CN114261404A

CN114261404A - Automatic driving method and related device

Info

Publication number: CN114261404A
Application number: CN202010975573.1A
Authority: CN
Inventors: 车玉涵
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2020-09-16
Filing date: 2020-09-16
Publication date: 2022-04-01

Abstract

The application discloses an automatic driving method in the field of artificial intelligence, which can be applied to intelligent automobiles. The method comprises the following steps: acquiring first information of a first target vehicle and second information of a second target vehicle, wherein the self vehicle, the first target vehicle and the second target vehicle have collision risks, the first information comprises position information and speed information of the first target vehicle, and the second information comprises position information and speed information of the second target vehicle; establishing a dynamic game model at least according to first information and second information, wherein the first information is used for determining the collision loss in the dynamic game model, and the second information is used for determining the safety distance loss in the dynamic game model; determining a driving strategy according to the dynamic game model; and controlling the running of the vehicle according to the running strategy. Based on the scheme of this application, can improve the security of autopilot.

Description

Automatic driving method and related device

Technical Field

The present application relates to the field of automatic driving technologies, and in particular, to an automatic driving method and a related device.

Background

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making. Research in the field of artificial intelligence includes robotics, natural language processing, computer vision, decision and reasoning, human-computer interaction, recommendation and search, AI basic theory, and the like.

Automatic driving is a mainstream application in the field of artificial intelligence, and the automatic driving technology depends on the cooperative cooperation of computer vision, radar, a monitoring device, a global positioning system and the like, so that the motor vehicle can realize automatic driving without the active operation of human beings. Autonomous vehicles use various computing systems to assist in transporting passengers from one location to another. Some autonomous vehicles may require some initial input or continuous input from an operator, such as a pilot, driver, or passenger. Autonomous vehicles permit an operator to switch from a manual mode of operation to an autonomous mode or an intermediate mode. Because the automatic driving technology does not need human to drive the motor vehicle, the driving error of human can be effectively avoided theoretically, the occurrence of traffic accidents is reduced, and the transportation efficiency of the road can be improved. Therefore, the automatic driving technique is increasingly emphasized.

In the field of automatic driving, an automatic driving vehicle can execute a corresponding driving strategy according to an actual driving scene so as to ensure the safe driving of the automatic driving vehicle. However, in some special scenes, such as a left-turn-free protection intersection or a passageway ramp, the situation that the autonomous vehicle collides with the track of another vehicle sometimes occurs.

At present, when the track of an automatic driving vehicle conflicts with that of a target vehicle, the automatic driving vehicle often adopts a game strategy of one-to-one game with the target vehicle, driving risks outside the game process are ignored, and a driving strategy made by the automatic driving vehicle is easy to have certain dangerousness.

Disclosure of Invention

The application provides an automatic driving method and a related device, wherein a dynamic game model introducing conflict loss and safety distance loss is established by acquiring information of a game target vehicle having a game relation with a vehicle and information of a risk target vehicle influencing the decision of the vehicle. Because the risk influence of the driving environment on the game process is added into the dynamic game model, the risk decision and the game decision can be combined, the risk of the game process is effectively reduced, and the safety of automatic driving is improved.

The application provides an automatic driving method, which can be used in the automatic driving field in the field of artificial intelligence. The method can comprise the following steps: the method comprises the steps of obtaining first information of a first target vehicle and second information of a second target vehicle, wherein the first target vehicle and the second target vehicle can be vehicles in the surrounding environment of the self vehicle, the self vehicle has collision risks with the first target vehicle and the second target vehicle, and the self vehicle can eliminate the collision risks with the first target vehicle by accelerating to rush or decelerating to let the self vehicle run. In short, the vehicle and the first target vehicle have a direct game relationship, and in the game process, the driving strategies that the vehicle and the first target vehicle can make are acceleration rush to move or deceleration give way. The second target vehicle and the self vehicle do not have a direct game relationship, but the second target vehicle can influence and limit the decision and action range of the self vehicle in the game process. The first information may include position information and speed information of the first target vehicle, and the second information may include position information and speed information of the second target vehicle. And establishing a dynamic game model at least according to the first information and the second information, wherein the first information is used for determining the collision loss in the dynamic game model, and the second information is used for determining the safety distance loss in the dynamic game model. Wherein the collision loss represents the loss of the own vehicle when the own vehicle selects acceleration for overtaking and the first target vehicle also selects acceleration for overtaking; the safe distance loss represents the loss of the own vehicle when the own vehicle executes a strategy of accelerating and robbing the own vehicle or a strategy of decelerating and yielding the own vehicle so that the space distance between the own vehicle and the second target vehicle in front of or behind the own vehicle is changed. The loss of the vehicle refers to a measure of the change of one or two of road passing performance or safety of the vehicle caused by the selection of the game strategy of the vehicle in the game process. Determining a driving strategy according to the dynamic game model, wherein the driving strategy is acceleration line robbing or deceleration line giving; and controlling the running of the self vehicle according to the running strategy.

In the scheme, a dynamic game model introducing conflict loss and safety distance loss is established by acquiring information of game target vehicles having game relation with the self vehicle and information of risk target vehicles influencing decision making of the self vehicle. Because the risk influence of the driving environment on the game process is added into the dynamic game model, the risk decision and the game decision can be combined, the risk of the game process is effectively reduced, and the safety of automatic driving is improved.

In one possible embodiment, the automatic driving method further comprises: recognizing that the self-vehicle is in a preset scene, wherein the preset scene comprises: the method comprises the following steps of protecting a crossing driving scene, an entrance ramp driving scene and a lane merging driving scene without left turn. According to the scheme, when the fact that the self-vehicle is in the preset scene is recognized, the automatic driving method is selected to be automatically executed, decision-making accuracy under the special scene can be improved, and safety of automatic driving is improved.

In a possible implementation manner, the establishing a dynamic game model according to at least the first information and the second information may specifically include: acquiring Time To Conflict (TTC) corresponding To the vehicle from a track Conflict point; determining a TTC corresponding to the first target vehicle according to the first information; and determining the collision loss in the dynamic game model according to the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the first target vehicle. For the self-vehicle, the Distance To Conflict (DTC) corresponding To the self-vehicle can be determined according To the current position of the self-vehicle and the position of the track Conflict point. After the DTC corresponding to the vehicle is determined, the TTC corresponding to the vehicle can be determined according to the current speed and acceleration of the vehicle. Similarly, the DTC of the first target vehicle may also be determined according to the current location of the first target vehicle and the location of the trajectory conflict point, and then the corresponding TTC from the first target vehicle may be determined based on the DTC of the first target vehicle.

In the scheme, the collision loss in the dynamic game model is determined based on the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the first target vehicle, so that the collision loss which changes along with the TTC difference value can be introduced in the game process, and the decision accuracy of the dynamic game model is improved.

In a possible implementation manner, the establishing a dynamic game model according to at least the first information and the second information may specifically include: determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and the speed of the self-vehicle and the second information; the first safety distance loss is the front safety distance loss when the acceleration and preemption strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration and yielding strategy is executed. For example, when the second target vehicle is in front of the own vehicle, a first safe distance loss corresponding to the second target vehicle can be determined; when the second target vehicle is behind the own vehicle, the second safety distance loss corresponding to the second target vehicle can be determined. Specifically, the first safety distance loss or the second safety distance loss may be determined based on a spatial distance between the second target vehicle and the own vehicle, and a relative speed between the second target vehicle and the own vehicle. For example, the first safety distance loss or the second safety distance loss may be obtained by dividing a spatial distance between the second target vehicle and the own vehicle by a relative speed between the second target vehicle and the own vehicle. Generally, the larger the spatial distance between the second target vehicle and the own vehicle, the larger the first safety distance loss or the second safety distance loss; the greater the relative speed between the second target vehicle and the host vehicle, the smaller the first safety distance loss or the second safety distance loss. In the scheme, by introducing the loss of the safety distance into the dynamic game model, the game decision and the risk decision can be coupled, and the decision safety is improved.

In one possible embodiment, the automatic driving method further comprises: determining that the vehicle has the right of way according to the vehicle and the first target vehicle; and determining road right bank-robbing income according to the fact that the vehicle has the road right, wherein the road right bank-robbing income is used for establishing the dynamic game model. It is understood that vehicles with road rights can have a priority of passing during travel. Therefore, in the process of establishing the dynamic game model, road right bank-robbing benefits can be introduced, namely certain benefits are given to vehicles with road rights, so that the dynamic game model is more in line with the actual situation, and the reliability of the decision result is improved.

In one possible embodiment, before the obtaining the first information of the first target vehicle and the second information of the second target vehicle, the automatic driving method further includes: determining a first target vehicle set according to the position of the self vehicle, wherein the first target vehicle set comprises a plurality of target vehicles, and the distance between the target vehicles and the self vehicle is less than a preset threshold value; determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which predicted tracks are intersected with the planned track of the self vehicle; determining the first target vehicle and the second target vehicle in the second set of target vehicles. In the scheme, before the first information of the first target vehicle and the second information of the second target vehicle are obtained, the key obstacles corresponding to the vehicle can be identified according to the position and the planning track of the vehicle, and then the first target vehicle and the second target vehicle are determined in the key obstacles corresponding to the vehicle. By key obstacle identification, a large number of extraneous objects can be filtered out.

In a possible implementation, the determining the first target vehicle and the second target vehicle in the second target vehicle set specifically includes: acquiring a TTC corresponding to the self vehicle and a TTC corresponding to a third target vehicle, wherein the third target vehicle is a target vehicle in the second target vehicle set; determining a difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle; if the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle is smaller than or equal to a preset threshold value, and the third target vehicle is in a game vehicle lane set, determining that the third target vehicle is the first target vehicle; the game lane set is a lane set which has a game relation with a lane where the self vehicle is located.

In a possible implementation, the determining the first target vehicle and the second target vehicle in the second target vehicle set specifically includes: if a third target vehicle is in a risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, determining that the third target vehicle is a second target vehicle; wherein, the third target vehicle does target vehicle in the second target vehicle set, the car lane set includes the own lane the place ahead lane of own lane with the rear lane from the lane, the own lane does the lane that the own vehicle is located, the risk lane set includes the car lane set and the left lane set of the car lane set and/or the right lane set of the car lane set.

