210 to the second segment (segment 2).

  Each intersection between the line drawn in 21320 and the road segment specified in 21310 is specified in 21330 as the predicted entrance position of the destination. For example, in FIGS. 18-19, the intersection of the drawn line 18300 and the third segment (segment 3) is identified as the predicted entrance location 18400 of the destination, and the drawn line 18310 and the second segment (segment 2) intersect. Is identified as the predicted entrance location 18410 of the destination. In one embodiment, the predicted entrance location corresponds to the edge of the destination along a line that intersects the road segment. For example, predicted entrance positions 18400/18410 in FIGS. 18-19 correspond to each edge of building 17200 along line 18300/18310 perpendicular to each segment.

  The entrance position includes a set entrance position or a predicted entrance position, and is associated with the docking position of the building 21400 shown in FIG.

  FIG. 22 illustrates another portion of a vehicle traffic network that includes candidate docking locations according to the present disclosure. In some embodiments, the pedestrian traffic network information includes one or more portions of the pedestrian traffic network that occur concurrently with one or more walkable, partially walkable, or non-walkable areas of the vehicle traffic network. Identify. For example, FIG. 22 includes parking areas 22000/2002, roads 22100/22102/22104, and buildings 22200, each of which includes a pedestrian walkable area. The entrance position of the set entrance position or the predicted entrance position is indicated by a white diamond. One or more docking positions of the building 22200 are identified as described above. FIG. 22 includes a first selected area indicator 22300 and a second selected area indicator 22310. The first selected area indicator 22300 and the second selected area indicator 22310 indicate passenger input selected areas in the vehicle traffic network indicated by dashed circles.

  FIG. 23 relates to the present disclosure, and shows a navigation and routing method for an autonomous vehicle based on a plurality of docking positions. Navigation and routing of an autonomous traveling vehicle by using a plurality of docking positions is performed by an autonomous traveling vehicle such as the autonomous traveling vehicle 1000 shown in FIG. 1 or the autonomous traveling vehicle 2100 shown in FIG. For example, the processor 1330 of the controller 1300 of the autonomous traveling vehicle 1000 shown in FIG. 1 executes instructions stored in the memory 1340 of the controller 1300 of the autonomous traveling vehicle 1000 shown in FIG. Navigation and routing of the traveling vehicle is performed. In some embodiments, autonomous vehicle navigation and routing using multiple docking locations is similar to the autonomous vehicle navigation and routing shown in FIG. 6, with additions or variations of the features described herein.

  In one embodiment, autonomous vehicle navigation and routing using multiple docking locations includes identifying a first destination at 23100, determining a first target docking location at 23110, and first target docking at 23120. Identifying the route to the location, identifying the second destination at 23130, identifying the route to the second destination at 23140, identifying the second target docking location at 23150, and identifying the second destination at 23160. 2 Specifying the route to the target docking position, traveling to the first target docking position at 23200, executing the docking operation at 23210, traveling to the second destination at 23220, and parking at 23230 , Running to the second target docking position at 23240, performing the second docking operation at 23250, or a combination thereof. .

  In some embodiments, identifying the passenger docking location includes identifying a series of target docking locations. For example, a building 222000 shown in FIG. 22 is a shopping center or a movie theater, and a passenger of an autonomous vehicle selects a shopping center as a destination, and sets an entrance position of the shopping center such as an entrance position close to the right side to an exit position. As the first target docking position that can be used as the second target docking position that can be used as the boarding position. In an embodiment, the autonomous vehicle specifies a pedestrian movement route between the first target docking position and the second docking position.

  In one embodiment, the first destination is identified at 23100. The identification of the first destination includes the identification of the traffic network information shown in FIG. The identification of the first destination includes the identification of the position of the set destination indicated in the traffic network information of the destination such as an address, a mail address, a vehicle traffic network address or a GPS address. For example, the first destination is specified based on the input such as selection of the first destination by user input.

  In one embodiment, a first target docking position is determined at 23110 based on the first destination identified at 23100. For example, the first target docking position that can be used as the getting-off position is specified as shown in FIG. In some embodiments, the first destination is associated with a plurality of entrance locations and / or a plurality of docking locations, and the identification of the first target docking location is determined by a respective docking location associated with each entrance location of the first destination. Includes the generation of one or more candidate routes for the location. In one embodiment, the generation of candidate routes is similar to the routing of FIG. In one embodiment, the generation of candidate routes includes the use of the combination of vehicle route and pedestrian route shown in FIG. In some implementations, the autonomous vehicle automatically identifies the entry position and / or the first target docking position. In some implementations, the entry position and / or the first target docking position are identified based on input, such as a selection by passenger input of the entry position and / or first target position.

  In one embodiment, the route to the first target docking location is identified at 23120. For example, a route is selected from the candidate routes generated at 23110. In one embodiment, identifying the route to the first target docking location is similar to identifying the route shown in FIG.

  In one embodiment, a second destination that is a parking area for the first destination is identified at 23130. For example, the first destination such as the building 22200 shown in FIG. 22 is associated with one or more parking areas such as the parking area 22000/2002 shown in FIG. 22 and specified as the second destination. In one embodiment, the second destination is identified based on the first target docking location identified at 23110. In one embodiment, the second destination is identified based on the second target docking location identified at 23150.

  In one embodiment, a route from the first target docking location to the second destination is generated at 23140. The generation of the route from the first target docking position to the second destination is similar to the routing shown in FIG.

  The second target docking position used for passenger boarding / exiting operations is identified at 23150. In some implementations, the autonomous vehicle automatically identifies the entrance position and / or the second target docking position. In some implementations, the entry position and / or the second target docking position are identified based on the input, for example, selecting the entry position and / or the second target docking position by passenger input.

