Automatic driving method and related device
By establishing a dynamic game model, combining conflict loss and safe distance loss, autonomous driving vehicles optimize driving strategies in trajectory conflict scenarios, solving the problems of low traffic efficiency and decision-making risks in the existing technology, and improving the safety and efficiency of autonomous driving .
Patent Information
- Application Number
- CN202411045633.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2020-09-16
- Publication Date
- 2024-12-06
AI Technical Summary
The adoption of conservative strategies by autonomous vehicles in trajectory conflict scenarios leads to inefficient traffic efficiency and decision-making risks. The existing game strategies ignore risks outside the driving environment, which may lead to the danger of driving strategies.
By obtaining information about the target car with a game relationship with the bicycle, a dynamic game model that introduces conflict loss and safe distance loss is established, and combining risk decisions and game decisions, a driving strategy for accelerating or slowing down and giving way is determined.
It effectively reduces the risk of the game process, improves the safety and traffic efficiency of autonomous driving, and is suitable for special scenarios such as crossings and entrances and exit ramps without left turn.
Smart Images

Figure CN119078872A_ABST
Abstract
Description
[0001] This application is a divisional application. The application number of the original application is 202010975573.1, and the original application date is September 16, 2020. The entire contents of the original application are incorporated into this application by reference. Technical Field
[0002] The present application relates to the field of autonomous driving technology, and in particular to an autonomous driving method and related devices. Background Art
[0003] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use 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 respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that 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-making and reasoning, human-computer interaction, recommendation and search, basic AI theory, etc.
[0004] Autonomous driving is a mainstream application in the field of artificial intelligence. Autonomous driving technology relies on the collaboration of computer vision, radar, monitoring devices, and global positioning systems to enable motor vehicles to achieve autonomous driving without the need for active human operation. Autonomous vehicles use various computing systems to help transport passengers from one location to another. Some autonomous vehicles may require some initial or continuous input from operators (such as navigators, drivers, or passengers). Autonomous vehicles allow operators to switch from manual mode to autonomous driving mode or a mode in between. Since autonomous driving technology does not require humans to drive motor vehicles, it can theoretically effectively avoid human driving errors, reduce the occurrence of traffic accidents, and improve highway transportation efficiency. Therefore, autonomous driving technology is receiving more and more attention.
[0005] In the field of autonomous driving, autonomous vehicles can execute corresponding driving strategies according to actual driving scenarios to ensure safe driving of autonomous vehicles. However, in some special scenarios, such as intersections without left-turn protection or entrance and exit ramps, sometimes autonomous vehicles may conflict with other vehicles.
[0006] At present, when there is a trajectory conflict between an autonomous driving vehicle and a target vehicle, the autonomous driving vehicle often adopts a one-on-one game strategy with the target vehicle, ignoring the driving risks outside the game process, which can easily lead to certain dangers in the driving strategy made by the autonomous driving vehicle. Summary of the invention
[0007] The present application provides an automatic driving method and related devices, which establish a dynamic game model that introduces conflict loss and safety distance loss by acquiring information about a game target vehicle that has a game relationship with the self-vehicle and information about a risk target vehicle that affects the decision of the self-vehicle. Since the risk impact of the driving environment on the game process is added to the dynamic game model, risk decision-making and game decision-making can be combined, effectively reducing the risk of the game process and improving the safety of automatic driving.
[0008] The first aspect of the present application provides an automatic driving method, which can be used in the field of automatic driving in the field of artificial intelligence. The method may include: obtaining first information of a first target vehicle and second information of a second target vehicle, the first target vehicle and the second target vehicle may be vehicles in the surrounding environment of the vehicle, the vehicle has a collision risk with the first target vehicle and the second target vehicle, and the vehicle can eliminate the collision risk with the first target vehicle by accelerating to grab or decelerating to give way. In short, the vehicle has a direct game relationship with the first target vehicle, and during the game process, the driving strategy that the vehicle and the first target vehicle can make is to accelerate to grab or decelerate to give way. There is no direct game relationship between the second target vehicle and the vehicle, but the second target vehicle will affect and limit the decision-making and action range of the vehicle during the game process. The first information may include the position information and speed information of the first target vehicle, and the second information may include the position information and speed information of the second target vehicle. At least based on the first information and the second information, a dynamic game model is established, the first information is used to determine the conflict loss in the dynamic game model, and the second information is used to determine the safety distance loss in the dynamic game model. The conflict loss indicates the loss of the ego vehicle when the ego vehicle chooses to accelerate and overtake and the first target vehicle also chooses to accelerate and overtake; the safety distance loss indicates the loss of the ego vehicle when the ego vehicle executes the strategy of accelerating and overtaking or the strategy of slowing down and yielding, which causes the spatial distance between the ego vehicle and the second target vehicle in front or behind to change. The loss of the ego vehicle refers to the measurement of the possible changes in one or both of the road passability or safety of the ego vehicle caused by the choice of the game strategy of the ego vehicle during the game process. The driving strategy is determined according to the dynamic game model, and the driving strategy is accelerating and overtaking or slowing down and yielding; the driving of the ego vehicle is controlled according to the driving strategy.
[0009] In this scheme, by obtaining the information of the game target vehicle that has a game relationship with the ego vehicle and the information of the risk target vehicle that affects the decision of the ego vehicle, a dynamic game model that introduces conflict loss and safety distance loss is established. Since the risk impact of the driving environment on the game process is added to the dynamic game model, risk decision-making and game decision-making can be combined, effectively reducing the risk of the game process and improving the safety of autonomous driving.
[0010] In a possible implementation, the autonomous driving method further includes: recognizing that the vehicle is in a preset scenario, wherein the preset scenario includes: a driving scenario at an intersection without left-turn protection, an entrance and exit ramp driving scenario, and a lane merging driving scenario. In this solution, by selecting to automatically execute the autonomous driving method of this application when recognizing that the vehicle is in a preset scenario, the decision accuracy in special scenarios can be improved, and the safety of autonomous driving can be improved.
[0011] In a possible implementation, the establishment of a dynamic game model based at least on the first information and the second information may specifically include: obtaining the time (Time To Conflict, TTC) corresponding to the ego vehicle from the trajectory conflict point; determining the TTC corresponding to the first target vehicle based on the first information; and determining the conflict loss in the dynamic game model based on the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle. For the ego vehicle, the distance (Distance To Conflict, DTC) corresponding to the trajectory conflict point of the ego vehicle can be determined based on the current position of the ego vehicle and the position of the trajectory conflict point. After determining the DTC corresponding to the ego vehicle, the TTC corresponding to the ego vehicle can be determined based on the current speed and acceleration of the ego vehicle. Similarly, the DTC of the first target vehicle can also be determined based on the current position of the first target vehicle and the position of the trajectory conflict point, and then based on the DTC of the first target vehicle, the TTC corresponding to the first target vehicle is determined.
[0012] In this scheme, by determining the conflict loss in the dynamic game model based on the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle, the conflict loss that varies with the TTC difference can be introduced in the game process to improve the decision-making accuracy of the dynamic game model.
[0013] In a possible implementation, the establishment of the dynamic game model at least according to the first information and the second information may specifically include: determining the first safety distance loss or the second safety distance loss in the dynamic game model according to the position and speed of the self-vehicle and the second information; wherein the first safety distance loss is the front safety distance loss when the acceleration grab strategy is executed, and the second safety distance loss is the rear safety distance loss when the deceleration yield strategy is executed. For example, when the second target vehicle is in front of the self-vehicle, the first safety distance loss corresponding to the second target vehicle can be determined; when the second target vehicle is behind the self-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 can be determined based on the spatial distance between the second target vehicle and the self-vehicle, and the relative speed between the second target vehicle and the self-vehicle. For example, the first safety distance loss or the second safety distance loss can be obtained by dividing the spatial distance between the second target vehicle and the self-vehicle by the relative speed between the second target vehicle and the self-vehicle. Generally speaking, the greater the spatial distance between the second target car and the ego car, the greater the first safety distance loss or the second safety distance loss; the greater the relative speed between the second target car and the ego car, the smaller the first safety distance loss or the second safety distance loss. In this scheme, by introducing the safety distance loss in the dynamic game model, the game decision and risk decision can be coupled to improve the safety of the decision.
[0014] In a possible implementation, the automatic driving method further includes: determining whether the self-vehicle has the right of way based on the self-vehicle and the first target vehicle; determining the right of way grabbing benefit based on the self-vehicle having the right of way, and the right of way grabbing benefit is used to establish the dynamic game model. It is understandable that a vehicle with the right of way can have the priority of passing during driving. Therefore, in the process of establishing the dynamic game model, the right of way grabbing benefit can be introduced, that is, a certain benefit is given to the vehicle with the right of way, so that the dynamic game model is more in line with the actual situation and improves the reliability of the decision-making results.
[0015] In a possible implementation, before 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, the first target vehicle set including a plurality of target vehicles whose distance from the self-vehicle is less than a preset threshold; determining a second target vehicle set according to the planned trajectory of the self-vehicle and the first target vehicle set, the second target vehicle set including a plurality of target vehicles whose predicted trajectories intersect with the planned trajectory of the self-vehicle; and determining the first target vehicle and the second target vehicle in the second target vehicle set. In this solution, before obtaining the first information of the first target vehicle and the second information of the second target vehicle, the key obstacles corresponding to the self-vehicle can be identified according to the position of the self-vehicle and the planned trajectory, and then the first target vehicle and the second target vehicle can be determined in the key obstacles corresponding to the self-vehicle. Through key obstacle identification, a large number of irrelevant targets can be filtered out.
[0016] In a possible implementation, determining the first target vehicle and the second target vehicle in the second target vehicle set specifically includes: obtaining the TTC corresponding to the ego vehicle and the TTC corresponding to a third target vehicle, the third target vehicle being a target vehicle in the second target vehicle set; determining the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle; if the difference between the TTC corresponding to the ego 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 lane set, determining the third target vehicle to be the first target vehicle; wherein the game lane set is a lane set having a game relationship with the lane in which the ego vehicle is located.