A second aspect of the present application provides an automatic driving apparatus comprising: the system comprises an acquisition unit, a processing unit and a control unit, wherein the acquisition unit is used for acquiring first information of a first target vehicle and second information of a second target vehicle, the self vehicle, the first target vehicle and the second target vehicle have collision risks, the first information comprises position information and speed information of the first target vehicle, and the second information comprises position information and speed information of the second target vehicle; the processing unit is used for establishing a dynamic game model at least according to the first information and the second information, wherein the first information is used for determining collision loss in the dynamic game model, the second information is used for determining safety distance loss in the dynamic game model, the collision loss is the loss of the self vehicle when the self vehicle and the first target vehicle select acceleration and preemption, and the safety distance loss is the loss of the self vehicle when the safety distance between the self vehicle and the second target vehicle changes; the processing unit is further used for determining a driving strategy according to the dynamic game model; and the control unit is used for controlling the running of the vehicle according to the running strategy.

In one possible embodiment, the autopilot device further comprises: the recognition unit is used for recognizing that the self-vehicle is in a preset scene, and the preset scene comprises: the method comprises the following steps of protecting a crossing driving scene, an entrance ramp driving scene and a lane merging driving scene without left turn.

In one possible embodiment, the host vehicle can eliminate the risk of collision with the first target vehicle by accelerating to grab the vehicle or decelerating to give way.

In a possible implementation, the processing unit is specifically configured to: acquiring time TTC from a track conflict point corresponding to the self-vehicle; determining a TTC corresponding to the first target vehicle according to the first information; and determining the collision loss in the dynamic game model according to the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the first target vehicle.

In a possible implementation, the processing unit is specifically configured to: determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and the speed of the self-vehicle and the second information; the first safety distance loss is the front safety distance loss when the acceleration and preemption strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration and yielding strategy is executed.

In a possible implementation, the processing unit is further configured to: determining that the vehicle has the right of way according to the vehicle and the first target vehicle; and determining road right bank-robbing income according to the fact that the vehicle has the road right, wherein the road right bank-robbing income is used for establishing the dynamic game model.

In a possible implementation, the processing unit is further configured to: determining a first target vehicle set according to the position of the self vehicle, wherein the first target vehicle set comprises a plurality of target vehicles, and the distance between the target vehicles and the self vehicle is less than a preset threshold value; determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which predicted tracks are intersected with the planned track of the self vehicle; determining the first target vehicle and the second target vehicle in the second set of target vehicles.

In a possible implementation manner, the obtaining unit is further configured to obtain a TTC corresponding to the self vehicle and a TTC corresponding to a third target vehicle, where the third target vehicle is a target vehicle in the second set of target vehicles; the processing unit is further configured to determine a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle; the processing unit is further configured to determine that the third target vehicle is the first target vehicle if a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle is less than or equal to a preset threshold and the third target vehicle is in a game vehicle lane set; the game lane set is a lane set which has a game relation with a lane where the self vehicle is located.

In a possible implementation, the processing unit is specifically configured to: if a third target vehicle is in a risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, determining that the third target vehicle is a second target vehicle; wherein, the third target vehicle does target vehicle in the second target vehicle set, the car lane set includes the own lane the place ahead lane of own lane with the rear lane from the lane, the own lane does the lane that the own vehicle is located, the risk lane set includes the car lane set and the left lane set of the car lane set and/or the right lane set of the car lane set.

A third aspect of the present application provides an autopilot device, which may include a processor, a memory coupled to the processor, the memory storing program instructions that, when executed by the processor, perform the autopilot method described in the first aspect or any one of the possible implementations of the first aspect.

A fourth aspect of the present application provides a computer-readable storage medium, which may include a program that, when run on a computer, causes the computer to perform the autopilot method described in the first aspect or any one of the possible implementations of the first aspect.

A fifth aspect of the present application provides an autonomous vehicle that may comprise a processing circuit and a memory circuit configured to perform the autonomous driving method described in the first aspect or any one of the possible embodiments of the first aspect.

A sixth aspect of the present application provides circuitry that may include processing circuitry configured to perform the autopilot method described in the first aspect or any one of the possible implementations of the first aspect.

A seventh aspect of the present application provides a computer program that, when running on a computer, causes the computer to execute the automatic driving method described in the first aspect or any one of the possible embodiments of the first aspect.

An eighth aspect of the present application provides a chip system that may include a processor for enabling an autopilot device to perform the functions referred to in the above aspects, e.g., to transmit or process data and/or information referred to in the above methods. In one possible design, the system-on-chip may further include a memory, storage, for storing program instructions and data necessary for the server or communication device. The chip system may be formed by a chip, or may include a chip and other discrete devices.

For specific implementation steps of the second aspect to the ninth aspect and various possible implementation manners and beneficial effects brought by each possible implementation manner in the present application, reference may be made to descriptions in various possible implementation manners in the first aspect, and details are not repeated here.

Drawings

FIG. 1 is a schematic structural diagram of a vehicle according to an embodiment of the present disclosure;

fig. 2 is a schematic view of a crossing driving scene without left turn protection according to an embodiment of the present application;

fig. 3 is a schematic view of a driving scenario of an entrance ramp according to an embodiment of the present application;

fig. 4 is a schematic view of a lane-merging driving scenario provided in an embodiment of the present application;

FIG. 5 is a schematic flow chart of an automatic driving method provided herein;

fig. 6 is a schematic flowchart of a method for identifying a target vehicle according to an embodiment of the present disclosure;

fig. 7 is a schematic diagram of a lane dividing method according to an embodiment of the present application;

fig. 8 is a schematic diagram of a gaming lane dividing manner provided in the embodiment of the present application;

fig. 9 is a schematic diagram of another gaming lane dividing manner provided in the embodiment of the present application;

FIG. 10 is a schematic diagram of a method for filtering out a first target vehicle and a second target vehicle according to an embodiment of the present disclosure;

fig. 11 is a schematic structural diagram of an autopilot device 1100 according to an embodiment of the present application;

FIG. 12 is a schematic diagram of an autonomous vehicle according to an embodiment of the present disclosure;

fig. 13 is a schematic structural diagram of a chip according to an embodiment of the present disclosure.

Detailed Description

The terms "first," "second," and the like in the description and in the claims of the present application and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances and are merely descriptive of the various embodiments of the application and how objects of the same nature can be distinguished. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

Embodiments of the present application are described below with reference to the accompanying drawings. As can be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solution provided in the embodiments of the present application is also applicable to similar technical problems.

In order to facilitate understanding of the present solution, in the embodiment of the present application, first, a structure of a vehicle is described with reference to fig. 1, and the automatic driving method provided in the embodiment of the present application may be applied to the vehicle shown in fig. 1. Referring to fig. 1, fig. 1 is a schematic structural diagram of a vehicle according to an embodiment of the present disclosure.

In one embodiment, the autonomous vehicle 100 may be configured in a fully or partially autonomous mode. For example, the autonomous vehicle 100 may control itself while in the autonomous mode, and may determine a current state of the vehicle and its surroundings by human operation, determine a possible behavior of at least one other vehicle in the surroundings, and determine a confidence level corresponding to a likelihood that the other vehicle performs the possible behavior, controlling the autonomous vehicle 100 based on the determined information. When the autonomous vehicle 100 is in the autonomous mode, the autonomous vehicle 100 may be placed into operation without human interaction.

Autonomous vehicle 100 may include various subsystems such as a travel system 102, a sensor system 104, a control system 106, one or more peripherals 108, as well as a power supply 110, a computer system 112, and a user interface 116. Alternatively, the autonomous vehicle 100 may include more or fewer subsystems, and each subsystem may include multiple elements. In addition, each of the sub-systems and elements of autonomous vehicle 100 may be interconnected by wires or wirelessly.

The travel system 102 may include components that provide powered motion to the autonomous vehicle 100. In one embodiment, the propulsion system 102 may include an engine 118, an energy source 119, a transmission 120, and wheels/tires 121. The engine 118 may be an internal combustion engine, an electric motor, an air compression engine, or other types of engine combinations, such as a hybrid engine of a gasoline engine and an electric motor, or a hybrid engine of an internal combustion engine and an air compression engine. The engine 118 converts the energy source 119 into mechanical energy.

Examples of energy sources 119 include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electrical power. The energy source 119 may also provide energy to other systems of the autonomous vehicle 100.

The transmission 120 may transmit mechanical power from the engine 118 to the wheels 121. The transmission 120 may include a gearbox, a differential, and a drive shaft. In one embodiment, the transmission 120 may also include other devices, such as a clutch. Wherein the drive shaft may comprise one or more shafts that may be coupled to one or more wheels 121.

The sensor system 104 may include a number of sensors that sense information about the environment surrounding the autonomous vehicle 100. For example, the sensor system 104 may include a positioning system 122 (which may be a GPS system, a beidou system, or other positioning system), an Inertial Measurement Unit (IMU) 124, a radar 126, a laser range finder 128, and a camera 130. The sensor system 104 may also include sensors that are monitored for internal systems of the autonomous vehicle 100 (e.g., an in-vehicle air quality monitor, a fuel gauge, an oil temperature gauge, etc.). Sensor data from one or more of these sensors may be used to detect the object and its corresponding characteristics (position, shape, orientation, velocity, etc.). Such detection and identification is a key function of safe operation of the autonomous vehicle 100.

The positioning system 122 may be used to estimate the geographic location of the autonomous vehicle 100. The IMU 124 is used to sense position and orientation changes of the autonomous vehicle 100 based on inertial acceleration. In one embodiment, IMU 124 may be a combination of an accelerometer and a gyroscope.

The radar 126 may utilize radio signals to sense objects within the surrounding environment of the autonomous vehicle 100. In some embodiments, in addition to sensing objects, radar 126 may also be used to sense the speed and/or heading of an object.

The laser rangefinder 128 may utilize a laser to sense objects in the environment in which the autonomous vehicle 100 is located. In some embodiments, the laser rangefinder 128 may include one or more laser sources, laser scanners, and one or more detectors, among other system components.

The camera 130 may be used to capture multiple images of the surrounding environment of the autonomous vehicle 100. The camera 130 may be a still camera or a video camera.

The control system 106 is for controlling the operation of the autonomous vehicle 100 and its components. The control system 106 may include various elements including a steering system 132, a throttle 134, a braking unit 136, a sensor fusion algorithm 138, a computer vision system 140, a route control system 142, and an obstacle avoidance system 144.