  In one embodiment, a route from the second destination to the second target docking location is generated at 23160. The generation of the route from the second destination to the second target docking position is similar to the routing shown in FIG.

  Although not shown individually in FIG. 23, in one embodiment, one or more pedestrian routes from the first target docking position to the second target docking position are generated and presented to the passengers.

  In one embodiment, the autonomous vehicle travels from the departure point to the first target docking position at 23200 by the route identified at 23120. The travel from the departure point to the first target docking position is similar to the travel shown in FIG.

  In one embodiment, at 23210, the autonomous vehicle performs a docking operation. For example, the autonomous vehicle stops at the first target docking position, and one or more passengers get off the autonomous vehicle. At 23220, the autonomous vehicle travels from the first target docking position to the second destination, and at 23230, parks at the second destination. In one embodiment, the autonomous vehicle travels from the second destination to the second target docking position at 23240, and at 23250, the second docking operation is performed at the second target docking position. For example, the autonomous traveling vehicle stops at the second target docking position, and one or more passengers get on the autonomous traveling vehicle.

  FIG. 24 illustrates a method for identifying an expanded passenger docking location according to the present disclosure. The identification of the extended passenger docking position is performed by an autonomous traveling vehicle such as the autonomous traveling vehicle 1000 illustrated in FIG. 1 or the autonomous traveling vehicle 2100 illustrated in FIG. 2. For example, the processor 1330 of the controller 1300 of the autonomous vehicle 1000 shown in FIG. 1 executes the instruction stored in the memory 1340 of the controller 1300 of the autonomous vehicle 1000 shown in FIG. 1 and specifies the expanded passenger docking position. To do. In one embodiment, the identification of the extended passenger docking position is similar to the identification of the docking position shown in FIG. 6 by adding or modifying the features described herein.

  In one embodiment, the identification of the expanded passenger docking location is identified by identifying traffic network information at 24100, generating a vehicle decision model at 24200, identifying pedestrian traffic network information at 24300, and determining pedestrians at 24200. Including model generation, extension of the vehicle determination model based on the pedestrian determination model at 24500, determination of the target docking position at 24600, or a combination thereof.

  In one embodiment, traffic network information as shown in FIGS. 3-5, 10-12, 17-19, and 22 is identified by 24100. For example, an autonomous vehicle control unit such as the controller 1300 shown in FIG. 1 reads traffic network information from a data storage unit such as the memory 1340 shown in FIG. 1 via a communication system such as the communication network 2300 shown in FIG. Receives traffic network information from an external data source, such as communication device 2400 shown in FIG. In some embodiments, the traffic network information includes docking location information indicating one or more docking locations within the vehicle traffic network.

  In one embodiment, a vehicle decision model is generated at 24200. The generation of the vehicle determination model includes the identification of the first destination similar to the identification of the destination shown in FIG. 6 or FIG. The identification of the first destination includes identification of a point of interest as the first destination such as the building 3100 shown in FIG. 3, the building 4100 shown in FIG. 4, the building 5100 shown in FIG. 5, or the building 22200 shown in FIG. .

  In one embodiment, the generation of the vehicle decision model is similar to the route generation shown in FIG. 6 by adding or modifying features described herein. The autonomous traveling vehicle creates a vehicle determination model for setting a route between the departure point and each docking position associated with the first destination by using the traffic network information. The vehicle determination model includes one or more candidate routes between the departure point and each docking location associated with the first destination.

  In one embodiment, pedestrian traffic network information is identified at 24300. The identification of pedestrian traffic network information is similar to the vehicle traffic network identification at 24100. The pedestrian traffic network information indicates an area close to the first destination. In one embodiment, the area included in the pedestrian traffic network information is identified based on a set metric such as, for example, a maximum pedestrian travel time determined based on user input. In one embodiment, the pedestrian traffic network information includes first destination entrance location information and / or docking location information. In one embodiment, the identification of traffic network information at 24100 and the identification of pedestrian traffic network information at 24300 are combined.

  In one embodiment, the pedestrian decision model is generated at 24200. For example, the autonomous vehicle generates a second determination model for setting a pedestrian route between each docking position associated with the first destination and the corresponding entrance position using the pedestrian traffic network information. To do. The generation of the pedestrian determination model is similar to the generation of the vehicle determination model in 22200 except that the pedestrian determination model is generated based on the pedestrian traffic network information, and each of the pedestrian determination models is associated with the first destination. One or more candidate routes between the docking location and each entrance location associated with the first destination are based on a metric specified in pedestrian routing.

  In one embodiment, the autonomous vehicle generates one or more candidate pedestrian routes between each entrance position identified at the first destination and each routing state of the pedestrian determination model. The autonomous vehicle generates an estimated cost in each routing state of the pedestrian determination model based on the pedestrian travel time. In one embodiment, the generation of the pedestrian decision model is similar to the route optimization shown in FIG. 6 in that the optimum between each identified entry location at the first destination and each routing state of the pedestrian decision model. Including route determination. For example, pedestrian route optimization involves the application of a shortest path algorithm for all pairs.

  In one embodiment, at 24500, the vehicle determination model generated at 24200 is extended based on the pedestrian determination model generated at 24400. The extension of the vehicle determination model by the pedestrian determination model includes a combination of the routing state, operation, the estimated cost from the vehicle determination model by the routing state, the docking operation, and the estimated cost from the pedestrian determination model.

  In one embodiment, the target docking location is identified at 24600 based on the enhanced decision model generated at 24500. In some embodiments, identifying the target docking location includes generating one or more routes, such as an optimal route from the target docking location to the entry location of the first destination. In one embodiment, identifying the target docking location includes identifying a target docking location for a first destination in the vehicle traffic network based on the enhanced decision model generated at 24500.