[0017] In a possible implementation, 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 trajectory lane set of the third target vehicle and the self-vehicle lane set have the same lane, then determining the third target vehicle as the second target vehicle; wherein the third target vehicle is a target vehicle in the second target vehicle set, the self-vehicle lane set includes a self-vehicle lane, a lane in front of the self-vehicle lane, and a lane behind the self-vehicle lane, the self-vehicle lane is the lane where the self-vehicle is located, and the risk lane set includes the self-vehicle lane set and a lane set to the left of the self-vehicle lane set and / or a lane set to the right of the self-vehicle lane set.
[0018] According to a second aspect of the present application, an automatic driving device is provided, including: an acquisition unit, configured to acquire first information of a first target vehicle and second information of a second target vehicle, wherein there is a risk of collision between the self-vehicle and 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; a processing unit, configured to establish a dynamic game model based on at least the first information and the second information, the first information being used to determine a conflict loss in the dynamic game model, and the second information being used to determine a safety distance loss in the dynamic game model, the conflict loss being the loss of the self-vehicle when both the self-vehicle and the first target vehicle choose to accelerate to overtake, and the safety distance loss being 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 configured to determine a driving strategy based on the dynamic game model; and a control unit is configured to control the driving of the self-vehicle based on the driving strategy.
[0019] In a possible implementation, the automatic driving device further includes: an identification unit for identifying that the vehicle is in a preset scenario, wherein the preset scenarios include: a driving scenario at an intersection without left-turn protection, an entrance and exit ramp driving scenario, and a lane merging driving scenario.
[0020] In a possible implementation manner, the ego vehicle can eliminate the risk of collision with the first target vehicle by accelerating to overtake or decelerating to yield.
[0021] In a possible implementation, the processing unit is specifically used to: obtain the time TTC corresponding to the ego vehicle from the trajectory conflict point; determine the TTC corresponding to the first target vehicle based on the first information; and determine the conflict loss in the dynamic game model based on the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
[0022] In a possible implementation, the processing unit is specifically used to determine a first safety distance loss or a second safety distance loss in the dynamic game model based on the position and speed of the vehicle and the second information; wherein the first safety distance loss is the front safety distance loss when executing an acceleration-to-overtake strategy, and the second safety distance loss is the rear safety distance loss when executing a deceleration-to-yield strategy.
[0023] In a possible implementation, the processing unit is further used to: determine whether the ego vehicle has the right of way based on the ego vehicle and the first target vehicle; determine the benefit of grabbing the right of way based on the ego vehicle having the right of way, and the benefit of grabbing the right of way is used to establish the dynamic game model.
[0024] In a possible implementation, the processing unit is further used to: determine a first target vehicle set based on the position of the ego vehicle, the first target vehicle set including multiple target vehicles whose distances from the ego vehicle are less than a preset threshold; determine a second target vehicle set based on the planned trajectory of the ego vehicle and the first target vehicle set, the second target vehicle set including multiple target vehicles whose predicted trajectories intersect with the planned trajectory of the ego vehicle; and determine the first target vehicle and the second target vehicle in the second target vehicle set.
[0025] In a possible implementation, the acquisition unit is further used to acquire the TTC corresponding to the ego vehicle and the TTC corresponding to a third target vehicle, wherein the third target vehicle is a target vehicle in the second target vehicle set; the processing unit is further used to determine the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle; the processing unit is further used to determine that the third target vehicle is the first target vehicle if the difference between the TTC corresponding to the ego 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 lane set; wherein the game lane set is a lane set that has a game relationship with the lane in which the ego vehicle is located.
[0026] In a possible implementation, the processing unit is specifically used to: if the third target vehicle is in a risk lane set, and the trajectory lane set of the third target vehicle and the self-vehicle lane set have the same lane, then determine that the third target vehicle is the second target vehicle; wherein the third target vehicle is a target vehicle in the second target vehicle set, the self-vehicle lane set includes the self-vehicle lane, the lane in front of the self-vehicle lane, and the lane behind the self-vehicle lane, the self-vehicle lane is the lane where the self-vehicle is located, and the risk lane set includes the self-vehicle lane set and the left lane set of the self-vehicle lane set and / or the right lane set of the self-vehicle lane set.
[0027] A third aspect of the present application provides an autonomous driving device, which may include a processor, wherein the processor is coupled to a memory, and the memory stores program instructions. When the program instructions stored in the memory are executed by the processor, the autonomous driving method described in the first aspect or any possible implementation of the first aspect is performed.
[0028] In a fourth aspect, the present application provides a computer-readable storage medium, which may include a program. When the program is run on a computer, the computer executes the autonomous driving method described in the above-mentioned first aspect or any possible implementation method of the first aspect.
[0029] In a fifth aspect, the present application provides an autonomous driving vehicle, which may include a processing circuit and a storage circuit, and the processing circuit and the storage circuit are configured to execute the autonomous driving method described in the first aspect or any possible implementation of the first aspect.
[0030] In a sixth aspect, the present application provides a circuit system, which may include a processing circuit, and the processing circuit is configured to execute the autonomous driving method described in the first aspect or any possible implementation of the first aspect.
[0031] In a seventh aspect, the present application provides a computer program which, when running on a computer, enables the computer to execute the automatic driving method described in the first aspect or any possible implementation of the first aspect.
[0032] In an eighth aspect of the present application, a chip system is provided, which may include a processor for supporting the autonomous driving device to implement the functions involved in the above aspects, for example, sending or processing the data and / or information involved in the above methods. In one possible design, the chip system may also include a memory for storing program instructions and data necessary for a server or communication device. The chip system may be composed of a chip, or may include a chip and other discrete devices.
[0033] For the specific implementation steps of the second to ninth aspects of the present application and various possible implementation methods, as well as the beneficial effects brought about by each possible implementation method, please refer to the description of the various possible implementation methods in the first aspect, and will not be repeated here one by one. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application;
[0035] Figure 2 A schematic diagram of a driving scenario at an intersection without left-turn protection provided in an embodiment of the present application;
[0036] Figure 3 A schematic diagram of an entrance and exit ramp driving scenario provided in an embodiment of the present application;
[0037] Figure 4 A schematic diagram of a lane merging driving scenario provided in an embodiment of the present application;
[0038] Figure 5 A schematic diagram of a process flow of an autonomous driving method provided in this application;
[0039] Figure 6 A flowchart of a method for identifying a target vehicle provided in an embodiment of the present application;
[0040] Figure 7 A schematic diagram of a lane division method provided in an embodiment of the present application;
[0041] Figure 8 A schematic diagram of a game lane division method provided in an embodiment of the present application;
[0042] Fig. 9 A schematic diagram of another gaming lane division method provided in an embodiment of the present application;
[0043] Fig.10 A schematic diagram of filtering out a first target vehicle and a second target vehicle provided in an embodiment of the present application;
[0044] Fig.11 A schematic diagram of the structure of an automatic driving device 1100 provided in an embodiment of the present application;
[0045] Fig.12 A schematic diagram of the structure of an autonomous driving vehicle provided in an embodiment of the present application;
[0046] Fig.13 A schematic diagram of the structure of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION
[0047] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances. This is merely a way of distinguishing objects with the same properties when describing the embodiments of the present application. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, so that a process, method, system, product, or device that includes a series of units is not necessarily limited to those units, but may include other units that are not explicitly listed or inherent to these processes, methods, products, or devices.
[0048] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0049] In order to facilitate the understanding of this solution, the present application embodiment first combines Figure 1 The structure of the vehicle is introduced. The automatic driving method provided in the embodiment of the present application can be applied as follows: Figure 1 See the vehicle shown. Figure 1 , Figure 1 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application.
[0050] In one embodiment, the autonomous driving vehicle 100 may be configured to be in a fully or partially autonomous driving mode. For example, the autonomous driving vehicle 100 may control itself while in the autonomous driving mode, and may determine the current state of the vehicle and its surroundings through human operation, determine the possible behavior of at least one other vehicle in the surroundings, and determine the confidence level corresponding to the possibility of the other vehicle performing the possible behavior, and control the autonomous driving vehicle 100 based on the determined information. When the autonomous driving vehicle 100 is in the autonomous driving mode, the autonomous driving vehicle 100 may be set to operate without human interaction.
[0051] The autonomous vehicle 100 may include various subsystems, such as a travel system 102, a sensor system 104, a control system 106, one or more peripheral devices 108, and a power supply 110, a computer system 112, and a user interface 116. Optionally, the autonomous vehicle 100 may include more or fewer subsystems, and each subsystem may include multiple elements. In addition, each subsystem and element of the autonomous vehicle 100 may be interconnected by wire or wirelessly.
[0052] The propulsion system 102 may include components that provide powered movement for 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 a combination of other types of engines, such as a hybrid engine consisting of a gasoline engine and an electric motor, or a hybrid engine consisting of an internal combustion engine and an air compression engine. The engine 118 converts the energy source 119 into mechanical energy.
[0053] Examples of energy source 119 include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electricity. Energy source 119 can also provide energy for other systems of autonomous vehicle 100.
[0054] The transmission 120 can 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. Among them, the drive shaft may include one or more shafts that can be coupled to one or more wheels 121.
[0055] The sensor system 104 may include several sensors that sense information about the environment surrounding the autonomous vehicle 100. For example, the sensor system 104 may include a positioning system 122 (the positioning system may be a GPS system, or a Beidou system or other positioning systems), an inertial measurement unit (IMU) 124, a radar 126, a laser rangefinder 128, and a camera 130. The sensor system 104 may also include sensors of the internal systems of the monitored 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 objects and their corresponding characteristics (position, shape, direction, speed, etc.). Such detection and recognition are key functions for the safe operation of the autonomous autonomous vehicle 100.