The steering system 132 is operable to adjust the heading of the autonomous vehicle 100. For example, in one embodiment, a steering wheel system.

The throttle 134 is used to control the operating speed of the engine 118 and thus the speed of the autonomous vehicle 100.

The brake unit 136 is used to control the deceleration of the autonomous vehicle 100. The brake unit 136 may use friction to slow the wheel 121. In other embodiments, the brake unit 136 may convert the kinetic energy of the wheel 121 into an electric current. The brake unit 136 may also take other forms to slow the rotational speed of the wheels 121 to control the speed of the autonomous vehicle 100.

The computer vision system 140 may be operable to process and analyze images captured by the camera 130 to identify objects and/or features in the environment surrounding the autonomous vehicle 100. The objects and/or features may include traffic signals, road boundaries, and obstacles. The computer vision system 140 may use object recognition algorithms, Motion from Motion (SFM) algorithms, video tracking, and other computer vision techniques. In some embodiments, the computer vision system 140 may be used to map an environment, track objects, estimate the speed of objects, and so forth.

The route control system 142 is used to determine a travel route for the autonomous vehicle 100. In some embodiments, the route control system 142 may combine data from the sensors 138, the GPS 122, and one or more predetermined maps to determine a travel route for the autonomous vehicle 100.

The obstacle avoidance system 144 is used to identify, evaluate, and avoid or otherwise negotiate potential obstacles in the environment of the vehicle 100.

Of course, in one example, the control system 106 may additionally or alternatively include components other than those shown and described. Or may reduce some of the components shown above.

The autonomous vehicle 100 interacts with external sensors, other vehicles, other computer systems, or users through peripherals 108. The peripheral devices 108 may include a wireless communication system 146, an in-vehicle computer 148, a microphone 150, and/or speakers 152.

In some embodiments, the peripheral devices 108 provide a means for a user of the autonomous vehicle 100 to interact with the user interface 116. For example, the onboard computer 148 may provide information to a user of the autonomous vehicle 100. The user interface 116 may also operate the in-vehicle computer 148 to receive user input. The in-vehicle computer 148 may be operated via a touch screen. In other cases, peripheral devices 108 may provide a means for autonomous vehicle 100 to communicate with other devices located within the vehicle. For example, the microphone 150 may receive audio (e.g., voice commands or other audio input) from a user of the autonomous vehicle 100. Similarly, the speaker 152 may output audio to a user of the autonomous vehicle 100.

The wireless communication system 146 may communicate wirelessly with one or more devices, either directly or via a communication network. For example, the wireless communication system 146 may use 3G cellular communication, such as CDMA, EVD0, GSM/GPRS, or 4G cellular communication, such as LTE. Or 5G cellular communication. The wireless communication system 146 may communicate with a Wireless Local Area Network (WLAN) using WiFi. In some embodiments, the wireless communication system 146 may utilize an infrared link, bluetooth, or ZigBee to communicate directly with the device. Other wireless protocols, such as various vehicle communication systems, for example, the wireless communication system 146 may include one or more Dedicated Short Range Communications (DSRC) devices that may include public and/or private data communications between vehicles and/or roadside stations.

The power supply 110 may provide power to various components of the autonomous vehicle 100. In one embodiment, power source 110 may be a rechargeable lithium ion or lead acid battery. One or more battery packs of such batteries may be configured as a power source to provide power to various components of the autonomous vehicle 100. In some embodiments, the power source 110 and the energy source 119 may be implemented together, such as in some all-electric vehicles.

Some or all of the functions of the autonomous vehicle 100 are controlled by the computer system 112. The computer system 112 may include at least one processor 113, the processor 113 executing instructions 115 stored in a non-transitory computer readable medium, such as a data storage device 114. The computer system 112 may also be a plurality of computing devices that control individual components or subsystems of the autonomous vehicle 100 in a distributed manner.

The processor 113 may be any conventional processor, such as a commercially available CPU. Alternatively, the processor may be a dedicated device such as an ASIC or other hardware-based processor. Although fig. 1 functionally illustrates processors, memories, and other elements of the computer 110 in the same blocks, those of ordinary skill in the art will appreciate that the processors, computers, or memories may actually comprise multiple processors, computers, or memories that may or may not be stored within the same physical housing. For example, the memory may be a hard disk drive or other storage medium located in a different housing than the computer 110. Thus, references to a processor or computer are to be understood as including references to a collection of processors or computers or memories which may or may not operate in parallel. Rather than using a single processor to perform the steps described herein, some components, such as the steering component and the retarding component, may each have their own processor that performs only computations related to the component-specific functions.

In various aspects described herein, the processor may be located remotely from the vehicle and in wireless communication with the vehicle. In other aspects, some of the processes described herein are executed on a processor disposed within the vehicle and others are executed by a remote processor, including taking the steps necessary to perform a single maneuver.

In some embodiments, the data storage device 114 may include instructions 115 (e.g., program logic), the instructions 115 being executable by the processor 113 to perform various functions of the autonomous vehicle 100, including those described above. The data storage 114 may also contain additional instructions, including instructions to send data to, receive data from, interact with, and/or control one or more of the propulsion system 102, the sensor system 104, the control system 106, and the peripherals 108.

In addition to instructions 115, data storage device 114 may also store data such as road maps, route information, the location, direction, speed of the vehicle, and other such vehicle data, among other information. Such information may be used by the autonomous vehicle 100 and the computer system 112 during operation of the autonomous vehicle 100 in autonomous, semi-autonomous, and/or manual modes.

A user interface 116 for providing information to or receiving information from a user of the autonomous vehicle 100. Optionally, the user interface 116 may include one or more input/output devices within the collection of peripheral devices 108, such as a wireless communication system 146, an on-board vehicle computer 148, a microphone 150, and a speaker 152.

The computer system 112 may control the functions of the autonomous vehicle 100 based on inputs received from various subsystems (e.g., the travel system 102, the sensor system 104, and the control system 106) and from the user interface 116. For example, the computer system 112 may utilize input from the control system 106 in order to control the steering unit 132 to avoid obstacles detected by the sensor system 104 and the obstacle avoidance system 144. In some embodiments, the computer system 112 is operable to provide control over many aspects of the autonomous vehicle 100 and its subsystems.

Alternatively, one or more of these components described above may be mounted or associated separately from the autonomous vehicle 100. For example, the data storage device 114 may exist partially or completely separate from the vehicle 1100. The above components may be communicatively coupled together in a wired and/or wireless manner.

Optionally, the above components are only an example, in an actual application, components in the above modules may be added or deleted according to an actual need, and fig. 1 should not be construed as limiting the embodiment of the present application.

Autonomous cars traveling on the road, such as autonomous vehicle 100 above, may identify objects within their surrounding environment to determine an adjustment to the current speed. The object may be another vehicle, a traffic control device, or another type of object. In some examples, each identified object may be considered independently, and based on the respective characteristics of the object, such as its current speed, acceleration, separation from the vehicle, etc., may be used to determine the speed at which the autonomous vehicle is to be adjusted.

Alternatively, the autonomous vehicle 100 or a computing device associated with the autonomous vehicle 100 (e.g., the computer system 112, the computer vision system 140, the data storage 114 of fig. 1) may predict behavior of the identified object based on characteristics of the identified object and the state of the surrounding environment (e.g., traffic, rain, ice on the road, etc.). Optionally, each identified object depends on the behavior of each other, so it is also possible to predict the behavior of a single identified object taking all identified objects together into account. The autonomous vehicle 100 is able to adjust its speed based on the predicted behavior of the identified object. In other words, the autonomous vehicle is able to determine what steady state the vehicle will need to adjust to (e.g., accelerate, decelerate, or stop) based on the predicted behavior of the object. In this process, other factors may also be considered to determine the speed of the autonomous vehicle 100, such as the lateral position of the autonomous vehicle 100 in the road being traveled, the curvature of the road, the proximity of static and dynamic objects, and so forth.

In addition to providing instructions to adjust the speed of the autonomous vehicle, the computing device may also provide instructions to modify the steering angle of the autonomous vehicle 100 to cause the autonomous vehicle to follow a given trajectory and/or maintain a safe lateral and longitudinal distance from objects in the vicinity of the autonomous vehicle (e.g., cars in adjacent lanes on a road).

The autonomous vehicle 100 may be a car, a truck, a motorcycle, a bus, a boat, an airplane, a helicopter, a lawn mower, an amusement car, a playground vehicle, construction equipment, an electric car, a golf cart, a train, etc., and the embodiment of the present invention is not particularly limited.

With the further development and the continuous evolution and iteration of the automatic driving technology, the requirements of people on the automatic driving vehicle are higher and higher, and the behaviors of the automatic driving vehicle are required to be more intelligent and humanized. In the related art of autonomous driving, safety is generally the primary criterion, and it is desirable to provide an overly redundant safety margin at all times. Therefore, the driving strategy of the automatic driving vehicle tends to be conservative, and the automatic driving vehicle is easy to be stopped in some special scenes, sometimes even blocks the traffic, and causes the situation of manual takeover.

The special scene may be a scene in which the autonomous vehicle and the target vehicle have a track conflict, for example, a left-turn-free intersection driving scene, an entrance-exit ramp driving scene, a lane merging driving scene, and the like.

For example, referring to fig. 2, fig. 2 is a schematic view of a driving scene of a left turn-free protected intersection according to an embodiment of the present application. As shown in fig. 2, at the intersection without the left turn protection, the planned trajectory of the own vehicle (i.e., the own vehicle) is straight, and the planned trajectory of the target vehicle is left turn. Since the same green light is shared by the straight running and the left turning, the situation that the tracks of the self vehicle and the target vehicle conflict with each other often occurs.

For example, referring to fig. 3, fig. 3 is a schematic diagram of an entrance ramp driving scenario provided in an embodiment of the present application. As shown in fig. 3, on the on-off ramp, the planned trajectory of the host vehicle is straight, and the planned trajectory of the target vehicle is a lane on which the host vehicle runs and merges from the ramp. In this case, the collision of the own vehicle and the target vehicle track may occur.

For example, referring to fig. 4, fig. 4 is a schematic view of a lane merging driving scenario provided in an embodiment of the present application. As shown in fig. 4, in a scene where two lanes are merged into a single lane, the planned trajectory of the host vehicle is straight, and the planned trajectory of the target vehicle is the lane after the target vehicle enters the merged lane. In this case, the collision between the own vehicle and the target vehicle track usually occurs.