  For example, a building such as 22200 shown in FIG. 22 is specified as the first destination. One of the predicted entrance positions on the upper right of the building 22200 in FIG. 22 is identified as the entrance position of the first destination, and docking of the docking position cluster identified based on the candidate docking position shown on the road 22100 in FIG. A target docking position corresponding to the position cluster median is identified based on the entry position.

  In some embodiments, the autonomous vehicle identifies an entrance position and / or a target docking position based on one or more metrics. For example, an autonomous vehicle is a target docking based on minimizing overall travel costs, including estimated vehicle travel costs such as estimated vehicle travel time and estimated pedestrian travel costs such as estimated pedestrian travel time. Identify the location.

  In some embodiments, the entry position and / or the target docking position is identified based on input, such as input by a passenger. For example, the passenger selects a set or predicted entrance. In another example, the passenger selects a target docking position. In yet another example, the passenger may identify an area that includes multiple entrances and / or multiple docking locations, such as the selected first selected area indicator 22300 and second selected area indicator 22310 shown in FIG. The autonomous vehicle specifies a target docking position in the selected area.

The above aspects, examples and implementations have been described so that the disclosure can be readily understood, and is not limited thereto. Further, the present disclosure can be applied with various modifications and equivalent arrangements within the scope of the claims, and all of the above modifications and equivalent structures are allowed in a broad interpretation.

Claims (85)