[0056] The positioning system 122 may be used to estimate the geographic location of the autonomous vehicle 100. The IMU 124 is used to sense the position and orientation changes of the autonomous vehicle 100 based on inertial acceleration. In one embodiment, the IMU 124 may be a combination of an accelerometer and a gyroscope.
[0057] Radar 126 may utilize radio signals to sense objects within the surrounding environment of 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.
[0058] The laser rangefinder 128 may utilize lasers 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, a laser scanner, and one or more detectors, among other system components.
[0059] 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.
[0060] The control system 106 controls 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 brake unit 136 , a sensor fusion algorithm 138 , a computer vision system 140 , a path control system 142 , and an obstacle avoidance system 144 .
[0061] The steering system 132 is operable to adjust the forward direction of the autonomous vehicle 100. For example, in one embodiment, it may be a steering wheel system.
[0062] The throttle 134 is used to control the operating speed of the engine 118 and thereby control the speed of the autonomous vehicle 100 .
[0063] The brake unit 136 is used to control the deceleration of the autonomous driving vehicle 100. The brake unit 136 can use friction to slow down the wheel 121. In other embodiments, the brake unit 136 can convert the kinetic energy of the wheel 121 into electric current. The brake unit 136 can also take other forms to slow down the rotation speed of the wheel 121 to control the speed of the autonomous driving vehicle 100.
[0064] The computer vision system 140 can be operated to process and analyze the images captured by the camera 130 in order to identify objects and / or features in the environment surrounding the autonomous vehicle 100. The objects and / or features may include traffic signs, road boundaries, and obstacles. The computer vision system 140 may use object recognition algorithms, Structure from Motion (SFM) algorithms, video tracking, and other computer vision techniques. In some embodiments, the computer vision system 140 can be used to map the environment, track objects, estimate the speed of objects, and the like.
[0065] The route control system 142 is used to determine the driving route of the autonomous vehicle 100. In some embodiments, the route control system 142 may combine data from the sensor 138, the GPS 122, and one or more predetermined maps to determine the driving route for the autonomous vehicle 100.
[0066] The obstacle avoidance system 144 is used to identify, evaluate, and avoid or otherwise negotiate potential obstacles in the environment of the vehicle 100 .
[0067] Of course, in one example, the control system 106 may include additional or alternative components other than those shown and described, or may also reduce some of the components shown above.
[0068] The autonomous vehicle 100 interacts with external sensors, other vehicles, other computer systems, or users through peripherals 108. The peripherals 108 may include a wireless communication system 146, an onboard computer 148, a microphone 150, and / or a speaker 152.
[0069] In some embodiments, the peripheral device 108 provides a means for a user of the autonomous vehicle 100 to interact with the user interface 116. For example, the onboard computer 148 can provide information to the user of the autonomous vehicle 100. The user interface 116 can also operate the onboard computer 148 to receive input from the user. The onboard computer 148 can be operated via a touch screen. In other cases, the peripheral device 108 can provide a means for the autonomous vehicle 100 to communicate with other devices located in the vehicle. For example, the microphone 150 can receive audio (e.g., voice commands or other audio input) from the user of the autonomous vehicle 100. Similarly, the speaker 152 can output audio to the user of the autonomous vehicle 100.
[0070] The wireless communication system 146 can communicate wirelessly with one or more devices directly or via a communication network. For example, the wireless communication system 146 can use 3G cellular communication, such as CDMA, EVDO, GSM / GPRS, or 4G cellular communication, such as LTE. Or 5G cellular communication. The wireless communication system 146 can communicate with a wireless local area network (WLAN) using WiFi. In some embodiments, the wireless communication system 146 can communicate directly with the device using an infrared link, Bluetooth, or ZigBee. 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, which may include public and / or private data communications between vehicles and / or roadside stations.
[0071] The power source 110 can provide power to various components of the autonomous vehicle 100. In one embodiment, the power source 110 can be a rechargeable lithium-ion or lead-acid battery. One or more battery packs of such batteries can 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 can be implemented together, such as in some all-electric vehicles.
[0072] Some or all functions of the autonomous vehicle 100 are controlled by a computer system 112. The computer system 112 may include at least one processor 113 that executes 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.
[0073] 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. Figure 1 The processor, memory, and other elements of the computer 110 in the same block are functionally illustrated, but it will be appreciated by those of ordinary skill in the art that the processor, computer, or memory may actually include multiple processors, computers, or memories that may or may not be stored in the same physical housing. For example, the memory may be a hard drive or other storage medium located in a housing different from the computer 110. Therefore, references to a processor or computer will be understood to include references to a collection of processors or computers or memories that may or may not operate in parallel. Different from using a single processor to perform the steps described herein, some components such as the steering assembly and the deceleration assembly may each have their own processor that performs only calculations related to the functions specific to the component.
[0074] In various aspects described herein, the processor may be located remote from the vehicle and in wireless communication with the vehicle. In other aspects, some of the processes described herein are performed on a processor disposed within the vehicle and others are performed by a remote processor, including taking the necessary steps to perform a single maneuver.
[0075] In some embodiments, data storage 114 may include instructions 115 (e.g., program logic) that may be executed by processor 113 to perform various functions of autonomous vehicle 100, including those described above. Data storage 114 may also include additional instructions, including instructions to send data to, receive data from, interact with, and / or control one or more of propulsion system 102, sensor system 104, control system 106, and peripherals 108.
[0076] In addition to instructions 115, data storage device 114 may also store data such as road maps, route information, the vehicle's location, direction, speed, and other such vehicle data, as well as other information. Such information may be used by autonomous vehicle 100 and computer system 112 during operation of autonomous vehicle 100 in autonomous, semi-autonomous, and / or manual modes.
[0077] The user interface 116 is used to provide information to or receive information from a user of the autonomous vehicle 100. Optionally, the user interface 116 may include one or more input / output devices within the set of peripheral devices 108, such as a wireless communication system 146, a vehicle computer 148, a microphone 150, and a speaker 152.
[0078] The computer system 112 may control functions of the autonomous vehicle 100 based on input 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 may be operable to provide control over many aspects of the autonomous vehicle 100 and its subsystems.
[0079] Alternatively, one or more of the above components may be installed or associated separately from the autonomous vehicle 100. For example, the data storage device 114 may be partially or completely separate from the vehicle 1100. The above components may be communicatively coupled together in a wired and / or wireless manner.
[0080] Optionally, the above components are only examples. In actual applications, the components in the above modules may be added or deleted according to actual needs. Figure 1 It should not be understood as limiting the embodiments of the present application.
[0081] An autonomous vehicle traveling on a road, such as the autonomous vehicle 100 above, can identify objects in its surrounding environment to determine adjustments to the current speed. The objects can be other vehicles, traffic control devices, or other types of objects. In some examples, each identified object can be considered independently, and based on the object's respective characteristics, such as its current speed, acceleration, spacing from the vehicle, etc., can be used to determine the speed to be adjusted by the autonomous vehicle.
[0082] Optionally, the autonomous vehicle 100 or a computing device associated with the autonomous vehicle 100 (eg, Figure 1 The computer system 112, computer vision system 140, data storage device 114) can predict the behavior of the identified objects based on the characteristics of the identified objects and the state of the surrounding environment (e.g., traffic, rain, ice on the road, etc.). Optionally, each of the identified objects depends on the behavior of each other, so all of the identified objects can also be considered together to predict the behavior of a single identified object. The autonomous driving vehicle 100 can adjust its speed based on the predicted behavior of the identified objects. In other words, the autonomous driving car can determine what stable state the vehicle will need to adjust to (e.g., accelerate, decelerate, or stop) based on the predicted behavior of the objects. In this process, other factors can also be considered to determine the speed of the autonomous driving vehicle 100, such as the lateral position of the autonomous driving vehicle 100 in the road it is traveling on, the curvature of the road, the proximity of static and dynamic objects, etc.
[0083] 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 so that the autonomous vehicle follows a given trajectory and / or maintains a safe lateral and longitudinal distance from objects near the autonomous vehicle (e.g., cars in adjacent lanes on the road).
[0084] The above-mentioned autonomous driving vehicle 100 can be a car, a truck, a motorcycle, a bus, a ship, an airplane, a helicopter, a lawn mower, an entertainment vehicle, an amusement park vehicle, construction equipment, a tram, a golf cart, a train, etc., and the embodiments of the present application are not particularly limited.
[0085] With the further development and continuous evolution of autonomous driving technology, people's requirements for autonomous driving vehicles are getting higher and higher, requiring the behavior of autonomous driving vehicles to be more intelligent and anthropomorphic. In the relevant technologies of autonomous driving, safety is usually the primary criterion, and it is expected to provide an excessively redundant safety margin at all times. As a result, the driving strategy of autonomous driving vehicles tends to be conservative. Autonomous driving vehicles are prone to stagnation in some special scenarios, and sometimes even block traffic, resulting in a situation where humans take over.
[0086] Among them, the aforementioned special scenarios may refer to scenarios where there is a trajectory conflict between the autonomous driving vehicle and the target vehicle, such as driving scenarios at intersections without left-turn protection, driving scenarios on entrance and exit ramps, and driving scenarios where lanes merge.
[0087] For example, see Figure 2 , Figure 2 A schematic diagram of a driving scenario at an intersection without left-turn protection provided in an embodiment of the present application. Figure 2 As shown in the figure, at an intersection without left-turn protection, the planned trajectory of the ego vehicle (i.e., the vehicle itself) is to go straight, and the planned trajectory of the target vehicle is to turn left. Since going straight and turning left share the same green light, there is often a conflict between the trajectory of the ego vehicle and the target vehicle.