In the related art, when the vehicle is subjected to track conflict, the decision of automatically driving the vehicle is very conservative, and the vehicle is selected to slow down and give the way in most cases, so that the passing efficiency is very low. Therefore, the appropriate interactive game strategy is selected, the passing efficiency in a track conflict scene can be greatly improved, the taking-over times caused by conservative decision are reduced, and the more intelligent embodiment of automatic driving is realized.

At present, when the trajectory of an automatic driving vehicle conflicts with that of a target vehicle, the automatic driving vehicle often adopts a game strategy of one-to-one game with the target vehicle, and influences of a driving environment on a game process are ignored, namely driving risks except the one-to-one game process are ignored, so that certain dangerousness exists in a driving strategy made by the automatic driving vehicle.

For example, when the autonomous vehicle selects a driving maneuver to accelerate for preemption during the game, the acceleration for preemption of the autonomous vehicle may cause the autonomous vehicle to reduce the safe distance from the vehicle in front of the autonomous vehicle, thereby resulting in a risk of collision of the autonomous vehicle with the vehicle in front of the autonomous vehicle. For another example, when the autonomous vehicle selects a driving maneuver for deceleration yielding during the game, the deceleration yielding of the autonomous vehicle may reduce the safe distance between the autonomous vehicle and the vehicle behind the autonomous vehicle, thereby resulting in a risk of collision between the autonomous vehicle and the vehicle behind the autonomous vehicle.

That is, the autonomous vehicle is at risk of colliding with its surrounding vehicles outside of the one-to-one gaming process, and existing gaming strategies typically ignore the driving risks outside of the one-to-one gaming process.

In view of the above, the present embodiment provides an automatic driving method, which may be applied to the automatic driving vehicle 100 shown in fig. 1 to ensure safe driving of the automatic driving vehicle in a trajectory conflict scenario.

For the sake of understanding, the technical terms mentioned in the embodiments of the present application will be described below.

And (4) crossing: an intersection region where two or more roads intersect.

The intersection is protected without left turn: the vehicle goes straight and the vehicle turns left at the intersection sharing the same green light, and no independent left-turn light is provided.

Own vehicle (ego): the automatic driving vehicle is taken as the self vehicle.

Target car (object): automatically driving a vehicle detected in the vicinity of the vehicle.

Dynamic game: a game theory that has precedence in the game process and the subsequent actors can make corresponding decisions according to the behaviors of the previous actors is also called a multi-stage game.

And (4) sub-game: in the dynamic game, all participants take actions one after another to form a new game, and each game in the game is called a sub-game.

Sub-game refinement nash equalization: sub-game refinement nash balance is formed if and only if the participant's strategy is in the series of sub-games (second generation, third generation …) where each sub-game forms nash balance. Sub-game refinement Nash equilibrium requires that the participant's decision be optimal in the sub-game at each time node.

Referring to fig. 5, fig. 5 is a schematic flow chart of an automatic driving method provided in the present application. As shown in fig. 5, an automatic driving method provided in an embodiment of the present application may include the following steps:

501. first information of a first target vehicle and second information of a second target vehicle are acquired.

The first target vehicle and the second target vehicle can be vehicles in the surrounding environment of the self vehicle, the first target vehicle and the second target vehicle have collision risks, and the self vehicle can accelerate to rush to run or decelerate to let run to eliminate the collision risks with the first target vehicle. That is, the vehicle and the first target vehicle have a direct game relationship, and in the game process, the driving strategies that the vehicle and the first target vehicle can make are acceleration racing or deceleration yielding. The second target vehicle and the self vehicle do not have a direct game relationship, but the second target vehicle can influence and limit the decision and action range of the self vehicle in the game process.

The first information of the first target vehicle includes position information and speed information of the first target vehicle, and the second information of the second target vehicle includes position information and speed information of the second target vehicle. The position information of the first target vehicle and the second target vehicle may be accomplished by a Global Positioning System (GPS), a real-time kinematic (RTK), a camera, a laser radar, and the like. In one possible implementation, the specific location of the vehicle may be determined by determining the location where the vehicle may exist by combining a map, GPS location information, and millimeter wave measurement information stored in advance, and calculating the probability of occurrence of the location where the vehicle may exist. It should be noted that the solution provided in the present application may obtain the position information of the vehicle through various ways, and the embodiments of the present application may be applied to the way of obtaining the position information of the vehicle in the related art. The speed information of the first target vehicle and the second target vehicle can be measured by a speed measuring radar or a sensor on the self vehicle. The speed information of the first target vehicle and the second target vehicle may include, for example, information of real-time speed and acceleration of the first target vehicle and the second target vehicle.

In one possible embodiment, the self-vehicle can rebuild the dynamic game model when recognizing that the self-vehicle is in the preset scene. And under the preset scene, the self vehicle and a target vehicle close to the self vehicle have collision risks, and the self vehicle can eliminate the collision risks with the target vehicle by accelerating to rush or decelerating to give way. Illustratively, the preset scenes include, but are not limited to, a left turn-free intersection driving scene, an entrance ramp driving scene, and a lane merging driving scene. The preset scene may also be other scenes meeting the foregoing conditions, and this embodiment does not specifically limit the preset scene.

For convenience of description, the present application will be described below by taking a preset scene as a left turn-free protection intersection driving scene as an example.

502. And establishing a dynamic game model at least according to the first information and the second information, wherein the first information is used for determining the collision loss in the dynamic game model, and the second information is used for determining the safety distance loss in the dynamic game model.

After the first information of the first target vehicle and the second information of the second target vehicle are acquired, a dynamic game model can be established based on the first information and the second information so as to iteratively obtain a driving strategy of the self vehicle.

Since the first target vehicle has a direct game relationship with the vehicle, the collision loss in the dynamic game model can be determined based on the first information of the first target vehicle to represent the loss of the vehicle when the vehicle chooses to accelerate and grab the vehicle and the first target vehicle also chooses to accelerate and grab the vehicle. The loss of the vehicle refers to a measure of the change of one or two of road passing performance or safety of the vehicle caused by the selection of the game strategy of the vehicle in the game process.

In one possible implementation, the collision loss may be determined according To a Time To Collision (TTC) from the trajectory Conflict point corresponding To the own vehicle and a TTC corresponding To the first target vehicle.

For example, after the planned trajectory of the host vehicle and the predicted trajectory of the first target vehicle are acquired, a trajectory conflict point, that is, a point where the planned trajectory and the predicted trajectory intersect, may be determined based on the planned trajectory of the host vehicle and the predicted trajectory of the first target vehicle. For the self-vehicle, the Distance To Conflict (DTC) corresponding To the self-vehicle can be determined according To the current position of the self-vehicle and the position of the track Conflict point. After the DTC corresponding to the vehicle is determined, the TTC corresponding to the vehicle can be determined according to the current speed and acceleration of the vehicle.

Specifically, the DTC corresponding to the own vehicle can be determined according to formula (1), where formula (1) is as follows:

wherein, X_iAnd Y_iRespectively representing the abscissa and the ordinate of the ith point of the track; x_i+1And Y_i+1Respectively representing the abscissa and the ordinate of the (i + 1) th point of the track; idx_cIndicating the serial number of the track intersection in the track.

The TTC corresponding to the own vehicle may be determined based on formula (2), where formula (2) is as follows:

where a represents acceleration and v represents velocity.

Similarly, the DTC and TTC of the first target vehicle can also be found according to the above equations (1) and (2).

After the TTC corresponding to the own vehicle and the TTC of the first target vehicle are obtained, a difference value between the two TTCs can be determined, and a collision loss in the dynamic game model is determined according to the difference value.

For example, the difference between the TTC corresponding to the own vehicle and the TTC corresponding to the first target vehicle may be determined based on equation (3), where equation (3) is as follows:

δT＝|TTC_e-TTC_oequation (3)

Wherein δ T represents a difference between a TTC corresponding to the own vehicle and a TTC corresponding to the first target vehicle, and the TTC_eIndicating TTC corresponding to the vehicle_oIndicating the TTC corresponding to the first target vehicle.

After the difference value δ T is acquired, the collision loss C may be found based on a preset function and the difference value δ T_e(δ T). The preset function can be a linear function, an exponential function or a logarithmic function, and the embodiment of the application does not deal with the difference value delta T and the collision loss C_eThe functional relationship between (δ T) is specifically defined. For example, the predetermined function may be an inverse proportional function, i.e., C_e(δ T) ═ 1/δ T. Generally, the larger the difference δ T, the smaller the collision loss, and the smaller the difference δ T, the larger the collision loss.

Based on the driving information of the self vehicle and the second information of the second target vehicle, a safe distance loss in the dynamic game model can be determined, wherein the safe distance loss is used for representing a strategy that the self vehicle performs acceleration and preemption or a strategy that performs deceleration and yielding, and the safe distance loss causes the loss of the self vehicle when the spatial distance between the self vehicle and the second target vehicle in front of or behind the self vehicle changes.

For example, during actual driving, a first safety distance loss or a second safety distance loss corresponding to the second target vehicle may be determined according to the relative position of the second target vehicle and the own vehicle.

The first safe distance loss represents a loss of the own vehicle when the own vehicle executes an acceleration preemption strategy, and the probability of a transmission collision between the own vehicle and a second target vehicle in front of the own vehicle is increased due to a decrease in the front safe distance between the autonomous vehicle and the second target vehicle in front of the autonomous vehicle. The second safe distance loss represents a loss of the own vehicle when the own vehicle executes a deceleration passing strategy, and the probability that the own vehicle collides with a second target vehicle behind the own vehicle increases due to a decrease in the rear safe distance between the autonomous vehicle and the second target vehicle behind the autonomous vehicle. For example, when the second target vehicle is in front of the own vehicle, a first safe distance loss corresponding to the second target vehicle can be determined; when the second target vehicle is behind the own vehicle, the second safety distance loss corresponding to the second target vehicle can be determined.

Specifically, the first safety distance loss or the second safety distance loss may be determined based on a spatial distance between the second target vehicle and the own vehicle, and a relative speed between the second target vehicle and the own vehicle. For example, the first safety distance loss or the second safety distance loss may be obtained by dividing a spatial distance between the second target vehicle and the own vehicle by a relative speed between the second target vehicle and the own vehicle. Generally, the larger the spatial distance between the second target vehicle and the own vehicle, the larger the first safety distance loss or the second safety distance loss; the greater the relative speed between the second target vehicle and the host vehicle, the smaller the first safety distance loss or the second safety distance loss.