An autonomous traveling vehicle comprising a processor and a track controller,
The processor executes instructions stored on a non-transitory computer readable medium,
Traffic network information indicating a vehicle traffic network including the first destination,
The traffic network information includes docking position information indicating a plurality of docking positions,
The traffic network information is specified so that each docking position of the plurality of docking positions corresponds to each position of the vehicle traffic network,
Based on the traffic network information, a target docking position is determined from the plurality of docking positions for the first destination,
Using the traffic network information, specify a route from the starting point in the vehicle traffic network to the target docking position,
The trajectory controller
The autonomous traveling vehicle, wherein the autonomous traveling vehicle is operated so as to move from the departure point to the target docking position using the route. The autonomous traveling vehicle according to claim 1,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles. The autonomous traveling vehicle according to claim 2,
The autonomous traveling vehicle, wherein the docking position information is determined based on the driving information by filtering the driving information. The autonomous traveling vehicle according to claim 2,
The autonomously traveling vehicle according to claim 1, wherein the filtering of the driving information includes specifying one operation as a candidate docking operation from the plurality of operations when a stationary time associated with the operation exceeds a minimum docking time. The autonomous traveling vehicle according to claim 2,
The driving information filtering includes specifying one operation as a candidate docking operation from the plurality of operations when a maximum docking time exceeds a stationary time associated with the operation. . The autonomous traveling vehicle according to claim 1,
The processor executes instructions stored on a non-transitory computer readable medium,
The docking position information is based on driving information of a plurality of vehicles.
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
The docking position information is determined based on the driving information by filtering the driving information,
The filtering of the driving information specifies one operation from the plurality of operations as a candidate docking operation when a stationary time associated with the operation exceeds a minimum docking time and a maximum docking time exceeds the stationary time. The autonomous traveling vehicle characterized by specifying the said traffic network information. The autonomous traveling vehicle according to claim 2 or 6,
The driving information associated with each vehicle of the plurality of vehicles includes a plurality of event indicators indicating a series of events of the vehicle;
One of the plurality of operations includes a plurality of events in the series of events,
The plurality of events includes a stop event followed by a start event;
The autonomous traveling vehicle, wherein the stationary time associated with the operation indicates a time difference between the start event and the stop event. The autonomous traveling vehicle according to claim 2 or 6,
Filtering the driving information includes specifying one operation as a candidate docking operation from the plurality of operations when the driving information includes a vehicle type indicator associated with the operation,
The autonomous vehicle according to claim 1, wherein the vehicle type indicator indicates that the vehicle is a transport vehicle. The autonomous traveling vehicle according to claim 2 or 6,
When the driving information includes a vehicle operation type indicator associated with the operation, the filtering of the driving information includes specifying one operation as a candidate docking operation from the plurality of operations,
The autonomous vehicle according to claim 1, wherein the vehicle operation type indicator indicates that the vehicle is a low-occupancy transport vehicle. The autonomous traveling vehicle according to claim 2 or 6,
The filtering of the driving information is performed by docking one operation from the plurality of operations when the driving information of the vehicle associated with the operation indicates a change in vehicle occupancy during a stationary time associated with the operation. An autonomous traveling vehicle including specifying as an action. The autonomous traveling vehicle according to claim 10,
The autonomous traveling vehicle, wherein the change in the vehicle occupancy is detected based on a signal from a passenger detection sensor of the vehicle. The autonomous traveling vehicle according to claim 2 or 6,
The driving information filtering is performed by docking one operation from the plurality of operations when the driving information of the vehicle associated with the operation indicates a passenger-side door opening event during a stationary time associated with the operation. An autonomous traveling vehicle including specifying as an action. The autonomous traveling vehicle according to claim 2 or 6,
The driving information filtering
Identifying one action from the plurality of actions;
Identifying a location associated with the action;
Identify traffic network information corresponding to the location;
Based on the traffic network information, determine whether the location is dockable,
If the position is dockable, the autonomous traveling vehicle includes specifying the operation as a candidate docking operation. The autonomous traveling vehicle according to claim 13,
The determination of whether the position is dockable is
Determining that the location is dockable if the traffic network information indicates that the location is adjacent to a pedestrian road;
If the traffic network information indicates that stopping or boarding at the location is prohibited, the location is determined to be undocked,
The autonomous traveling vehicle, wherein the traffic network information includes determining that the position can be docked when a spatial difference between the position and the intersection is larger than an intersection docking buffer distance. The autonomous traveling vehicle according to claim 1,
The processor executes instructions stored on a non-transitory computer readable medium,
The docking position information is based on driving information of a plurality of vehicles.
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
The driving information associated with each vehicle of the plurality of vehicles includes a plurality of event indicators indicating a series of events of the vehicle;
One operation in the plurality of operations includes a plurality of events in the series of events,
The plurality of operations includes a stop event and a subsequent start event;
The quiescent time associated with the action indicates the time difference between the start event and the stop event;
The docking position information specifies the traffic network information to be determined based on the driving information by filtering the driving information,
The driving information filtering
Identifying one action from the plurality of actions;
Identifying a location associated with the action;
Identify traffic network information corresponding to the location;
Based on the traffic network information, determine whether the location is dockable,
Including
The determination as to whether the position is dockable is
If the traffic network information indicates that the location is adjacent to a pedestrian road, the location is determined to be dockable;
If the traffic network information indicates that stopping or boarding at the location is prohibited, the location is determined to be undocked,
If the traffic network information indicates that the spatial difference between the location and the intersection is greater than an intersection docking buffer distance, determine that the location is dockable;
When the position is dockable, and the stationary time is larger than the minimum docking time and the maximum docking time is larger than the stationary time, the operation vehicle is specified as a candidate docking operation. The autonomous traveling vehicle according to any one of claims 1, 6, and 15,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second destination of the vehicle traffic network in response to performing a docking operation at the target docking position;
Using the traffic network information to identify a subsequent route from the target docking location of the vehicle traffic network to the second destination;
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle using the route to travel the autonomous traveling vehicle from the target docking position to the second destination. The autonomous traveling vehicle according to claim 15,
The driving information filtering
When the driving information includes a vehicle type indicator of the vehicle associated with the operation, and the vehicle type indicator indicates that the vehicle is a transport vehicle, one operation is selected from the plurality of operations. Identified as a candidate docking action,
If the driving information includes a vehicle operation type indicator for the vehicle associated with the operation, and the vehicle operation type indicator indicates that the vehicle is a low occupancy transport vehicle, If one operation is identified as a candidate docking operation and the driving information of the vehicle associated with the operation indicates a change in the vehicle occupancy during the stationary time associated with the operation, the change in occupancy is: Detected based on a signal from a passenger detection sensor of the vehicle, specifying one operation as a candidate docking operation from the plurality of operations,