[0088] For example, see Figure 3 , Figure 3 A schematic diagram of an entrance and exit ramp driving scenario provided in an embodiment of the present application. Figure 3 As shown in Figure 1, at the entrance and exit ramps, the planned trajectory of the ego vehicle is to go straight, and the planned trajectory of the target vehicle is to merge from the ramp into the lane where the ego vehicle is traveling. In this case, there will also be a conflict between the ego vehicle and the target vehicle's trajectory.
[0089] For example, see Figure 4 , Figure 4 A schematic diagram of a lane merging driving scenario provided in an embodiment of the present application. Figure 4As shown in the figure, in the scenario where two lanes merge into a single lane, the planned trajectory of the ego vehicle is to go straight, and the planned trajectory of the target vehicle is to enter the merged lane. In this case, there is usually a conflict between the trajectory of the ego vehicle and the target vehicle.
[0090] In related technologies, when faced with trajectory conflicts, autonomous vehicles tend to make very conservative decisions, and in most cases they will conservatively choose to slow down and give way, which will result in very low traffic efficiency. Therefore, choosing an appropriate interactive game strategy can greatly improve traffic efficiency in trajectory conflict scenarios, reduce the number of takeovers caused by conservative decision-making, and is also a manifestation of more intelligent autonomous driving.
[0091] At present, when a trajectory conflict occurs between an autonomous driving vehicle and a target vehicle, the autonomous driving vehicle often adopts a one-on-one game strategy with the target vehicle, while ignoring the impact of the driving environment on the game process, that is, ignoring the driving risks outside the one-on-one game process, which can easily lead to certain risks in the driving strategies made by the autonomous driving vehicle.
[0092] For example, when the autonomous vehicle chooses the driving strategy of accelerating to overtake during the game, the acceleration of the autonomous vehicle will reduce the safe distance between the autonomous vehicle and the vehicle in front of it, resulting in a risk of collision between the autonomous vehicle and the vehicle in front of it. For another example, when the autonomous vehicle chooses the driving strategy of slowing down to give way during the game, the slowing down of the autonomous vehicle will reduce the safe distance between the autonomous vehicle and the vehicle behind it, resulting in a risk of collision between the autonomous vehicle and the vehicle behind it.
[0093] That is to say, outside of the one-on-one game process, autonomous driving vehicles also face the risk of collision with vehicles around them, while existing game strategies usually ignore driving risks outside of the one-on-one game process.
[0094] In view of this, the present application provides an automatic driving method, which can be applied to Figure 1 In the autonomous driving vehicle 100 shown in the figure, the safe driving of the autonomous driving vehicle in the trajectory conflict scenario is ensured.
[0095] To facilitate understanding, the technical terms mentioned in the embodiments of the present application are first introduced below.
[0096] Intersection: An intersection where two or more roads meet.
[0097] Intersection without left-turn protection: An intersection where vehicles going straight and vehicles turning left share the same green light, and there is no independent left-turn light.
[0098] Ego: The self-driving vehicle is the ego.
[0099] Target vehicle (object): A vehicle detected around the autonomous vehicle.
[0100] Dynamic game: A game theory in which there is a sequence in the game process, and the later actors will make corresponding decisions based on the behavior of the earlier actors, also known as "multi-stage game".
[0101] Subgame: refers to a set of new games formed after all participants in a dynamic game have taken an action successively. Each game in this set is called a subgame.
[0102] Subgame perfect Nash equilibrium: A subgame perfect Nash equilibrium is formed if and only if the strategies of the participants in the series of subgames (second generation, third generation, etc.) form a Nash equilibrium in each subgame. Subgame perfect Nash equilibrium requires that the decisions of the participants are optimal in each subgame at each time node.
[0103] See also Figure 5 , Figure 5 This is a flow chart of an automatic driving method provided by this application. Figure 5 As shown, an automatic driving method provided in an embodiment of the present application may include the following steps:
[0104] 501. Obtain first information of a first target vehicle and second information of a second target vehicle.
[0105] Among them, the first target car and the second target car can be vehicles in the surrounding environment of the self-vehicle, and the self-vehicle has a collision risk with the first target car and the second target car, and the self-vehicle can eliminate the collision risk with the first target car by accelerating to rush or slowing down to give way. In other words, the self-vehicle and the first target car have a direct game relationship, and in the game process, the driving strategies that the self-vehicle and the first target car can make are accelerating to rush or slowing down to give way. There is no direct game relationship between the second target car and the self-vehicle, but the second target car will affect and limit the decision-making and action range of the self-vehicle in the game process.
[0106] The first information of the first target vehicle includes the position information and speed information of the first target vehicle, and the second information of the second target vehicle includes the position information and speed information of the second target vehicle. The position information of the first target vehicle and the second target vehicle can be completed by means of global positioning system (global position system, GPS), real-time dynamic positioning (real-time kinematic, RTK), camera and laser radar. In a possible implementation, the possible position of the vehicle can be judged by combining the pre-stored map, GPS position information and millimeter wave measurement information, and the probability of occurrence of the possible position of the vehicle is calculated, so as to determine the specific position of the vehicle. It should be noted that the scheme provided by the present application can obtain the position information of the vehicle in a variety of ways, and the method of obtaining the position information of the vehicle in the relevant technology can be adopted in the embodiments of the present application. The speed information of the first target vehicle and the second target vehicle can be measured by the speed measuring radar or sensor on the vehicle. The speed information of the first target vehicle and the second target vehicle can, for example, include the real-time speed and acceleration of the first target vehicle and the second target vehicle.
[0107] In a possible embodiment, the vehicle may establish a dynamic game model when it recognizes that it is in a preset scenario. In this preset scenario, there is a risk of collision between the vehicle and a target vehicle near the vehicle, and the vehicle can eliminate the risk of collision with the target vehicle by accelerating to overtake or slowing down to yield. Exemplarily, the preset scenario includes but is not limited to driving scenarios at intersections without left-turn protection, driving scenarios at entrance and exit ramps, and driving scenarios for lane merging. The preset scenario may also be other scenarios that meet the aforementioned conditions, and this embodiment does not specifically limit the preset scenario.
[0108] For ease of description, the following will describe an embodiment of the present application by taking a preset scenario of driving at an intersection without left-turn protection as an example.
[0109] 502. Establish a dynamic game model based on at least the first information and the second information, wherein the first information is used to determine a conflict loss in the dynamic game model, and the second information is used to determine a safety distance loss in the dynamic game model.
[0110] After obtaining the first information of the first target vehicle and the second information of the second target vehicle, a dynamic game model may be established based on the first information and the second information to iteratively obtain the driving strategy of the vehicle.
[0111] Since the first target vehicle has a direct game relationship with the ego vehicle, the conflict 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 ego vehicle when the ego vehicle chooses to accelerate and overtake and the first target vehicle also chooses to accelerate and overtake. The loss of the ego vehicle refers to a measure of the possible changes in one or both of the road passability or safety of the ego vehicle caused by the choice of the ego vehicle's game strategy during the game process.
[0112] In a possible implementation, the conflict loss may be determined according to the time to conflict (TTC) corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
[0113] For example, after obtaining the planned trajectory of the ego vehicle and the predicted trajectory of the first target vehicle, the trajectory conflict point, i.e., the point where the planned trajectory and the predicted trajectory intersect, can be determined based on the planned trajectory of the ego vehicle and the predicted trajectory of the first target vehicle. For the ego vehicle, the distance to the trajectory conflict point (Distance To Conflict, DTC) corresponding to the ego vehicle can be determined according to the current position of the ego vehicle and the position of the trajectory conflict point. After determining the DTC corresponding to the ego vehicle, the TTC corresponding to the ego vehicle can be determined according to the current speed and acceleration of the ego vehicle.
[0114] Specifically, the DTC corresponding to the vehicle can be determined according to formula (1), which is as follows:
[0115]
[0116] Among them, Xi and Yi represent the horizontal coordinate and vertical coordinate of the i-th point of the trajectory respectively; Xi+1 and Yi+1 represent the horizontal coordinate and vertical coordinate of the i+1-th point of the trajectory respectively; Idxc represents the sequence number of the intersection point in the trajectory.
[0117] The TTC corresponding to the ego vehicle can be determined based on formula (2), which is as follows:
[0118]
[0119] Among them, a represents acceleration and v represents velocity.
[0120] Similarly, the DTC and TTC of the first target vehicle can also be calculated according to the above formula (1) and formula (2).
[0121] After obtaining the TTC corresponding to the ego vehicle and the TTC of the first target vehicle, the difference between the two TTCs may be determined, and the conflict loss in the dynamic game model may be determined according to the difference.
[0122] Exemplarily, the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle may be determined based on formula (3), which is as follows:
[0123] δT=|TTC e -TTC o | Formula (3)
[0124] Among them, δT represents the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle, TTCe represents the TTC corresponding to the ego vehicle, and TTCo represents the TTC corresponding to the first target vehicle.
[0125] After obtaining the difference δT, the conflict loss Ce(δT) can be calculated based on a pre-set function and the difference δT. The pre-set function can be a linear function, an exponential function or a logarithmic function. The embodiment of the present application does not specifically limit the functional relationship between the difference δT and the conflict loss Ce(δT). For example, the pre-set function can be an inverse proportional function, that is, Ce(δT)=1 / δT. Generally speaking, the larger the difference δT, the smaller the conflict loss, and the smaller the difference δT, the larger the conflict loss.
[0126] The safety distance loss in the dynamic game model can be determined based on the driving information of the ego vehicle and the second information of the second target vehicle. The safety distance loss is used to represent the loss of the ego vehicle when the ego vehicle executes the strategy of accelerating to overtake or slowing down to give way, which causes the spatial distance between the ego vehicle and the second target vehicle in front or behind to change.
[0127] For example, during the actual driving process, the first safety distance loss or the 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.