In one possible embodiment, it may be further determined whether the own vehicle has the right of way based on the own vehicle and the first target vehicle. And under the condition that the vehicle is determined to have the road right, the road right robbing income of the vehicle can be determined, and the road right robbing income can be used for establishing the dynamic game model. It is understood that vehicles with road rights can have a priority of passing during travel. Therefore, in the process of establishing the dynamic game model, road right bank-robbing benefits can be introduced, namely certain benefits are given to vehicles with road rights, so that the dynamic game model is more in line with the actual situation, and the reliability of the decision result is improved. Generally, a vehicle in the traffic flow has a right of way with respect to an isolated vehicle (i.e., no other vehicle is in front of or behind the vehicle); a straight-ahead vehicle has road rights relative to a turning vehicle, such as a left or right turn vehicle. In practical situations, the own vehicle can determine whether the own vehicle has the right of way according to the driving scene and the traffic flow situation.

503. And determining a driving strategy according to the dynamic game model.

After the dynamic game model is established, the driving strategy of the self vehicle can be determined based on the dynamic game model, and the driving strategy is acceleration and line grabbing or deceleration and line giving.

For ease of understanding, the process of establishing a dynamic gaming model and determining a driving strategy based on the dynamic gaming model will be described below.

Firstly, determining a plurality of elements of a game system: game subject, game subject strategy set and lost revenue variable.

The game main body comprises a self vehicle and a first target vehicle; the game main body strategy set B comprises acceleration line grabbing or deceleration line giving; the lost revenue variables in the gaming body gaming process include a revenue variable and a loss variable. Specifically, the revenue variable includes a base loss G of the own vehicle_eRight of way and gain of right of way for rush to move R_e(ii) a Loss variables include back-off loss F_eCollision loss C_e(δ T), mutual yield loss M_eAnd a first safety distance loss S_fAnd a second safety distance loss S_a。

The gains and losses are defined as follows:

basic profit G_e: and the constant is the basic income in the game relation.

Road right rush income R_e: and a constant value is used for showing the influence of the right of way on the passing priority when the party with the right of way has the robbing intention.

Avoidance loss F_e: and a constant value, namely the loss of the self vehicle when the self vehicle selects deceleration for yielding and the first target vehicle selects acceleration for robbing.

Loss of collision C_e(δ T): the function is that when the self vehicle selects acceleration to rush to walk and the first target vehicle selects acceleration to rush to walk, the self vehicle is lost; the δ T is δ T in equation (4).

Mutual giveLoss M_e: and a constant value, when the speed reduction yielding is selected by the self vehicle and the speed reduction yielding is also selected by the first target vehicle, the loss of the self vehicle is caused.

First safety distance loss S_f(D_f,δv_f,B_e): and when the vehicle selects accelerating and snatching, the safety distance loss caused by the change of the space distance between the vehicle and the second target vehicle in the front direction is reduced. Exemplarily, the first safety distance loss S_fOne possible form of (2) is shown in equation (4):

wherein v is_egoSIs the longitudinal speed of the vehicle; v. of_frontSThe longitudinal speed of a second target vehicle in front of the self vehicle; delta v_fThe speed difference between the self vehicle and the second target vehicle is obtained; ttr (time To risk) represents a collision risk time, which is a time when a second target vehicle ahead of the host vehicle brakes at a deceleration of a (a is a positive number) and collides with the second target vehicle ahead of the host vehicle at a constant speed. D_fIs the longitudinal relative distance between the self-vehicle and a second target vehicle in front of the self-vehicle. S_fFor the first safety distance loss, k1, k2 and k3 are each parameters of the steepness of the regulating function. k is a radical of_BFor the preemption coefficient, the value range of offset is (0,1), gw (grabway) represents the preemption coefficient, and yd (yield) represents the yielding coefficient. For example, when the value of offset is 0.2, when the own vehicle policy B is satisfied_eWhen GW is k_BIs 1.2; when own vehicle strategy B_eWhen is TD, k_BIs 0.8.

It is understood that, in the case where it is determined that there are a plurality of second target vehicles ahead of the host vehicle, the first safe distance loss corresponding to the plurality of second target vehicles may be determined, respectively. In the process of establishing the dynamic game model, one first safety distance loss with the largest loss value is selected from the first safety distance losses corresponding to the second target vehicles to establish an income equation in the dynamic game model.

Second safety distance loss S_a(D_a,δv_a,B_e): and when the self vehicle selects deceleration and gives way, the safety distance loss caused by the change of the space distance between the self vehicle and the second target vehicle is generated at the rear part. Exemplarily, the second safety distance loss S_aOne possible form of (2) is shown in equation (5):

wherein v is_egoSIs the longitudinal speed of the vehicle; v. of_afterSIs the longitudinal speed of a second target vehicle behind the host vehicle; delta v_fIs the speed difference between the vehicle and the second target vehicle. TTR represents collision risk time, which is when a second target vehicle behind the own vehicle travels with acceleration of a (a is a positive number) and the own vehicle collides with the second target vehicle at a constant speed; d_aIs the longitudinal relative distance between the self-vehicle and a second target vehicle behind the self-vehicle. S_aFor the second safety distance loss, k1, k2 and k3 are parameters of the steepness of the regulating function, respectively. k is a radical of_BFor the preemption coefficient, the value range of offset is (0,1), GW represents the preemption coefficient, and YD represents the yielding coefficient.

Similarly, in a case where it is determined that there are a plurality of second target vehicles behind the own vehicle, the second safe distance loss corresponding to the plurality of second target vehicles may be determined, respectively. In the process of establishing the dynamic game model, one second safety distance loss with the largest loss value is selected from the second safety distance losses corresponding to the second target vehicles to establish an income equation in the dynamic game model.

And secondly, establishing an income equation after determining a plurality of elements of a game system.

The probability of accelerating and grabbing the own vehicle is assumed to be x, and the probability of decelerating and yielding the own vehicle is assumed to be (1-x); the probability of the first target vehicle accelerating and grabbing the line is y, and the probability of the first target vehicle decelerating and yielding the line is (1-y). According to the gains and losses defined in the foregoing, the pure strategy expected gains E for accelerating the robbery are selected by the self vehicle_e1And pure strategy expected yield E of selecting deceleration yield_e2As shown in equation (6) and equation (7), respectively:

E_e1＝G_e-C_ey+R_e-S_f+S_aformula (6)

E_e2＝G_e-M_e+(M_e-F_e)y+S_f-S_aFormula (7)

Therefore, according to the formula (6) and the formula (7), the expected profit of the hybrid strategy when the vehicle accelerates and snatchs with the probability of x and gives way with the probability of (1-x) can be deduced

The hybrid strategy expects a benefit

Can be shown as equation (8):

and thirdly, after the profit equation is established, establishing a dynamic replication equation.

The dynamic replication equation is an iterative equation of each frame of data, and is a differential delta x of the game object yield probability of the current frame of data, the self-vehicle yield probability and a yield equation, namely the variable quantity of x

Specifically, according to equations (6) and (8), a dynamic replication equation for obtaining the iteration probability of the own vehicle can be derived, and the dynamic replication equation can be specifically expressed as equation (9):

after the dynamic replication equation is obtained, the probability x of accelerated line grabbing of the own vehicle can be iterated according to the dynamic replication equation and the probability y of accelerated line grabbing of the first target vehicle predicted and obtained by the current frame, and finally the probability x is set through the set probability threshold value x_thresholdAnd a probability differential threshold valueδx_thresholdTo determine the driving strategy.

Illustratively, the probability threshold for speeding up the preemptive decision may be set to x_grabway1And x_grabway2The probability threshold of the deceleration yielding decision is x_yield1And x_yield2The probability differential threshold is δ x_thresholdWherein x is_yield2<x_yield1<x_grabway1<x_grabway2. When the probability x of accelerating the vehicle to rush the line is in (x)_yield1，x_grabway1) In between, continue to use the dynamic replication equation to carry on the iteration of x; when the probability x of accelerating the vehicle to rush the line is in (x)_yield2，x_yield1) And (x)_grabway1，x_grabway2) In time, the dynamic replication equation is judged

Whether or not it is less than the probability differential threshold value deltax_threshold. If dynamic replication equation

Less than a probability differential threshold δ x_thresholdThen it can be considered converged and a decision can be made that x is located at (x)_yield2，x_yield1) When the driving strategy is determined to be deceleration yielding, the x is positioned at (x)_grabway1，x_grabway2) Determining that the driving strategy is accelerated and robbed; if dynamic replication equation

Not less than a probability differential threshold value deltax_thresholdThen the iteration of x continues using the dynamic replication equation. When x is<x_yield2Or x>x_grabway2It can also be considered converged to make a decision, i.e. at x<x_yield2When determining the driving strategy as deceleration yielding, at x>x_grabway2And determining the driving strategy as accelerating and preempting.

504. And controlling the running of the vehicle according to the running strategy.

It is understood that, after the running strategy is determined, the running of the own vehicle may be controlled according to the running strategy. For example, when the driving strategy is acceleration preemption, the speed of the own vehicle is increased so as to control the own vehicle to pass through the track conflict point before the first target vehicle reaches the track conflict point. When the driving strategy is deceleration yielding, the speed of the self-vehicle is reduced, and the self-vehicle is controlled to decelerate yielding, so that the self-vehicle is controlled to reach the track conflict point after the first target vehicle point passes through the track conflict point.

In this embodiment, a dynamic game model that introduces collision loss and safety distance loss is established by acquiring information of a game target vehicle having a game relationship with a host vehicle and information of a risk target vehicle that affects decision making of the host vehicle. Because the risk influence of the driving environment on the game process is added into the dynamic game model, the risk decision and the game decision can be combined, the risk of the game process is effectively reduced, and the safety of automatic driving is improved.

The process of establishing the dynamic game model and determining the driving strategy based on the dynamic game model is described above, and the process of identifying the first target vehicle and the second target vehicle by the own vehicle will be described in detail below.