If the driving information of the vehicle associated with the operation indicates a passenger-side door opening event during a stationary time associated with the operation, one operation is identified as a candidate docking operation from the plurality of operations. Autonomous vehicle characterized by including. The autonomous traveling vehicle according to claim 1,
The processor executes instructions stored on a non-transitory computer readable medium,
The autonomous traveling vehicle, wherein the traffic network information is specified so that at least one of the plurality of docking positions is associated with the first destination. The autonomous traveling vehicle according to claim 18,
The processor executes instructions stored on a non-transitory computer readable medium,
The docking position information includes a set of docking positions of the plurality of docking positions;
Each docking position of a set of docking positions is associated with the first destination,
The target docking position of the set of docking positions is
Identifying a plurality of docking position clusters based on the set of docking positions, and identifying the target docking position based on the plurality of docking position clusters,
To be associated with the first destination,
Identifying the traffic network information;
An autonomous traveling vehicle that determines whether to use the target docking position based on the traffic network information. The autonomous traveling vehicle according to claim 18,
The processor executes instructions stored on a non-transitory computer readable medium,
The docking position information includes a set of docking positions of the plurality of docking positions;
Each docking position of a set of docking positions is associated with the first destination,
The target docking position of the set of docking positions is
Identifying a plurality of docking position clusters based on the set of docking positions, and identifying the target docking position based on the plurality of docking position clusters,
Associating with the first destination such that the distance between the first destination and the target docking position is minimized,
Identifying the traffic network information;
An autonomous traveling vehicle that determines whether to use the target docking position based on the traffic network information. The autonomous traveling vehicle according to any one of claims 18 to 20,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second destination of the vehicle traffic network in response to performing a docking operation at the target docking position;
Using the traffic network information to identify a subsequent route from the target docking location of the vehicle traffic network to the second destination;
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle using the route to travel the autonomous traveling vehicle from the target docking position to the second destination. The autonomous traveling vehicle according to any one of claims 18 to 20,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
Each of the docking positions of the plurality of docking positions corresponds to each of the plurality of actions. The autonomous traveling vehicle according to claim 22,
The processor executes instructions stored on the non-transitory computer readable medium,
If one of the plurality of operations corresponding to the docking position indicates the first destination, and the distance between the target docking position and the first destination is within a set threshold, the plurality of docking positions The autonomous traveling vehicle, wherein the traffic network information is specified so that the target docking position is associated with the first destination. The autonomous traveling vehicle according to claim 22,
The vehicle traffic network has a plurality of candidate destinations,
The first destination is one of the plurality of candidate destinations;
The processor executes instructions stored on the non-transitory computer readable medium,
Specifying the traffic network information so that the target docking positions of the plurality of docking positions are associated with the first destination when the distance between the first destination and the target docking position is minimized; A featured autonomous vehicle. The autonomous traveling vehicle according to claim 20 or 24,
Minimizing the distance between the first destination and the target docking position is:
By determining each distance between the target docking position and each candidate destination of the plurality of candidate destinations based on the traffic network information, a plurality of distances are generated,
The autonomous traveling vehicle is performed by selecting the candidate destination corresponding to the minimum distance from the plurality of distances as the first destination. The autonomously traveling vehicle according to claim 25,
The autonomous traveling vehicle, wherein the determination of each distance between the target docking position and one of the plurality of candidate destinations includes determination of a distance between the target docking position and the first destination. The autonomous vehicle according to claim 26,
Determination of the distance between the target docking position and the first destination is:
When the traffic network information specifies a set entrance position of the first destination, determine a distance between the target docking position and the set entrance position of the first destination;
The distance between the target docking position and the predicted entrance position of the first destination when the traffic network information excludes the set entrance position of the first destination and the traffic network information specifies the predicted entrance position Decide
When the traffic network information excludes the set entrance position of the first destination and the predicted entrance position of the first destination, and the traffic network information specifies the edge position of the first destination, Determining a distance between the target docking position and the edge position of the first destination;
The traffic network information excludes a set entrance position of the first destination, the traffic network information excludes the predicted entrance position of the first destination, and the traffic network information is an edge of the first destination. An autonomously traveling vehicle comprising: determining a distance between the target docking position and a set position of the first destination when specifying a part position. The autonomous vehicle according to claim 27,
The autonomous traveling vehicle, wherein the set position of the first destination is a global positioning system (GPS) position. The autonomous traveling vehicle according to claim 19 or 20,
Identification of the target docking position based on the plurality of docking position clusters is:
An autonomous traveling vehicle comprising: specifying the selected docking position cluster of the plurality of docking position clusters so that the size of the selected docking position cluster is maximized. The autonomously traveling vehicle according to claim 29,
Identification of the target docking position based on the plurality of docking position clusters is:
Identifying a docking position cluster arithmetic average of the selected docking position clusters;
Use of the docking position cluster arithmetic average as the target docking position. The autonomously traveling vehicle according to claim 29,
Identification of the target docking position based on the plurality of docking position clusters is:
Using the selected docking position cluster as the target docking position. The autonomous traveling vehicle according to claim 22,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information to include a set of docking positions of the plurality of docking positions;
Each docking position of the set of docking positions is associated with the first destination,
The association between the target docking position and the first destination is:
Identifying a plurality of docking position clusters based on the set of docking positions;
Based on the plurality of docking position clusters by associating the target docking position with the first destination;
The association between the target docking position and the first destination is:
An autonomous traveling vehicle comprising: specifying the selected docking position cluster of the plurality of docking position clusters so that the size of the selected docking position cluster is maximized. An autonomous traveling vehicle according to claim 32,
The association of the target docking position with the first destination is:
Identifying a docking position cluster arithmetic average of the selected docking position clusters;
Use of the docking position cluster arithmetic average as the target docking position. An autonomous traveling vehicle according to claim 32,
The association between the target docking position and the first destination is:
Using the selected docking position cluster as the target docking position. The autonomous traveling vehicle according to claim 18,
The processor executes instructions stored on the non-transitory computer readable medium,