[0128] The first safety distance loss represents the loss of the vehicle when the vehicle executes the acceleration strategy, the front safety distance between the autonomous driving vehicle and the second target vehicle in front of it decreases, and the probability of a collision between the vehicle and the second target vehicle in front increases. The second safety distance loss represents the loss of the vehicle when the vehicle executes the deceleration strategy, the rear safety distance between the autonomous driving vehicle and the second target vehicle behind it decreases, and the probability of a collision between the vehicle and the second target vehicle behind it increases. For example, when the second target vehicle is in front of the vehicle, the first safety distance loss corresponding to the second target vehicle can be determined; when the second target vehicle is behind the vehicle, the second safety distance loss corresponding to the second target vehicle can be determined.
[0129] Specifically, the first safety distance loss or the second safety distance loss can be determined based on the spatial distance between the second target vehicle and the vehicle, and the relative speed between the second target vehicle and the vehicle. For example, the first safety distance loss or the second safety distance loss can be obtained by dividing the spatial distance between the second target vehicle and the vehicle by the relative speed between the second target vehicle and the vehicle. Generally speaking, the larger the spatial distance between the second target vehicle and the vehicle, the larger the first safety distance loss or the second safety distance loss; the larger the relative speed between the second target vehicle and the vehicle, the smaller the first safety distance loss or the second safety distance loss.
[0130] In one possible embodiment, it is also possible to determine whether the self-vehicle has the right of way based on the self-vehicle and the first target vehicle. In the case of determining that the self-vehicle has the right of way, the right of way grabbing benefit of the self-vehicle can be determined, and the right of way grabbing benefit can be used to establish a dynamic game model. It is understandable that a vehicle with the right of way can have priority in passing during driving. Therefore, in the process of establishing a dynamic game model, the right of way grabbing benefit can be introduced, that is, a certain benefit is given to the vehicle with the right of way, so that the dynamic game model is more in line with the actual situation and improves the reliability of the decision-making results. Generally speaking, relative to isolated vehicles (that is, there are no other vehicles in front and behind the vehicle), vehicles in the traffic flow have the right of way; relative to turning vehicles, such as vehicles turning left or right, vehicles going straight have the right of way. In actual situations, the self-vehicle can determine whether the self-vehicle has the right of way based on the driving scene and traffic conditions.
[0131] 503. Determine the driving strategy based on the dynamic game model.
[0132] After the dynamic game model is established, the driving strategy of the vehicle can be determined based on the dynamic game model, and the driving strategy is to accelerate to overtake or decelerate to yield.
[0133] For ease of understanding, the following will introduce the process of establishing a dynamic game model and determining a driving strategy based on the dynamic game model.
[0134] 1. Determine the multiple elements of the game system: game subjects, game subject strategy sets and profit and loss variables.
[0135] Among them, the game subjects are the ego vehicle and the first target vehicle; the game subject strategy set B includes accelerating to overtake or slowing down to yield; the profit and loss variables of the game subject in the game process include profit variables and loss variables. Specifically, the profit variable includes the basic loss Ge of the ego vehicle and the road right-over profit Re; the loss variable includes the avoidance loss Fe, the conflict loss Ce(δT), the mutual yield loss Me, the first safety distance loss Sf and the second safety distance loss Sa.
[0136] Gains and losses are defined as follows:
[0137] Basic income Ge: constant, which is the basic income in the game relationship.
[0138] The benefit of right-of-way grabbing Re: constant, the benefit when the party with the right of way has the intention to grab the right of way, reflects the impact of the right of way on the right of way priority.
[0139] Avoidance loss Fe: constant, the loss of the ego vehicle when the ego vehicle chooses to slow down and give way, and the first target vehicle chooses to speed up and rush over.
[0140] Conflict loss Ce(δT): function, when the ego vehicle chooses to accelerate and overtake, and the first target vehicle chooses to accelerate and overtake, the loss of the ego vehicle; this δT is the δT in formula (4).
[0141] Mutual yield loss Me: constant, the loss of the ego vehicle when the ego vehicle chooses to slow down and yield, and the first target vehicle also chooses to slow down and yield.
[0142] First safety distance loss Sf(Df,δvf,Be): function, when the ego vehicle chooses to accelerate and overtake, the safety distance loss in front due to the change in the spatial distance between the ego vehicle and the second target vehicle. For example, a possible form of the first safety distance loss Sf is shown in formula (4):
[0143]
[0144] Wherein, vegoS is the longitudinal speed of the ego vehicle; vfrontS is the longitudinal speed of the second target vehicle in front of the ego vehicle; δvf is the speed difference between the ego vehicle and the second target vehicle; TTR (Time To Risk) represents the collision risk time, which refers to the collision time between the ego vehicle and the second target vehicle in front of the ego vehicle when the second target vehicle in front of the ego vehicle brakes at a deceleration of a (a is a positive number) and the ego vehicle is moving at a constant speed. Df is the longitudinal relative distance between the ego vehicle and the second target vehicle in front of the ego vehicle. Sf is the first safety distance loss, k1, k2 and k3 are parameters for adjusting the steepness of the function. kB is the grabway coefficient, the value range of offset is (0,1), GW (grabway) represents the grabway coefficient, and YD (yield) represents the yield coefficient. For example, when the offset value is 0.2, when the ego vehicle strategy Be is GW, kB is 1.2; when the ego vehicle strategy Be is TD, kB is 0.8.
[0145] It is understandable that, when it is determined that there are multiple second target vehicles in front of the vehicle, the first safety distance losses corresponding to the multiple second target vehicles can be determined respectively. In the process of establishing the dynamic game model, the first safety distance loss with the largest loss value is selected from the first safety distance losses corresponding to the multiple second target vehicles to establish the profit equation in the dynamic game model.
[0146] Second safety distance loss Sa (Da, δva, Be): function, when the ego vehicle chooses to slow down and yield, the safety distance loss caused by the change in the spatial distance between the ego vehicle and the second target vehicle at the rear. For example, a possible form of the second safety distance loss Sa is shown in formula (5):
[0147]
[0148] Among them, vegoS is the longitudinal speed of the ego vehicle; vafterS is the longitudinal speed of the second target vehicle behind the ego vehicle; δvf is the speed difference between the ego vehicle and the second target vehicle. TTR represents the collision risk time, which refers to the collision time between the ego vehicle and the second target vehicle when the second target vehicle behind the ego vehicle accelerates at an acceleration of a (a is a positive number) and the ego vehicle is moving at a constant speed; Da is the longitudinal relative distance between the ego vehicle and the second target vehicle behind the ego vehicle. Sa is the second safety distance loss, k1, k2 and k3 are the parameters for adjusting the steepness of the function respectively. kB is the grab coefficient, the value range of offset is (0,1), GW represents the grab coefficient, and YD represents the yield coefficient.
[0149] Similarly, when it is determined that there are multiple second target vehicles behind the vehicle, the second safety distance losses corresponding to the multiple second target vehicles can be determined respectively. In the process of establishing the dynamic game model, a second safety distance loss with the largest loss value is selected from the second safety distance losses corresponding to the multiple second target vehicles to establish the profit equation in the dynamic game model.
[0150] 2. After determining the multiple elements of the game system, establish the profit equation.
[0151] Assume that the probability of the ego car accelerating to overtake is x, and the probability of slowing down to yield is (1-x); the probability of the first target car accelerating to overtake is y, and the probability of slowing down to yield is (1-y). According to the above-defined gains and losses, the pure strategy expected gain Eel of the ego car choosing to accelerate to overtake and the pure strategy expected gain Ee2 of the ego car choosing to slow down to yield are shown in formula (6) and formula (7) respectively:
[0152] E e1 =G e -C e y+R e -S f +S a Formula (6)
[0153] E e2 =G e -M e +(M e -F e )y+S f -S a Formula (7)
[0154] Therefore, according to formula (6) and formula (7), we can derive the expected return E of the mixed strategy when the ego car accelerates to grab the other car with probability x and slows down to give way with probability (1-x): e The expected return of this mixed strategy is E e It can be shown as formula (8):
[0155]
[0156] 3. After establishing the profit equation, establish the dynamic replication equation.
[0157] The dynamic replication equation is an iterative equation for each frame of data. It is the differential δx of the probability of the self-car giving way established based on the probability of the game object giving way, the probability of the self-car giving way and the profit equation of the current frame data, that is, the change in x, that is, Specifically, according to formulas (6) and (8), the dynamic replication equation of the iteration probability of the ego vehicle can be derived, and the dynamic replication equation can be specifically shown as formula (9):
[0158]
[0159] After obtaining the dynamic replication equation, the probability x of the vehicle accelerating to overtake can be iterated based on the dynamic replication equation and the probability y of the first target vehicle accelerating to overtake predicted in the current frame. Finally, the driving strategy is determined by setting the probability threshold xthreshold and the probability differential threshold δxthreshold.
[0160] For example, the probability thresholds of the acceleration and overtaking decision can be set to xgrabwayl and xgrabway2, the probability thresholds of the deceleration and yielding decision can be set to xyieldl and xyield2, and the probability differential threshold can be set to δxthreshold, where xyield2<xyieldl<xgrabwayl<xgrabway2. When the probability x of the ego vehicle accelerating and overtaking is between (xyieldl, xgrabwayl), continue to use the dynamic replication equation to iterate x; when the probability x of the ego vehicle accelerating and overtaking is between (xyield2, xyieldl) and (xgrabwayl, xgrabway2), the dynamic replication equation is determined. Is it less than the probability differential threshold δxthreshold? If the dynamic copy equation If it is less than the probability differential threshold δxthreshold, it can be considered to have converged and a decision can be made, that is, when x is at (xyield2, xyieldl), the driving strategy is to slow down and yield, and when x is at (xgrabwayl, xgrabway2), the driving strategy is to speed up and grab; if the dynamic replication equation If the probability differential threshold δxthreshold is not less than the threshold, the dynamic replication equation is continued to iterate x. When x<xyield2 or x>xgrabway2, it can also be considered to have converged, and a decision can be made, that is, when x<xyield2, the driving strategy is to slow down and yield, and when x>xgrabway2, the driving strategy is to speed up and grab.