Referring to fig. 6, fig. 6 is a schematic flowchart of a method for identifying a target vehicle according to an embodiment of the present application. Referring to fig. 6, the process of identifying the first target vehicle and the second target vehicle from the vehicle may include the steps of:

601. and determining a first target vehicle set according to the position of the own vehicle, wherein the first target vehicle set comprises a plurality of target vehicles with the distance to the own vehicle being less than a preset threshold value.

Before the first information of the first target vehicle and the second information of the second target vehicle are obtained, the key obstacles corresponding to the vehicle can be identified according to the position and the planned track of the vehicle, and then the first target vehicle and the second target vehicle are determined in the key obstacles corresponding to the vehicle. By key obstacle identification, a large number of extraneous objects can be filtered out. The key obstacle is a movable target such as a vehicle, a pedestrian, a riding vehicle and the like which has high correlation with the own vehicle and may affect the driving behavior of the own vehicle. For convenience of description, the following description will be given taking an example in which the key obstacle is a vehicle.

In this embodiment, there may be a plurality of ways to determine the first target vehicle set according to the position of the own vehicle.

In one possible embodiment, the first set of target vehicles may be determined by euclidean distances between the own vehicle and other vehicles. For example, the euclidean distance between the host vehicle and the vehicles in the surrounding environment may be determined based on the location of the host vehicle and the locations of the vehicles in the surrounding environment of the host vehicle. When the Euclidean distance between the own vehicle and a certain vehicle is smaller than a preset Euclidean distance threshold value, the vehicle can be determined to be a vehicle in the first target vehicle set. Specifically, it can be determined whether the euclidean distance between the own vehicle and a certain vehicle is less than a preset euclidean distance threshold value by the formula (10).

Wherein, X_eIndicating the abscissa, Y, of the vehicle_eIndicating the ordinate, X, of the vehicle_oShowing the abscissa, Y, of the target vehicle_oRepresenting the ordinate, D, of the target vehicle_thresholdIs the euclidean distance threshold.

In one possible embodiment, the first set of target vehicles may be determined by taking the planned trajectory of the host vehicle as a central axis, forming a range box, such as a bendable rectangular box, around the periphery of the host vehicle, and determining whether the target vehicle is within the range box by ray method. Specifically, the manner of determining whether the target vehicle is within the range box by the ray method may be as shown in equation (11).

Wherein, Y_cA vertical coordinate representing the intersection point of the ray and the range frame is vertically and upwards taken by taking the target vehicle as a starting point; x_(P,i)The abscissa value of the ith point of the demonstration enclosure is shown; x_(P,i+1)Table demonstrates the abscissa of the (i + 1) th point of the bounding boxA value; y is_(P,i)The ordinate value of the ith point of the demonstration enclosure is shown; y is_(P,i+1)The longitudinal coordinate value of the (i + 1) th point of the demonstration enclosure is shown; s_pTable demonstrates the number of points contained in the bounding box; i denotes a number for traversing the range box; count is the number of intersections. In equation (11), whether the target vehicle is within the range box may be determined based on the parity of the count. If count is odd, then it may be determined that the target vehicle is within the range box; if count is even, then the target vehicle may be determined to be outside the range box.

602. And determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which the predicted tracks are intersected with the planned track of the self vehicle.

After the first set of target vehicles is determined, target vehicles in the first set of target vehicles, of which predicted trajectories intersect with the planned trajectory of the own vehicle, can be further determined, so that a second set of target vehicles is determined. For example, for each target vehicle in the first set of target vehicles, it may be determined whether the predicted trajectory of the target vehicle intersects the planned trajectory of the own vehicle based on equation (12).

Wherein S is_EPIs the number of points contained in the planned trajectory of the own vehicle, S_OPIs the number of points included in the predicted trajectory of the target vehicle, i and j are coefficients of the traversal trajectory, X_(EP，i)Abscissa, Y, representing the i-th point of the vehicle_(EP，i)Ordinate, X, representing i-th point of own vehicle_(OP，i)Abscissa, Y, representing the i-th point of the target vehicle_(OP，i)The ordinate of the i-th point of the target vehicle is represented. r and s are coefficients of the intersection of two straight lines defined by two points, which are projected on a line segment formed by the two points. When r is more than or equal to 0 and less than or equal to 1 and s is more than or equal to 0 and less than or equal to 1, the planned trajectory of the own vehicle can be determined to be intersected with the predicted trajectory of the target vehicle.

603. A first target vehicle and a second target vehicle are determined in a second set of target vehicles.

After the second target vehicle set is determined, a first target vehicle having a direct game relationship with the own vehicle and a second target vehicle influencing the game process of the own vehicle can be determined in the second target vehicle set.

In one possible implementation, the first target vehicle in the second set of target vehicles may be determined based on a difference between the TTCs of the own vehicle and the target vehicle and a lane in which the target vehicle is located.

For example, the TTC corresponding to the vehicle and the TTC corresponding to the third target vehicle, which is the target vehicle in the second set of target vehicles, may be obtained first. The manner of obtaining the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle may be as described in the above embodiments, and details thereof are not repeated herein. Then, a difference between the TTC corresponding to the own vehicle and the TTC corresponding to the third target vehicle may be determined. When the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle is smaller than or equal to a preset threshold value and the third target vehicle is in the game vehicle lane set, determining that the third target vehicle is the first target vehicle; the game lane set is a lane set which has a game relation with a lane where the self vehicle is located. That is to say, when the time difference between the self vehicle and the third target vehicle reaching the track conflict point is small and the third target vehicle is located in the game lane set, it can be determined that the third target vehicle is the first target vehicle.

For ease of understanding, the lane referred to in the present embodiment will be described in detail below. Referring to fig. 7, fig. 7 is a schematic view illustrating a lane dividing method according to an embodiment of the present disclosure. As shown in fig. 7, the lanes defined in the present embodiment have a certain length, and there is a connection relationship between the lanes. In fig. 7, a1, a2, a3, b1, b2, b3, c1, c2, and c3 are all independent lanes, and a set of lanes may be referred to as a lane set.

When the self-vehicle is in different scenes, the self-vehicle can determine a corresponding game vehicle road set according to the current scene. For example, referring to fig. 8, fig. 8 is a schematic diagram of a gaming lane dividing method provided in an embodiment of the present application. As shown in fig. 8, when the own vehicle is in a driving scene at an intersection without left turn protection, the own vehicle can determine that a lane forming a game relationship with the lane where the own vehicle is located is a game lane. In fig. 8, lane f1 and lane f3 are both gaming lanes, i.e., the host vehicle may determine that the set of gaming lanes includes lane f1 and lane f 3.

Referring to fig. 9, fig. 9 is a schematic diagram of another gaming lane dividing method provided in the embodiment of the present application. As shown in fig. 9, when the own vehicle is in the on-off ramp driving scene, the own vehicle can also determine that the lane forming the game relationship with the lane where the own vehicle is located is the game lane. In fig. 9, the set of gaming lanes includes lane h 1.

In one possible implementation, a second target vehicle in the second set of target vehicles may be determined based on the lane in which the target vehicle is located.

For example, if the third target vehicle is in the risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, the third target vehicle is determined to be the second target vehicle. And the third target vehicle is the target vehicle in the second target vehicle set. The lane set of the vehicle may refer to a lane set within a certain longitudinal range of a lane where the vehicle is located, and the lane set of the vehicle may include, for example, a lane of the vehicle, a lane in front of the lane of the vehicle, and a lane behind the lane of the vehicle, where the lane of the vehicle is the lane where the vehicle is located. The risk lane set can refer to a lane where a vehicle which can affect the vehicle is located in the game process of the vehicle. The set of at-risk lanes may include, for example, a set of vehicle lanes and a set of left lanes of the set of vehicle lanes and/or a set of right lanes of the set of vehicle lanes.

Illustratively, as shown in fig. 8, in a left-turn-free intersection driving scenario, the set of vehicle lanes includes lane d1, lane d2, and lane d3, the set of risk lanes includes lane d1, lane d2, lane d3, lane e1, lane e2, and lane e3, and the set of trajectory lanes includes lane f3 and lane f 4.

As shown in fig. 9, in the on-off ramp driving scenario, the set of cars includes lane g1 and lane g2, the set of risk lanes includes lane g1, lane g2, lane i1 and lane i2, and the set of trajectory lanes includes lane h1 and lane g 2.

In one possible embodiment, after identifying the first target vehicle and the second target vehicle, if there is a spatial separation between the first target vehicle or the second target vehicle and the host vehicle, the identified first target vehicle and the second target vehicle may be considered to have no game-related relationship in nature. Thus, the portion of the identified target vehicle may be filtered out.

For example, for the identified first target vehicle, the predicted trajectory of the first target vehicle may be widened to a bendable rectangular frame; for the second target vehicle, the connecting line between the second target vehicle and the vehicle can be widened to obtain the bendable rectangular frame. After the bendable rectangular frame is obtained, whether the first target vehicle or the second target vehicle is spaced by other target vehicles can be determined by judging whether the target vehicles are covered by the bendable rectangular frame. Specifically, it may be determined whether the first target vehicle or the second target vehicle is spaced by the other target vehicle by traversing whether the other target vehicle is within the range of the bendable rectangular frame through equation (9).

Referring to fig. 10, fig. 10 is a schematic diagram illustrating a first target vehicle and a second target vehicle filtered according to an embodiment of the present disclosure. In fig. 10, the target vehicle 1 and the target vehicle 2 may be recognized as a first target vehicle, and the target vehicle 3 and the target vehicle 4 may be recognized as a second target vehicle. After the predicted trajectory of the target vehicle 2 is widened and the rectangular frame is obtained, it can be determined that the target vehicle 1 is covered by the rectangular frame. Therefore, it can be determined that the target vehicles 2 are spaced by the target vehicles 1, so that the target vehicles 2 can be filtered out. Further, by widening the connection line between the own vehicle and the target vehicle 4 and obtaining a rectangular frame, it can be determined that the target vehicle 3 is covered by the rectangular frame. Therefore, it can be determined that the target vehicles 4 are spaced by the target vehicles 3, so that the target vehicles 4 can be filtered out.

On the basis of the embodiments corresponding to fig. 5 to 10, in order to better implement the above-mentioned scheme of the embodiments of the present application, the following also provides related equipment for implementing the above-mentioned scheme. Referring to fig. 11 in particular, fig. 11 is a schematic structural diagram of an autopilot device 1100 according to an embodiment of the present disclosure. The autopilot device 1100 may include an acquisition unit 1101, a processing unit 1102, and a control unit 1103, and the autopilot device 1100 may further include a recognition unit 1104.