An autonomous traveling vehicle characterized in that the traffic network information is specified so that at least one of the plurality of docking positions is associated with the first destination based on a pedestrian travel time. An autonomous traveling vehicle according to claim 35,
The processor executes instructions stored on a non-transitory computer readable medium,
Each docking position of the plurality of docking positions is
Each position of the vehicle traffic network;
Corresponding to road segment information indicating a plurality of road segments of the vehicle traffic network,
Based on the pedestrian travel time, the traffic network information is identified such that target docking positions of the plurality of docking positions are associated with the first destination,
Determining whether to use the target docking position based on the traffic network information;
The association between the target docking position based on the pedestrian travel time and the first destination is
Associating each docking position of the plurality of docking positions with each road segment of the plurality of road segments;
A plurality of docking position clusters included in each road segment of the plurality of road segments, wherein each docking position cluster of the plurality of docking position clusters includes at least one of the docking positions of the plurality of docking positions. Identifying the plurality of docking position clusters,
The plurality of docking position cluster medians, wherein each docking position cluster median of the plurality of docking position cluster medians is the plurality of docking positions such that the median value of each docking position cluster of the plurality of docking position clusters is indicated. Identify the cluster median,
Each vehicle traffic network region of the plurality of vehicle traffic network regions includes a docking position cluster median;
Each vehicle traffic network region of the plurality of vehicle traffic network regions includes a plurality of points of the vehicle traffic network,
Each predicted pedestrian travel time between each of the plurality of points and the docking position cluster median associated with the vehicle traffic network region is the center of each docking position cluster of the points and the plurality of docking position cluster medians. Dividing the vehicle traffic network generating the plurality of vehicle traffic network regions based on the median values of the plurality of docking positions so that they are within each estimated pedestrian travel time between values, Autonomous traveling vehicle. An autonomous traveling vehicle according to claim 35,
The processor executes instructions stored on a non-transitory computer readable medium,
The traffic network information includes docking position information indicating a plurality of docking positions,
Road segment information indicating a plurality of road segments of the vehicle traffic network, and
Based on the pedestrian travel time, the traffic network information is identified such that target docking positions of the plurality of docking positions are associated with the first destination,
Determining whether to use the target docking position based on the traffic network information;
The association between the target docking position based on the pedestrian travel time and the first destination is
Associating each docking position of the plurality of docking positions with each road segment of the plurality of road segments;
A plurality of docking position clusters included in each road segment of the plurality of road segments, wherein each docking position cluster of the plurality of docking position clusters includes at least one of the docking positions of the plurality of docking positions. Identifying the plurality of docking position clusters,
The plurality of docking position cluster medians, wherein each docking position cluster median of the plurality of docking position cluster medians is the plurality of docking positions such that the median value of each docking position cluster of the plurality of docking position clusters is indicated. Identify the cluster median,
Each vehicle traffic network region of the plurality of vehicle traffic network regions includes a docking position cluster median;
Each vehicle traffic network region of the plurality of vehicle traffic network regions includes a plurality of points of the vehicle traffic network,
Each predicted pedestrian travel time between each of the plurality of points and the docking position cluster median associated with the vehicle traffic network region is the center of each docking position cluster of the points and the plurality of docking position cluster medians. Dividing the vehicle traffic network to generate the plurality of vehicle traffic network regions based on the plurality of docking position cluster medians, so that each estimated pedestrian travel time between values
When the traffic network information includes entrance position information for specifying a set entrance position, the set entrance position of the vehicle traffic network is specified as the first destination,
When the traffic network information excludes the entrance position information specifying the set entrance position, a predicted entrance position is generated as the first destination,
By associating the position of the first destination with the median docking position cluster of the vehicle traffic network area;
The generation of the predicted entrance position is
Identifying the set of road segments from the plurality of road segments such that each road segment of the set of road segments is proximate to the first destination;
Generating a set of lines such that each line of the set of lines is perpendicular to a road segment of the set of road segments and intersects the first destination;
Each candidate predicted entry location represents an intersection of the line of the set of lines and a road segment of the set of road segments;
When the generation of the predicted entrance position includes identification of a plurality of predicted entrance positions, the generation of the predicted entrance position is performed for the docking position cluster median value of the vehicular traffic network region including the first destination. Based on the pedestrian travel time between the median position cluster and each predicted entrance position, including ordering of the plurality of candidate predicted entrance positions,
An autonomously traveling vehicle comprising: specifying a candidate predicted entrance position. An autonomous traveling vehicle according to claim 35,
The processor executes instructions stored on the non-transitory computer readable medium,
An autonomous traveling vehicle characterized in that the traffic network information is specified so as to include road segment information indicating a plurality of road segments of the vehicle traffic network. The autonomous traveling vehicle according to claim 38,
The processor executes instructions stored on the non-transitory computer readable medium,
Autonomy characterized by specifying the traffic network information so that the target docking position is associated with the first destination by associating each docking position of the plurality of docking positions with the plurality of road segments. Traveling vehicle. The autonomous traveling vehicle according to claim 39,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the target docking location is associated with the first destination by identifying a plurality of docking location clusters for each of the plurality of road segments;
Each of the docking position clusters of the plurality of docking position clusters includes at least one docking position of the plurality of docking positions. The autonomous traveling vehicle according to claim 40, wherein
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the target docking position is associated with the first destination by selecting a plurality of docking position cluster medians;
The autonomous traveling vehicle, wherein each docking position cluster median of the plurality of docking position clusters represents a median of each docking position cluster of the plurality of docking position clusters. The autonomous traveling vehicle according to claim 41,
The processor executes instructions stored on the non-transitory computer readable medium,
The traffic network information is generated such that the target docking position is associated with the first destination by generating the plurality of vehicle traffic network areas by dividing the vehicle traffic network based on the plurality of docking position cluster medians. An autonomous traveling vehicle characterized by specifying the vehicle. An autonomous traveling vehicle according to claim 42,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information by dividing the vehicle traffic network such that each vehicle traffic network area of the plurality of vehicle traffic network areas includes a docking position cluster median of the plurality of docking position clusters. Autonomous vehicle characterized by An autonomous traveling vehicle according to claim 43,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information by dividing the vehicle traffic network such that each vehicle traffic network region of the plurality of vehicle traffic network regions includes a plurality of points of the vehicle traffic network;
Each distance between each point of the plurality of points and the docking position cluster median value associated with the vehicle traffic network area is calculated as follows: the point and each docking position cluster median value of the plurality of docking position cluster median values Autonomous traveling vehicle characterized by being within each distance. An autonomous traveling vehicle according to claim 44,