[0161] 504. Control the driving of the vehicle according to the driving strategy.
[0162] It is understandable that after the driving strategy is determined, the driving of the vehicle can be controlled according to the driving strategy. For example, when the driving strategy is to accelerate and overtake, the speed of the vehicle is increased so that the vehicle can pass the trajectory conflict point before the first target vehicle reaches the trajectory conflict point. When the driving strategy is to slow down and yield, the speed of the vehicle is reduced and the vehicle is controlled to slow down and yield so that the vehicle can reach the trajectory conflict point after the first target vehicle passes the trajectory conflict point.
[0163] In this embodiment, by acquiring information about the game target vehicle that has a game relationship with the self-vehicle and information about the risk target vehicle that affects the decision of the self-vehicle, a dynamic game model that introduces conflict loss and safety distance loss is established. Since the risk impact of the driving environment on the game process is added to the dynamic game model, risk decision-making and game decision-making can be combined, effectively reducing the risk of the game process and improving the safety of autonomous driving.
[0164] The above describes the process of establishing a dynamic game model and determining a driving strategy based on the dynamic game model. The following describes in detail the process of the ego vehicle identifying the first target vehicle and the second target vehicle.
[0165] See also Figure 6 , Figure 6 A flowchart of a method for identifying a target vehicle provided in an embodiment of the present application. Figure 6 The process of the self-vehicle identifying the first target vehicle and the second target vehicle may include the following steps:
[0166] 601. Determine a first target vehicle set according to a position of the ego vehicle, where the first target vehicle set includes a plurality of target vehicles whose distances from the ego vehicle are less than a preset threshold.
[0167] Before obtaining the first information of the first target vehicle and the second information of the second target vehicle, the key obstacles corresponding to the ego vehicle can be identified according to the position of the ego vehicle and the planned trajectory, and then the first target vehicle and the second target vehicle can be determined from the key obstacles corresponding to the ego vehicle. Through key obstacle identification, a large number of irrelevant targets can be filtered out. Among them, key obstacles refer to movable targets such as vehicles, pedestrians, and cyclists that are highly correlated with the ego vehicle and may have an impact on the driving behavior of the ego vehicle. For ease of description, the following will take the case where the key obstacle is a vehicle as an example for explanation.
[0168] In this embodiment, there may be multiple ways to determine the first target vehicle set according to the position of the ego vehicle.
[0169] In a possible implementation, the first target vehicle set can be determined by the Euclidean distance between the vehicle and other vehicles. Exemplarily, the Euclidean distance between the vehicle and the vehicles in the surrounding environment can be determined based on the position of the vehicle and the positions of the vehicles in the surrounding environment of the vehicle. When the Euclidean distance between the vehicle and a certain vehicle is less than a preset Euclidean distance threshold, the vehicle can be determined to be a vehicle in the first target vehicle set. Specifically, whether the Euclidean distance between the vehicle and a certain vehicle is less than a preset Euclidean distance threshold can be determined by formula (10).
[0170]
[0171] Among them, Xe represents the horizontal coordinate of the vehicle, Ye represents the vertical coordinate of the vehicle, Xo represents the horizontal coordinate of the target vehicle, Yo represents the vertical coordinate of the target vehicle, and Dthreshold is the Euclidean distance threshold.
[0172] In a possible implementation, a range frame, such as a bendable rectangular frame, may be made around the vehicle with the planned trajectory of the vehicle as the central axis, and a ray method may be used to determine whether the target vehicle is within the range frame, thereby determining the first target vehicle set.
[0173] Specifically, the method of determining whether the target vehicle is within the range box by using the ray method can be as shown in formula (11).
[0174]
[0175] Where Yc represents the ordinate of the intersection of the ray drawn vertically upward from the target vehicle as the starting point and the range box; X(P,i) represents the abscissa of the i-th point in the range box; X(P,i+1) represents the abscissa of the i+1-th point in the range box; Y(P,i) represents the ordinate of the i-th point in the range box; Y(P,i+1) represents the ordinate of the i+1-th point in the range box; Sp represents the number of points contained in the range box; i represents the number used to traverse the range box; and count represents the number of intersections. In formula (11), whether the target vehicle is within the range box can be determined based on the parity of count. If count is an odd number, it can be determined that the target vehicle is within the range box; if count is an even number, it can be determined that the target vehicle is outside the range box.
[0176] 602. Determine a second target vehicle set according to the planned trajectory of the ego vehicle and the first target vehicle set, where the second target vehicle set includes a plurality of target vehicles whose predicted trajectories intersect with the planned trajectory of the ego vehicle.
[0177] After determining the first target vehicle set, the target vehicles whose predicted trajectories intersect with the planned trajectory of the ego vehicle in the first target vehicle set can be further determined, thereby determining the second target vehicle set. Exemplarily, for each target vehicle in the first target vehicle set, it can be determined based on formula (12) whether the predicted trajectory of the target vehicle intersects with the planned trajectory of the ego vehicle.
[0178]
[0179] Among them, SEP is the number of points contained in the planned trajectory of the ego vehicle, SOP is the number of points contained in the predicted trajectory of the target vehicle, i and j are the coefficients of the traversal trajectory, X(EP, i) represents the horizontal coordinate of the i-th point of the ego vehicle, Y(EP, i) represents the vertical coordinate of the i-th point of the ego vehicle, X(OP, i) represents the horizontal coordinate of the i-th point of the target vehicle, and Y(OP, i) represents the vertical coordinate of the i-th point of the target vehicle. r and s are the coefficients of the intersection of two straight lines determined by the two points, projected on the line segment formed by the two points. When 0≤r≤1 and 0≤s≤1, it can be determined that the planned trajectory of the ego vehicle intersects with the predicted trajectory of the target vehicle.
[0180] 603. Determine a first target vehicle and a second target vehicle in the second target vehicle set.
[0181] After the second target vehicle set is determined, the first target vehicle having a direct game relationship with the ego vehicle and the second target vehicle having an impact on the game process of the ego vehicle can be determined in the second target vehicle set.
[0182] In a possible implementation, the first target vehicle in the second target vehicle set may be determined based on the difference between the TTC of the ego vehicle and the target vehicle and the lane where the target vehicle is located.
[0183] Exemplarily, first, the TTC corresponding to the self-vehicle and the TTC corresponding to the third target vehicle can be obtained, and the third target vehicle is a target vehicle in the second target vehicle set. The method of obtaining the TTC corresponding to the self-vehicle and the TTC corresponding to the third target vehicle can be as described in the above embodiment, and will not be repeated here. Then, the difference between the TTC corresponding to the self-vehicle and the TTC corresponding to the third target vehicle can be determined. When the difference between the TTC corresponding to the self-vehicle and the TTC corresponding to the third target vehicle is less than or equal to the preset threshold, and the third target vehicle is in the game lane set, the third target vehicle can be determined to be the first target vehicle; wherein the game lane set is a lane set that has a game relationship with the lane where the self-vehicle is located. In other words, when the time difference between the self-vehicle and the third target vehicle arriving at the trajectory conflict point is small, and the third target vehicle is located in the game lane set, the third target vehicle can be determined to be the first target vehicle.
[0184] For ease of understanding, the lanes involved in this embodiment are described in detail below. Figure 7 , Figure 7 A schematic diagram of a lane division method provided in an embodiment of the present application. Figure 7 As shown, the lanes defined in this embodiment have a certain length, and there is a connection relationship between the lanes. Figure 7 In , a1, a2, a3, b1, b2, b3, c1, c2 and c3 are all independent lanes, and a set of multiple lanes can be called a lane set.
[0185] When the vehicle is in different scenes, the vehicle can determine the corresponding game lane set according to the current scene. Figure 8 , Figure 8 A schematic diagram of a game lane division method provided in an embodiment of the present application. Figure 8 As shown in the figure, when the vehicle is in a driving scenario at an intersection without left-turn protection, the vehicle can determine that the lane that forms a game relationship with the lane where the vehicle is located is the game lane. Figure 8 In the example, lane f1 and lane f3 are both game lanes, that is, the vehicle can determine that the game lane set includes lane f1 and lane f3.
[0186] See also Fig. 9 , Fig. 9 A schematic diagram of another game lane division method provided in an embodiment of the present application. Fig. 9 As shown in Figure 1, when the vehicle is in the entrance and exit ramp driving scenario, the vehicle can also determine that the lane that forms a game relationship with the lane where the vehicle is located is the game lane. Fig. 9 In , the game lane set includes lane h1.
[0187] In a possible implementation, the second target vehicle in the second target vehicle set may be determined based on the lane in which the target vehicle is located.
[0188] Exemplarily, if the third target vehicle is in the risk lane set, and the trajectory lane set of the third target vehicle and the self-vehicle lane set have the same lane, then the third target vehicle is determined to be the second target vehicle. Among them, the third target vehicle is a target vehicle in the second target vehicle set. The self-vehicle lane set may refer to a lane set within a certain longitudinal range of the lane where the self-vehicle is located. The self-vehicle lane set may include, for example, the self-vehicle lane, the lane in front of the self-vehicle lane, and the lane behind the self-vehicle lane, and the self-vehicle lane is the lane where the self-vehicle is located. The risk lane set may refer to the lane where the vehicle that will have an impact on the self-vehicle is located during the self-vehicle game process. The risk lane set may include, for example, the self-vehicle lane set and the left lane set of the self-vehicle lane set and / or the right lane set of the self-vehicle lane set.
[0189] For example, Figure 8 As shown, in the driving scenario at the intersection without left-turn protection, the ego vehicle lane set includes lane d1, lane d2 and lane d3, the risk lane set includes lane d1, lane d2, lane d3, lane e1, lane e2 and lane e3, and the trajectory lane set includes lane f3 and lane f4.