In one possible implementation, may include: the acquiring unit 1101 is configured to acquire first information of a first target vehicle and second information of a second target vehicle, where the own vehicle has a collision risk with the first target vehicle and the second target vehicle, the first information includes position information and speed information of the first target vehicle, and the second information includes position information and speed information of the second target vehicle;

the processing unit 1102 is configured to establish a dynamic game model according to at least first information and second information, where the first information is used to determine collision loss in the dynamic game model, and the second information is used to determine safety distance loss in the dynamic game model;

the processing unit 1102 is further configured to determine a driving strategy according to the dynamic game model;

and a control unit 1103 for controlling the running of the own vehicle according to the running strategy.

In one possible embodiment, the apparatus further comprises: the identifying unit 1104 is configured to identify that the vehicle is in a preset scene, where the preset scene includes one of a driving scene of a junction without left turn protection, a driving scene of an entrance ramp or a driving scene of a lane combination.

In a possible implementation, the processing unit 1102 is specifically configured to: obtaining time TTC from a corresponding track conflict point of the vehicle; determining a TTC corresponding to the first target vehicle according to the first information; and determining the collision loss in the dynamic game model according to the difference value between the TTC corresponding to the own vehicle and the TTC corresponding to the first target vehicle.

In a possible implementation, the processing unit 1102 is specifically configured to: determining first safety distance loss or second safety distance loss in the dynamic game model according to the position and the speed of the self-vehicle and the second information; the first safety distance loss is the front safety distance loss when the acceleration and preemption strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration and yielding strategy is executed.

In one possible implementation, the processing unit 1102 is further configured to: determining that the vehicle has the right of way according to the vehicle and the first target vehicle; and determining road right bank-robbing income according to the fact that the bicycle has the road right, wherein the road right bank-robbing income is used for building a dynamic game model.

In one possible implementation, the processing unit 1102 is further configured to: determining a first target vehicle set according to the position of the vehicle, wherein the first target vehicle set comprises a plurality of target vehicles, and the distance between each target vehicle and the vehicle is smaller than a preset threshold value; determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which predicted tracks are intersected with the planned track of the self vehicle; a first target vehicle and a second target vehicle are determined in a second set of target vehicles.

In a possible embodiment, the obtaining unit 1101 is further configured to obtain a TTC corresponding to the own vehicle and a TTC corresponding to a third target vehicle, where the third target vehicle is a target vehicle in the second set of target vehicles; the processing unit 1102 is further configured to determine a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle; the processing unit 1102 is further configured to determine that the third target vehicle is the first target vehicle if a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle is less than or equal to a preset threshold and the third target vehicle is in the game lane set; the game lane set is a lane set which has a game relation with a lane where the self vehicle is located.

In a possible implementation, the processing unit 1102 is specifically configured to: if the third target vehicle is in the risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, determining that the third target vehicle is the second target vehicle; the third target vehicle is a target vehicle in the second target vehicle set, the vehicle lane set comprises a self lane, a front lane of the self lane and a rear lane of the self lane, the self lane is a lane where the self vehicle is located, and the risk lane set comprises the vehicle lane set and a left lane set of the vehicle lane set and/or a right lane set of the vehicle lane set.

It should be noted that, the information interaction, the execution process, and the like between the units in the automatic driving device 1100 are based on the same concept as the method embodiments corresponding to fig. 5 to 10 in the present application, and specific contents may refer to the description in the foregoing method embodiments in the present application, and are not described again here.

Fig. 12 is a schematic structural diagram of the autonomous vehicle provided in the embodiment of the present application, where fig. 12 is a schematic structural diagram of the autonomous vehicle provided in the embodiment of the present application, and an autonomous device described in the embodiment corresponding to fig. 10 may be disposed on the autonomous vehicle 100 to implement functions of the autonomous vehicle in the embodiments corresponding to fig. 5 to fig. 10. Since in some embodiments the autonomous vehicle 100 may also include communication functionality, the autonomous vehicle 100 may include, in addition to the components shown in fig. 1: a receiver 1201 and a transmitter 1202, wherein the processor 113 may include an application processor 1131 and a communication processor 1132. In some embodiments of the present application, the receiver 1201, the transmitter 1202, the processor 113, and the memory 114 may be connected by a bus or other means.

The processor 113 controls operation of the autonomous vehicle. In a particular application, the various components of the autonomous vehicle 100 are coupled together by a bus system that may include a power bus, a control bus, a status signal bus, etc., in addition to a data bus. For clarity of illustration, the various buses are referred to in the figures as a bus system.

Receiver 1201 may be used to receive input numeric or character information and to generate signal inputs related to settings and function controls associated with the autonomous vehicle. The transmitter 1202 may be configured to output numeric or character information via the first interface; the transmitter 1202 is also operable to send instructions to the disk group via the first interface to modify data in the disk group; the transmitter 1202 may also include a display device such as a display screen.

In the embodiment of the present application, the processor 1131 is configured to execute an automatic driving method executed by the automatic driving vehicle in the embodiment corresponding to fig. 5. Specifically, the application processor 1131 is configured to perform the following steps:

acquiring first information of a first target vehicle and second information of a second target vehicle, wherein the self vehicle, the first target vehicle and the second target vehicle have collision risks, the first information comprises position information and speed information of the first target vehicle, and the second information comprises position information and speed information of the second target vehicle; establishing a dynamic game model at least according to first information and second information, wherein the first information is used for determining the collision loss in the dynamic game model, and the second information is used for determining the safety distance loss in the dynamic game model; determining a driving strategy according to the dynamic game model; and controlling the running of the vehicle according to the running strategy.

In one possible implementation, it is recognized that the vehicle is in a preset scene, where the preset scene includes one of a driving scene of a junction without left turn protection, a driving scene of an entrance ramp or a driving scene of a lane combination.

In one possible implementation, the time TTC from the trajectory conflict point corresponding to the own vehicle is obtained; determining a TTC corresponding to the first target vehicle according to the first information; and determining the collision loss in the dynamic game model according to the difference value between the TTC corresponding to the own vehicle and the TTC corresponding to the first target vehicle.

In one possible implementation mode, determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and the speed of the self vehicle and the second information; the first safety distance loss is the front safety distance loss when the acceleration and preemption strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration and yielding strategy is executed.

In one possible embodiment, it is determined that the own vehicle has the right of way based on the own vehicle and the first target vehicle; and determining road right bank-robbing income according to the fact that the bicycle has the road right, wherein the road right bank-robbing income is used for building a dynamic game model.

In one possible embodiment, a first target vehicle set is determined according to the position of the own vehicle, and the first target vehicle set comprises a plurality of target vehicles with the distance to the own vehicle being less than a preset threshold value; determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which predicted tracks are intersected with the planned track of the self vehicle; a first target vehicle and a second target vehicle are determined in a second set of target vehicles.

In one possible implementation, the TTC corresponding to the own vehicle and the TTC corresponding to a third target vehicle are obtained, and the third target vehicle is a target vehicle in the second set of target vehicles; determining a difference value between the TTC corresponding to the vehicle and the TTC corresponding to the third target vehicle; if the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle is smaller than or equal to a preset threshold value, and the third target vehicle is in a game vehicle lane set, determining that the third target vehicle is the first target vehicle; the game lane set is a lane set which has a game relation with a lane where the self vehicle is located.

In one possible implementation, if the third target vehicle is in the risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, determining that the third target vehicle is the second target vehicle; the third target vehicle is a target vehicle in the second target vehicle set, the vehicle lane set comprises a self lane, a front lane of the self lane and a rear lane of the self lane, the self lane is a lane where the self vehicle is located, and the risk lane set comprises the vehicle lane set and a left lane set of the vehicle lane set and/or a right lane set of the vehicle lane set.

It should be noted that, for specific implementation manners and advantageous effects of the automatic driving method executed by the application processor 1131, reference may be made to descriptions in each method embodiment corresponding to fig. 5 to fig. 10, and details are not repeated here.

Also provided in an embodiment of the present application is a computer-readable storage medium having stored therein a program for performing autonomous driving, which when executed on a computer causes the computer to perform the steps performed by an autonomous vehicle (or autonomous driving apparatus) in the method described in the embodiments of fig. 5 to 10 described above.

Embodiments of the present application also provide a computer program product, which when executed on a computer causes the computer to perform the steps performed by the autonomous vehicle in the methods described in the embodiments of fig. 5 to 10.

Further provided in embodiments of the present application is a circuit system including a processing circuit configured to perform the steps performed by the autonomous vehicle in the method described in the embodiments of fig. 5-10 above.

The autopilot device or the autopilot vehicle provided by the embodiment of the application can be specifically a chip, and the chip comprises: a processing unit, which may be, for example, a processor, and a communication unit, which may be, for example, an input/output interface, a pin or a circuit, etc. The processing unit may execute computer-executable instructions stored by the storage unit to cause a chip within the server to perform the autopilot method described in the embodiments illustrated in fig. 5-10 above. Alternatively, the storage unit may be a storage unit in the chip, such as a register, a cache, and the like, and the storage unit may also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (ROM) or another type of static storage device that can store static information and instructions, a Random Access Memory (RAM), and the like.

Specifically, referring to fig. 13, fig. 13 is a schematic structural diagram of a chip provided in the embodiment of the present application, where the chip may be represented as a neural network processor NPU 130, and the NPU 130 is mounted on a main CPU (Host CPU) as a coprocessor, and the Host CPU allocates tasks. The core portion of the NPU is an arithmetic circuit 1303, and the arithmetic circuit 1303 is controlled by a controller 1304 to extract matrix data in a memory and perform multiplication.

In some implementations, the arithmetic circuit 1303 includes a plurality of processing units (PEs) therein. In some implementations, the operational circuit 1303 is a two-dimensional systolic array. The arithmetic circuit 1303 may also be a one-dimensional systolic array or other electronic circuit capable of performing mathematical operations such as multiplication and addition. In some implementations, the arithmetic circuitry 1303 is a general-purpose matrix processor.

For example, assume that there is an input matrix A, a weight matrix B, and an output matrix C. The arithmetic circuit fetches the data corresponding to the matrix B from the weight memory 1302 and buffers the data on each PE in the arithmetic circuit. The arithmetic circuit takes the matrix a data from the input memory 1301 and performs matrix operation with the matrix B, and a partial result or a final result of the obtained matrix is stored in an accumulator (accumulator) 1308.