The processor executes instructions stored on the non-transitory computer readable medium,
Autonomy characterized by identifying the traffic network information by dividing the vehicle traffic network so that the distance between each of the plurality of points and the median docking position cluster is the estimated pedestrian travel time Traveling vehicle. An autonomous traveling vehicle according to claim 35 or 36,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second destination of the vehicle traffic network in response to performing a docking operation at the target docking position;
Using the traffic network information to identify a subsequent route from the target docking location of the vehicle traffic network to the second destination;
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle using the route to travel the autonomous traveling vehicle from the target docking position to the second destination. An autonomous traveling vehicle according to claim 35 or 36,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
Each of the docking positions of the plurality of docking positions corresponds to each of the plurality of actions. The autonomous traveling vehicle according to claim 36 or 45,
The processor executes instructions stored on the non-transitory computer readable medium,
An association between the target docking position of the plurality of docking positions and the first destination is:
For the vehicle traffic network area including the first destination, an entrance position of the first destination in each vehicle traffic network area of the plurality of vehicle traffic network areas is specified,
An autonomous traveling vehicle characterized in that the traffic network information is specified by associating the entrance position with a median docking position cluster corresponding to the vehicle traffic network area. An autonomous traveling vehicle according to claim 48,
The entry position is specified as follows:
If the set entrance position is indicated in the traffic network information, the set entrance position indicated in the traffic network information is specified as the entrance position;
When the traffic network information excludes the set entrance position of the first destination, generating an estimated entrance position of the first destination as the entrance position. . An autonomous traveling vehicle according to claim 49,
The generation of the predicted entrance position is
Identifying a set of road segments from the plurality of road segments;
Each line of the set of lines generates a set of lines perpendicular to a road segment of the set of road segments and intersecting the first destination;
Identifying a candidate predicted entrance location,
Each road segment of the set of road segments is proximate to the first destination;
Each candidate predicted entrance position is an autonomous traveling vehicle, wherein each line of the set of lines intersects a road segment of the set of road segments. 51. The autonomous traveling vehicle of claim 50, wherein
If the generation of the predicted entrance position includes the identification of a plurality of candidate predicted entrance positions,
For each docking position cluster median of the vehicle traffic network area including the first destination, the plurality of candidate predicted entrance positions are determined based on a pedestrian travel time between the docking position cluster median and each predicted entrance position. Autonomous vehicle characterized by including ordering. An autonomous traveling vehicle according to claim 37,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second destination of the vehicle traffic network in response to performing a docking operation at the target docking position;
Using the traffic network information to identify a subsequent route from the target docking location of the vehicle traffic network to the second destination;
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle using the route to travel the autonomous traveling vehicle from the target docking position to the second destination. An autonomous traveling vehicle according to claim 37,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
Each of the docking positions of the plurality of docking positions corresponds to each of the plurality of actions. The autonomous traveling vehicle according to claim 18,
The processor executes instructions stored on a non-transitory computer readable medium,
Based on the traffic network information and pedestrian travel time, a target docking position is determined from the plurality of docking positions for the first destination,
An autonomous traveling vehicle, wherein a route from a starting point in the vehicle traffic network to the target docking position is specified as the first route using the traffic network information. An autonomous traveling vehicle according to claim 54,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
Each of the docking positions of the plurality of docking positions corresponds to each of the plurality of actions. An autonomous traveling vehicle according to claim 54,
The determination of the target docking position is as follows:
Receiving user input indicating the entrance location of the first destination;
Use of the entrance position as the first destination. An autonomous traveling vehicle according to claim 54,
The processor executes instructions stored on the non-transitory computer readable medium,
An autonomous traveling vehicle characterized by specifying the traffic network information so that the pedestrian traffic network information includes pedestrian traffic network information indicating a pedestrian traffic network. The autonomous traveling vehicle according to claim 57, wherein:
The processor executes instructions stored on the non-transitory computer readable medium,
The autonomous traveling vehicle, wherein the traffic network information is specified so that a part of the pedestrian traffic network is close to the first destination. The autonomous traveling vehicle according to claim 57, wherein:
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the traffic network information indicates a set of candidate docking positions of the plurality of docking positions;
The autonomous traveling vehicle, wherein each candidate docking position of the set of candidate docking positions is associated with the first destination. The autonomous vehicle according to claim 59,
The determination of the target docking position is as follows:
Excluding one candidate docking position from the plurality of candidate docking positions when a predicted pedestrian travel time between the candidate docking position and the first destination exceeds a maximum pedestrian travel time. Autonomous traveling vehicle. The autonomous vehicle according to claim 59,
The processor executes instructions stored on the non-transitory computer readable medium,
The autonomous traveling vehicle characterized by specifying the traffic network information so that each candidate docking position of the set of candidate docking positions is close to each part of the pedestrian traffic network. An autonomous traveling vehicle according to claim 61,
The processor executes instructions stored on the non-transitory computer readable medium,
Based on the traffic network information, from the plurality of docking positions, by selecting the target docking position from the plurality of candidate docking positions based on the pedestrian travel time between the target docking position and the first destination An autonomous traveling vehicle characterized by determining a target docking position. An autonomous traveling vehicle according to claim 62,
Selection of the target docking position from the plurality of candidate docking positions is
Generating a pedestrian determination model indicating a plurality of candidate routes between the first destination and each of the plurality of candidate docking positions. An autonomous traveling vehicle according to claim 63, wherein
The pedestrian determination model includes a plurality of routing states,
The autonomous traveling vehicle, wherein each candidate docking position of the plurality of candidate docking positions corresponds to each routing state. 65. An autonomous traveling vehicle according to claim 64, wherein
Generating the pedestrian determination model includes generating a plurality of predicted costs;
The plurality of predicted costs indicate an estimated pedestrian travel time between the first destination and each routing state of the plurality of routing states. An autonomous traveling vehicle according to claim 65,
The determination of the target docking position is as follows:
An estimated travel time between the departure point and the target docking position;
An autonomous traveling vehicle comprising: determining a minimum travel time including a sum of the target docking position and a predicted pedestrian travel time between the first destination. An autonomous traveling vehicle according to claim 65,
The identification of the target docking position is as follows:
Generating a vehicle decision model indicating a plurality of candidate routes between the departure point and each candidate docking position of the plurality of candidate docking positions;
An autonomous traveling vehicle including generation of an extended determination model based on the pedestrian determination model and the vehicle determination model. An autonomous traveling vehicle according to claim 67,