[0190] like Fig. 9 As shown, in the entrance and exit ramp driving scenario, the ego vehicle lane set includes lane g1 and lane g2, the risk lane set includes lane g1, lane g2, lane i1 and lane i2, and the trajectory lane set includes lane h1 and lane g2.
[0191] In a possible embodiment, after the first target vehicle and the second target vehicle are identified, if there is a space gap between the first target vehicle or the second target vehicle and the vehicle, it can be considered that the identified first target vehicle and the second target vehicle do not actually have a game-related relationship. Therefore, this part of the identified target vehicles can be filtered out.
[0192] Exemplarily, for the first target vehicle identified, the predicted trajectory of the first target vehicle can be widened into a bendable rectangular frame; for the second target vehicle, the line connecting the second target vehicle and the vehicle can be widened to obtain a bendable rectangular frame. After obtaining the bendable rectangular frame, it can be determined whether the first target vehicle or the second target vehicle is separated by other target vehicles by judging whether there is a target vehicle covered by the bendable rectangular frame. Specifically, it can be determined whether the first target vehicle or the second target vehicle is separated by other target vehicles by traversing other target vehicles within the range of the bendable rectangular frame using formula (9).
[0193] See also Fig.10 , Fig.10 A schematic diagram of filtering out a first target vehicle and a second target vehicle provided in an embodiment of the present application. Fig.10 In the above example, target vehicle 1 and target vehicle 2 can be identified as the first target vehicle, and target vehicle 3 and target vehicle 4 can be identified as the second target vehicle. By widening the predicted trajectory of target vehicle 2 and obtaining a rectangular frame, it can be determined that target vehicle 1 is covered by the rectangular frame. Therefore, it can be determined that target vehicle 2 is separated by target vehicle 1, so that target vehicle 2 can be filtered out. In addition, by widening the line between the self-vehicle and the target vehicle 4 and obtaining a rectangular frame, it can be determined that target vehicle 3 is covered by the rectangular frame. Therefore, it can be determined that target vehicle 4 is separated by target vehicle 3, so that target vehicle 4 can be filtered out.
[0194] exist Figures 5 to 10 On the basis of the corresponding embodiments, in order to better implement the above solutions of the embodiments of the present application, the following also provides related devices for implementing the above solutions. Fig.11 , Fig.11 A schematic diagram of the structure of an automatic driving device 1100 provided in an embodiment of the present application. The automatic driving device 1100 may include an acquisition unit 1101, a processing unit 1102 and a control unit 1103, and the automatic driving device 1100 may also include an identification unit 1104.
[0195] In a possible implementation, the method may include: an acquisition unit 1101, configured to acquire first information of a first target vehicle and second information of a second target vehicle, wherein the 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;
[0196] A processing unit 1102 is configured to establish a dynamic game model based on at least the first information and the second information, wherein the first information is used to determine a conflict loss in the dynamic game model, and the second information is used to determine a safety distance loss in the dynamic game model;
[0197] The processing unit 1102 is further used to determine the driving strategy according to the dynamic game model;
[0198] The control unit 1103 is used to control the driving of the vehicle according to the driving strategy.
[0199] In a possible implementation, the device further includes: an identification unit 1104, configured to identify that the vehicle is in a preset scenario, wherein the preset scenario includes one of a driving scenario at an intersection without left-turn protection, an entrance and exit ramp driving scenario, or a lane merging driving scenario.
[0200] In one possible implementation, the processing unit 1102 is specifically used to: obtain the time TTC corresponding to the ego vehicle from the trajectory conflict point; determine the TTC corresponding to the first target vehicle based on the first information; and determine the conflict loss in the dynamic game model based on the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
[0201] In one possible implementation, the processing unit 1102 is specifically used to: determine the first safety distance loss or the second safety distance loss in the dynamic game model based on the position and speed of the vehicle and the second information; wherein the first safety distance loss is the front safety distance loss when executing the acceleration and overtaking strategy, and the second safety distance loss is the rear safety distance loss when executing the deceleration and yielding strategy.
[0202] In a possible implementation, the processing unit 1102 is further used to: determine whether the own vehicle has the right of way based on the own vehicle and the first target vehicle; determine the benefit of grabbing the right of way based on the own vehicle having the right of way, and the benefit of grabbing the right of way is used to establish a dynamic game model.
[0203] In one possible implementation, the processing unit 1102 is further used to: determine a first target vehicle set based on the position of the own vehicle, the first target vehicle set including multiple target vehicles whose distances from the own vehicle are less than a preset threshold; determine a second target vehicle set based on the planned trajectory of the own vehicle and the first target vehicle set, the second target vehicle set including multiple target vehicles whose predicted trajectories intersect with the planned trajectory of the own vehicle; and determine the first target vehicle and the second target vehicle in the second target vehicle set.
[0204] In one possible implementation, the acquisition unit 1101 is further used to acquire the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle, where the third target vehicle is a target vehicle in the second target vehicle set; the processing unit 1102 is further used to determine the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle; the processing unit 1102 is further used to determine that the third target vehicle is the first target vehicle if the difference between the TTC corresponding to the ego 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 lane set; wherein the game lane set is a lane set that has a game relationship with the lane in which the ego vehicle is located.
[0205] In one possible implementation, the processing unit 1102 is specifically used to: if the third target vehicle is in the risk lane set, and the trajectory lane set of the third target vehicle and the self-vehicle lane set have the same lane, then determine that the third target vehicle is the second target vehicle; wherein the third target vehicle is a target vehicle in the second target vehicle set, the self-vehicle lane set includes the self-vehicle lane, the lane in front of the self-vehicle lane, and the lane behind the self-vehicle lane, the self-vehicle lane is the lane where the self-vehicle is located, and the risk lane set includes the self-vehicle lane set and the left lane set of the self-vehicle lane set and / or the right lane set of the self-vehicle lane set.
[0206] It should be noted that the information interaction and execution process between the units in the automatic driving device 1100 are the same as those in the present application. Figures 5 to 10 The corresponding method embodiments are based on the same concept. For specific contents, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.
[0207] The present application also provides an automatic driving vehicle. Figure 1 For a description of Fig.12 , Fig.12 A schematic diagram of a structure of an autonomous driving vehicle provided in an embodiment of the present application, wherein the autonomous driving vehicle 100 may be deployed with Fig.10 The automatic driving device described in the corresponding embodiment is used to implement Figures 5 to 10 The functions of the autonomous driving vehicle in the corresponding embodiments. Since in some embodiments, the autonomous driving vehicle 100 may also include a communication function, the autonomous driving vehicle 100 may include Figure 1 The components shown in the figure may also include: 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 via a bus or other means.
[0208] The processor 113 controls the operation of the autonomous driving vehicle. In a specific application, the various components of the autonomous driving vehicle 100 are coupled together through a bus system, wherein the bus system may include a power bus, a control bus, and a status signal bus in addition to a data bus. However, for the sake of clarity, various buses are referred to as bus systems in the figure.
[0209] The receiver 1201 can be used to receive input digital or character information and generate signal input related to the relevant settings and function control of the autonomous driving vehicle. The transmitter 1202 can be used to output digital or character information through the first interface; the transmitter 1202 can also be used to send instructions to the disk group through the first interface to modify the data in the disk group; the transmitter 1202 can also include a display device such as a display screen.
[0210] In the embodiment of the present application, the application processor 1131 is used to execute Figure 5 The autonomous driving method performed by the autonomous driving vehicle in the corresponding embodiment. Specifically, the application processor 1131 is used to perform the following steps:
[0211] Obtain first information of a first target vehicle and second information of a second target vehicle, wherein there is a risk of collision between the self-vehicle and 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; establish a dynamic game model based on at least the first information and the second information, the first information is used to determine the conflict loss in the dynamic game model, and the second information is used to determine the safety distance loss in the dynamic game model; determine a driving strategy based on the dynamic game model; and control the driving of the self-vehicle based on the driving strategy.
[0212] In a possible implementation, it is recognized that the vehicle is in a preset scenario, where the preset scenario includes one of a driving scenario at an intersection without left-turn protection, a driving scenario at an entrance and exit ramp, or a driving scenario of a lane merging.
[0213] In one possible implementation, the time TTC corresponding to the ego vehicle from the trajectory conflict point is obtained; the TTC corresponding to the first target vehicle is determined according to the first information; and the conflict loss in the dynamic game model is determined according to the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
[0214] In one possible implementation, the first safety distance loss or the second safety distance loss in the dynamic game model is determined based on the position and speed of the vehicle and the second information; wherein the first safety distance loss is the front safety distance loss when executing the acceleration and overtaking strategy, and the second safety distance loss is the rear safety distance loss when executing the deceleration and yielding strategy.
[0215] In a possible implementation, based on the ego vehicle and the first target vehicle, it is determined that the ego vehicle has the right of way; based on the ego vehicle having the right of way, the benefit of grabbing the right of way is determined, and the benefit of grabbing the right of way is used to establish a dynamic game model.
[0216] In one possible implementation, a first target vehicle set is determined based on the position of the ego vehicle, the first target vehicle set including multiple target vehicles whose distances from the ego vehicle are less than a preset threshold; a second target vehicle set is determined based on the planned trajectory of the ego vehicle and the first target vehicle set, the second target vehicle set including multiple target vehicles whose predicted trajectories intersect with the planned trajectory of the ego vehicle; and the first target vehicle and the second target vehicle are determined in the second target vehicle set.
[0217] In one possible implementation, the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle are obtained, and the third target vehicle is a target vehicle in the second target vehicle set; the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle is determined; if the difference between the TTC corresponding to the ego 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 lane set, the third target vehicle is determined to be the first target vehicle; wherein the game lane set is a lane set that has a game relationship with the lane in which the ego vehicle is located.