The unified memory 1306 is used to store input data as well as output data. The weight data directly passes through a Direct Memory Access Controller (DMAC) 1305, and the DMAC is transferred to the weight memory 1302. The input data is also carried into the unified memory 1306 through the DMAC.

A Bus Interface Unit (BIU) 1310 for interaction of the AXI bus with the DMAC and the instruction fetch memory (IFB) 1309.

BIU1310 is used for instruction fetch 1309 to fetch instructions from external memory, and is also used for memory access controller 1305 to fetch the original data of input matrix a or weight matrix B from external memory.

The DMAC is mainly used to transfer input data in the external memory DDR to the unified memory 1306 or to transfer weight data into the weight memory 1302 or to transfer input data into the input memory 1301.

The vector calculation unit 1307 includes a plurality of operation processing units, and further processes such as vector multiplication, vector addition, exponential operation, logarithmic operation, magnitude comparison, and the like are performed on the outputs of the operation circuits, if necessary. The method is mainly used for non-convolution/full-connection layer network calculation in the neural network, such as batch normalization (batch normalization), pixel-level summation, up-sampling of a feature plane and the like.

In some implementations, vector calculation unit 1307 can store the processed output vector to unified memory 1306. For example, the vector calculation unit 1307 may apply a linear function and/or a nonlinear function to the output of the arithmetic circuit 1303, such as linear interpolation of the feature planes extracted by the convolution layer, and further such as a vector of accumulated values to generate an activation value. In some implementations, the vector calculation unit 1307 generates normalized values, pixel-level summed values, or both. In some implementations, the vector of processed outputs can be used as activation inputs to the arithmetic circuitry 1303, e.g., for use in subsequent layers in a neural network.

An instruction fetch buffer (issue fetch buffer)1309 is connected to the controller 1304 and is used to store instructions used by the controller 1304.

The unified memory 1306, input memory 1301, weight memory 1302 and instruction fetch memory 1309 are all On-Chip memories. The external memory is private to the NPU hardware architecture.

Here, the operation of each layer in the recurrent neural network may be performed by the operation circuit 1303 or the vector calculation unit 1307.

Wherein any of the aforementioned processors may be a general purpose central processing unit, a microprocessor, an ASIC, or one or more integrated circuits configured to control the execution of the programs of the method of the first aspect.

It should be noted that the above-described embodiments of the apparatus are merely schematic, where units illustrated as separate components may or may not be physically separate, and components illustrated as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiments of the apparatus provided in the present application, the connection relationship between the modules indicates that there is a communication connection therebetween, and may be implemented as one or more communication buses or signal lines.

Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by software plus necessary general hardware, and certainly can also be implemented by special hardware including application specific integrated circuits, special CLUs, special memories, special components and the like. Generally, functions performed by computer programs can be easily implemented by corresponding hardware, and specific hardware structures for implementing the same functions may be various, such as analog circuits, digital circuits, or dedicated circuits. However, for the present application, the implementation of a software program is more preferable. Based on such understanding, the technical solutions of the present application may be substantially embodied in the form of a software product, which is stored in a readable storage medium, such as a floppy disk, a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the method of the embodiments of the present application.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product.

The computer program product includes one or more computer instructions. The procedures or functions according to the embodiments of the present application are all or partially generated when the computer program instructions are loaded and executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., coaxial cable, fiber optic, digital subscriber line) or wirelessly (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium may be any available medium that a computer can store or a data storage device, such as a server, a data center, etc., that is integrated with one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

Claims

1. An automatic driving method, characterized by comprising:

acquiring first information of a first target vehicle and second information of a second target vehicle, wherein the self vehicle has collision risks with the first target vehicle and the second target vehicle, the first information comprises position information and speed information of the first target vehicle, and the second information comprises position information and speed information of the second target vehicle;

establishing a dynamic game model at least according to the first information and the second information, wherein the first information is used for determining collision loss in the dynamic game model, the second information is used for determining safety distance loss in the dynamic game model, the collision loss is the loss of the self vehicle when the self vehicle and the first target vehicle select accelerating and grabbing, and the safety distance loss is the loss of the self vehicle when the safety distance between the self vehicle and the second target vehicle changes;

determining a driving strategy according to the dynamic game model;

and controlling the running of the self vehicle according to the running strategy.

2. The automated driving method of claim 1, further comprising:

recognizing that the self-vehicle is in a preset scene, wherein the preset scene comprises: the method comprises the following steps of protecting a crossing driving scene, an entrance ramp driving scene and a lane merging driving scene without left turn.

3. The automated driving method according to claim 1 or 2, wherein the own vehicle can eliminate a risk of collision with the first target vehicle by accelerating for overtaking or decelerating for passing.

4. The autopilot method of any of claims 1 to 3 wherein the establishing a dynamic gaming model based on at least the first information and the second information comprises:

acquiring time TTC from a track conflict point corresponding to the self-vehicle;

determining a TTC corresponding to the first target vehicle according to the first information;

and determining the collision loss in the dynamic game model according to the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the first target vehicle.

5. The autopilot method of any of claims 1 to 4 wherein the establishing a dynamic gaming model based on at least the first information and the second information comprises:

determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and the speed of the self-vehicle and the second information;

the first safety distance loss is the front safety distance loss when the acceleration and preemption strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration and yielding strategy is executed.

6. The automated driving method according to any one of claims 1 to 5, characterized in that the method further comprises:

determining that the vehicle has the right of way according to the vehicle and the first target vehicle;

and determining road right rush-run income in the dynamic game model according to the fact that the self vehicle has the road right.

7. The automated driving method according to any one of claims 1 to 6, wherein before the acquiring the first information of the first target vehicle and the second information of the second target vehicle, the method further comprises:

determining a first target vehicle set according to the position of the self vehicle, wherein the first target vehicle set comprises a plurality of target vehicles, and the distance between the target vehicles and the self vehicle is less than a preset threshold value;

determining a second target vehicle set according to the planned track of the self vehicle and the first target vehicle set, wherein the second target vehicle set comprises a plurality of target vehicles of which predicted tracks are intersected with the planned track of the self vehicle;

determining the first target vehicle and the second target vehicle in the second set of target vehicles.

8. The autonomous driving method of claim 7, wherein the determining the first target vehicle and the second target vehicle in the second set of target vehicles comprises:

acquiring a TTC corresponding to the self vehicle and a TTC corresponding to a third target vehicle, wherein the third target vehicle is a target vehicle in the second target vehicle set;

determining a difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle;

if the difference value between the TTC corresponding to the self vehicle and the TTC corresponding to the third target vehicle is smaller than or equal to a preset threshold value, and the third target vehicle is in a game vehicle lane set, determining that the third target vehicle is the first target vehicle;

the game lane set is a lane set which has a game relation with a lane where the self vehicle is located.

9. The autonomous driving method of claim 7 or 8, wherein the determining the first target vehicle and the second target vehicle in the second set of target vehicles comprises:

if a third target vehicle is in a risk lane set and the same lane exists in the track lane set of the third target vehicle and the own vehicle lane set, determining that the third target vehicle is a second target vehicle;

wherein, the third target vehicle does target vehicle in the second target vehicle set, the car lane set includes the own lane the place ahead lane of own lane with the rear lane from the lane, the own lane does the lane that the own vehicle is located, the risk lane set includes the car lane set and the left lane set of the car lane set and/or the right lane set of the car lane set.

10. An autopilot device, comprising:

the system comprises an acquisition unit, a processing unit and a control unit, wherein the acquisition unit is used for acquiring first information of a first target vehicle and second information of a second target vehicle, the self vehicle, the first target vehicle and the second target vehicle have collision risks, the first information comprises position information and speed information of the first target vehicle, and the second information comprises position information and speed information of the second target vehicle;

the processing unit is used for establishing a dynamic game model at least according to the first information and the second information, wherein the first information is used for determining collision loss in the dynamic game model, the second information is used for determining safety distance loss in the dynamic game model, the collision loss is the loss of the self vehicle when the self vehicle and the first target vehicle select acceleration and preemption, and the safety distance loss is the loss of the self vehicle when the safety distance between the self vehicle and the second target vehicle changes;

the processing unit is further used for determining a driving strategy according to the dynamic game model;

and the control unit is used for controlling the running of the vehicle according to the running strategy.

11. The autopilot device of claim 10 wherein the device further comprises:

the recognition unit is used for recognizing that the self-vehicle is in a preset scene, and the preset scene comprises: the method comprises the following steps of protecting a crossing driving scene, an entrance ramp driving scene and a lane merging driving scene without left turn.

12. The autopilot device of claim 10 or 11 wherein the host vehicle can eliminate the risk of collision with the first target vehicle by speeding up or slowing down the ride.

13. The autopilot device of any one of claims 10 to 12 wherein the processing unit is specifically configured to:

14. The autopilot device of any one of claims 10 to 13 wherein the processing unit is specifically configured to:

15. The autopilot device of any one of claims 10 to 14 wherein the processing unit is further configured to:

and determining road right bank-robbing income according to the fact that the vehicle has the road right, wherein the road right bank-robbing income is used for establishing the dynamic game model.

16. The autopilot device of any one of claims 10 to 15 wherein the processing unit is further configured to:

17. The autopilot device of claim 16 wherein the obtaining unit is further configured to obtain a TTC corresponding to the host vehicle and a TTC corresponding to a third target vehicle, the third target vehicle being a target vehicle of the second set of target vehicles;

the processing unit is further configured to determine a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle;

the processing unit is further configured to determine that the third target vehicle is the first target vehicle if a difference between the TTC corresponding to the host vehicle and the TTC corresponding to the third target vehicle is less than or equal to a preset threshold and the third target vehicle is in a game vehicle lane set;

18. The autopilot device of claim 16 or 17, wherein the processing unit is specifically configured to:

19. An autopilot device comprising a processor coupled to a memory, the memory storing program instructions that when executed by the processor implement the method of any one of claims 1 to 9.

20. A computer-readable storage medium comprising a program which, when run on a computer, causes the computer to perform the method of any one of claims 1 to 9.

21. An autonomous vehicle, characterized in that the smart car comprises a processing circuit and a storage circuit, the processing circuit and the storage circuit being configured to perform the method of any of claims 1 to 9.