The action cost associated with using the candidate docking position as the target docking position includes the predicted pedestrian travel time between the target docking position and the first destination, and a set docking operation cost. An autonomous traveling vehicle that generates the extended determination model. An autonomous traveling vehicle according to claim 54,
The processor executes instructions stored on a non-transitory computer readable medium,
The traffic network information includes pedestrian traffic network information indicating a pedestrian traffic network,
So that part of the pedestrian traffic network is close to the first destination,
Identifying the traffic network information;
When the traffic network information indicates candidate docking positions of the plurality of docking positions associated with the first destination, the candidate docking position is used as a target docking position;
When the traffic network information indicates a set of candidate docking positions of the plurality of docking positions, and each candidate docking position of the candidate docking positions is associated with the first destination, the target docking position and the first purpose Based on the pedestrian travel time between the ground, by selecting the target docking position from the plurality of candidate docking positions,
The autonomous traveling vehicle, wherein the target docking position is determined from the plurality of docking positions for the first destination based on the traffic network information and a pedestrian travel time. An autonomous traveling vehicle according to claim 54 or 69,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second destination of the vehicle traffic network in response to performing a docking operation at the target docking position;
Identifying a second route from the target docking position of the vehicle traffic network to the second destination using the traffic network information;
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle, wherein the autonomous traveling vehicle travels from the target docking position to the second destination using the first route. The autonomous traveling vehicle according to claim 70, wherein
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying a second target docking position associated with the first destination;
An autonomous traveling vehicle generating a pedestrian movement route between the target docking position and the second target docking position. The autonomous traveling vehicle according to claim 71,
The processor executes instructions stored on the non-transitory computer readable medium,
Using the traffic network information to identify a third route from the second destination of the vehicle traffic network to the second target docking position;
The trajectory controller
An autonomous traveling vehicle, wherein the autonomous traveling vehicle travels from the second destination to the second target docking position using the third route. An autonomous traveling vehicle according to claim 54,
The processor executes instructions stored on a non-transitory computer readable medium,
The traffic network information includes pedestrian traffic network information indicating a pedestrian traffic network,
So that part of the pedestrian traffic network is close to the first destination,
Identifying the traffic network information;
When the traffic network information indicates candidate docking positions of the plurality of docking positions associated with the first destination, the candidate docking position is used as a target docking position;
When the traffic network information indicates a set of candidate docking positions of the plurality of docking positions, and each candidate docking position of the candidate docking positions is associated with the first destination, the target docking position and the first purpose Based on the pedestrian travel time between the ground, by selecting the target docking position from the plurality of candidate docking positions,
Determining the target docking position from the plurality of docking positions for the first destination based on the traffic network information and pedestrian travel time;
Identifying a second destination of the vehicle traffic network;
Identifying a second route from the target docking position of the vehicle traffic network to the second destination using the traffic network information;
Identifying a second target docking position associated with the first destination;
Using the traffic network information to identify a third route from the second destination of the vehicle traffic network to the second target docking position;
The track controller moves the autonomous vehicle from the target docking position to the second destination using the second route according to a docking operation at the target docking position.
The autonomously traveling vehicle is operated using the third route so as to move from the second destination to the second target docking position in accordance with execution of the operation at the second destination. Autonomous vehicle. An autonomous traveling vehicle according to claim 73,
The second destination is a parking area associated with the first destination;
The trajectory controller
Park the autonomous vehicle at the second destination;
An autonomous traveling vehicle, wherein the autonomous traveling vehicle travels from the target docking position to the second destination using the first route. An autonomous traveling vehicle according to claim 73,
The processor executes instructions stored on the non-transitory computer readable medium,
An autonomous traveling vehicle generating a pedestrian movement route between the target docking position and the second target docking position. An autonomous traveling vehicle according to claim 69 or 73,
The processor executes instructions stored on the non-transitory computer readable medium,
Identifying the traffic network information such that the docking position information is based on driving information of a plurality of vehicles;
The driving information includes a plurality of operations,
Each operation of the plurality of operations is associated with each vehicle of the plurality of vehicles,
Each of the docking positions of the plurality of docking positions corresponds to each of the plurality of actions. An autonomous traveling vehicle according to claim 69 or 73,
The determination of the target docking position is as follows:
Receiving user input indicating the entrance location of the first destination;
Use of the entrance position as the first destination. An autonomous traveling vehicle according to claim 69 or 73,
The determination of the target docking position is as follows:
Excluding one candidate docking position from the plurality of candidate docking positions when a predicted pedestrian travel time between the candidate docking position and the first destination exceeds a maximum pedestrian travel time. Autonomous traveling vehicle. An autonomous traveling vehicle according to claim 69 or 73,
The processor executes instructions stored on the non-transitory computer readable medium,
The autonomous traveling vehicle characterized by specifying the traffic network information so that each candidate docking position of the set of candidate docking positions is close to each part of the pedestrian traffic network. An autonomous traveling vehicle according to claim 69 or 73,
Selection of the target docking position from the plurality of candidate docking positions is
Generating a pedestrian determination model indicating a plurality of candidate routes between the first destination and each of the plurality of candidate docking positions. The autonomous vehicle according to claim 80, wherein
The pedestrian determination model includes a plurality of routing states,
The autonomous traveling vehicle, wherein each candidate docking position of the plurality of candidate docking positions corresponds to each routing state. The autonomous traveling vehicle according to claim 81,
Generating the pedestrian determination model includes generating a plurality of predicted costs;
The plurality of predicted costs indicate an estimated pedestrian travel time between the first destination and each routing state of the plurality of routing states. The autonomous traveling vehicle according to claim 82,
The determination of the target docking position is as follows:
An estimated travel time between the departure point and the target docking position;
An autonomous traveling vehicle comprising: determining a minimum travel time including a sum of the target docking position and a predicted pedestrian travel time between the first destination. The autonomous traveling vehicle according to claim 82,
The identification of the target docking position is as follows:
Generating a vehicle decision model indicating a plurality of candidate routes between the departure point and each candidate docking position of the plurality of candidate docking positions;
An autonomous traveling vehicle including generation of an extended determination model based on the pedestrian determination model and the vehicle determination model. The autonomous traveling vehicle according to claim 84,
The action cost associated with using the candidate docking position as the target docking position includes the predicted pedestrian travel time between the target docking position and the first destination, and a set docking operation cost. An autonomous traveling vehicle that generates the extended determination model.

Images