[0218] In one possible implementation, if the third target vehicle is in the risk lane set, and the trajectory lane set of the third target vehicle and the ego vehicle lane set have the same lane, then the third target vehicle is determined to be the second target vehicle; wherein the third target vehicle is a target vehicle in the second target vehicle set, the ego vehicle lane set includes the ego lane, the lane in front of the ego lane, and the lane behind the ego lane, the ego lane is the lane where the ego vehicle is located, and the risk lane set includes the ego vehicle lane set and the left lane set of the ego vehicle lane set and / or the right lane set of the ego vehicle lane set.
[0219] It should be noted that for the specific implementation of the automatic driving method executed by the application processor 1131 and the beneficial effects thereof, reference can be made to Figures 5 to 10 The descriptions in the corresponding method embodiments will not be repeated here one by one.
[0220] In an embodiment of the present application, a computer-readable storage medium is also provided, wherein a program for executing autonomous driving is stored in the computer-readable storage medium, and when the program is executed on a computer, the computer executes the above-mentioned Figures 5 to 10 The illustrated embodiments describe the steps performed by an autonomous driving vehicle (or autonomous driving device) in the method.
[0221] The present application also provides a computer program product which, when executed on a computer, enables the computer to execute the above Figures 5 to 10 The illustrated embodiment describes the steps performed by an autonomous vehicle in the method.
[0222] The present application also provides a circuit system including a processing circuit, wherein the processing circuit is configured to execute the above Figures 5 to 10 The illustrated embodiment describes the steps performed by an autonomous vehicle in the method.
[0223] The autonomous driving device or autonomous driving vehicle provided in the embodiment of the present application may be a chip, which includes: a processing unit and a communication unit. The processing unit may be, for example, a processor, and the communication unit may be, for example, an input / output interface, a pin or a circuit. The processing unit may execute the computer execution instructions stored in the storage unit to enable the chip in the server to execute the above Figures 5 to 10 The autonomous driving method described in the illustrated embodiment. Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc. 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 other types of static storage devices that can store static information and instructions, a random access memory (RAM), etc.
[0224] For details, please refer to Fig.13 , Fig.13 A schematic diagram of the structure of a chip provided in an embodiment of the present application, the chip can be expressed as a neural network processor NPU 130, NPU 130 is mounted on the host CPU (Host CPU) as a coprocessor, and the host CPU assigns tasks. The core part of the NPU is the operation circuit 1303, which is controlled by the controller 1304 to extract matrix data from the memory and perform multiplication operations.
[0225] In some implementations, the operation circuit 1303 includes multiple processing units (Process Engine, PE) inside. In some implementations, the operation circuit 1303 is a two-dimensional systolic array. The operation circuit 1303 can also be a one-dimensional systolic array or other electronic circuits capable of performing mathematical operations such as multiplication and addition. In some implementations, the operation circuit 1303 is a general-purpose matrix processor.
[0226] For example, assume there is an input matrix A, a weight matrix B, and an output matrix C. The operation circuit takes the corresponding data of matrix B from the weight memory 1302 and caches it on each PE in the operation circuit. The operation circuit takes the matrix A data from the input memory 1301 and performs matrix operation with matrix B, and the partial result or final result of the matrix is stored in the accumulator 1308.
[0227] The unified memory 1306 is used to store input data and output data. The weight data is directly transferred to the weight memory 1302 through the direct memory access controller (DMAC) 1305. The input data is also transferred to the unified memory 1306 through the DMAC.
[0228] The bus interface unit (BIU) 1310 is used for the interaction between the AXI bus and the DMAC and the instruction fetch buffer (IFB) 1309 .
[0229] BIU 1310 is used for the instruction fetch memory 1309 to obtain instructions from the external memory, and is also used for the storage unit access controller 1305 to obtain the original data of the input matrix A or the weight matrix B from the external memory.
[0230] DMAC is mainly used to transfer input data in the external memory DDR to the unified memory 1306 or to transfer weight data to the weight memory 1302 or to transfer input data to the input memory 1301.
[0231] The vector calculation unit 1307 includes multiple operation processing units, and further processes the output of the operation circuit when necessary, such as vector multiplication, vector addition, exponential operation, logarithmic operation, size comparison, etc. It is mainly used for non-convolutional / fully connected layer network calculations in neural networks, such as batch normalization, pixel-level summation, upsampling of feature planes, etc.
[0232] In some implementations, the vector calculation unit 1307 can store the processed output vector to the unified memory 1306. For example, the vector calculation unit 1307 can apply a linear function and / or a nonlinear function to the output of the operation circuit 1303, such as linear interpolation of the feature plane extracted by the convolution layer, and then, for example, a vector of accumulated values to generate an activation value. In some implementations, the vector calculation unit 1307 generates a normalized value, a pixel-level summed value, or both. In some implementations, the processed output vector can be used as an activation input to the operation circuit 1303, for example, for use in a subsequent layer in a neural network.
[0233] The controller 1304 is connected to an instruction fetch buffer 1309 for storing instructions used by the controller 1304 .
[0234] Unified memory 1306, input memory 1301, weight memory 1302 and instruction fetch memory 1309 are all on-chip memories. External memories are private to the NPU hardware architecture.
[0235] Among them, the operations of each layer in the recurrent neural network can be performed by the operation circuit 1303 or the vector calculation unit 1307.
[0236] The processor mentioned in any of the above places may be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the above-mentioned first aspect method.
[0237] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.
[0238] Through the description of the above implementation mode, the technicians in the relevant field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CLUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.
[0239] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.
[0240] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrations. Available media can be magnetic media, (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid state disk (SSD)), etc.
Claims
1. An automatic driving method, characterized in that: include: Acquire first information of a first target vehicle and second information of a second target vehicle, wherein the self-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; Establishing a dynamic game model according to the first information and the second information, wherein the first information and the second information are used to determine the loss of the vehicle in the dynamic game model; The driving of the vehicle is controlled according to a driving strategy, and the driving strategy is determined by the dynamic game model.
2. The automatic driving method according to claim 1, characterized in that: The method further comprises: It is recognized that the vehicle is in a preset scenario, and the preset scenario includes: a driving scenario at an intersection without left-turn protection, an entrance and exit ramp driving scenario, and a lane merging driving scenario.
3. The automatic driving method according to claim 1 or 2, characterized in that: The self-vehicle can eliminate the risk of collision with the first target vehicle by accelerating to overtake or decelerating to yield.
4. The automatic driving method according to any one of claims 1 to 3, characterized in that: The ego vehicle loss refers to a measure of possible changes in one or both of the road passability or safety of the ego vehicle due to the game strategy selected by the ego vehicle during the game.
5. According to the automatic driving method according to any one of claims 1 to 4, the ego-vehicle loss includes conflict loss and safety distance loss.
6. The automatic driving method according to claim 5, characterized in that: The conflict loss indicates the loss of the ego vehicle when both the ego vehicle and the target vehicle choose to accelerate to overtake.
7. The automatic driving method according to claim 5, characterized in that: The safety distance loss indicates a loss of the ego vehicle when a spatial distance between the ego vehicle and the target vehicle changes.
8. The automatic driving method according to any one of claims 1 to 7, characterized in that: The establishing of a dynamic game model according to the first information and the second information includes: Obtain the time TTC corresponding to the ego vehicle to the trajectory conflict point; Determine a TTC corresponding to the first target vehicle according to the first information; The conflict loss in the dynamic game model is determined according to the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
9. The automatic driving method according to any one of claims 1 to 8, characterized in that: The establishing of a dynamic game model according to the first information and the second information includes: Determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and speed of the ego vehicle and the second information; Among them, the first safety distance loss is the front safety distance loss when executing the acceleration and overtaking strategy, and the second safety distance loss is the rear safety distance loss when executing the deceleration and yielding strategy.
10. The automatic driving method according to any one of claims 1 to 9, characterized in that: The method further comprises: Determining, based on the ego vehicle and the first target vehicle, that the ego vehicle has the right of way; According to the vehicle having the right of way, the profit of grabbing the right of way in the dynamic game model is determined.
11. The automatic driving method according to any one of claims 1 to 10, characterized in that: Before obtaining the first information of the first target vehicle and the second information of the second target vehicle, the method further includes: Determine a first target vehicle set according to the position of the ego vehicle, wherein the first target vehicle set includes a plurality of target vehicles whose distances from the ego vehicle are less than a preset threshold; Determine a second target vehicle set according to the planned trajectory of the ego vehicle and the first target vehicle set, wherein the second target vehicle set includes a plurality of target vehicles whose predicted trajectories intersect with the planned trajectory of the ego vehicle; The first target vehicle and the second target vehicle are determined in the second target vehicle set.
12. The automatic driving method according to claim 11, characterized in that: The step of determining the first target vehicle and the second target vehicle in the second target vehicle set includes: Obtaining 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; Determine a difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the third target vehicle; If the difference between the TTC corresponding to the self-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 third target vehicle is determined to be the first target vehicle; The game lane set is a lane set that has a game relationship with the lane where the vehicle is located.
13. The automatic driving method according to claim 11 or 12, characterized in that: The step of determining the first target vehicle and the second target vehicle in the second target vehicle set includes: If the third target vehicle is in the risk lane set, and the trajectory lane set of the third target vehicle and the lane set of the ego vehicle have the same lane, then the third target vehicle is determined to be the second target vehicle; Among them, the third target vehicle is a target vehicle in the second target vehicle set, the self-vehicle lane set includes the self-vehicle lane, the lane in front of the self-vehicle lane and the lane behind the self-vehicle lane, the self-vehicle lane is the lane where the self-vehicle is located, and the risk lane set includes the self-vehicle lane set and the left lane set of the self-vehicle lane set and / or the right lane set of the self-vehicle lane set.
14. An automatic driving device, characterized in that: include: An acquisition unit, configured to acquire first information of a first target vehicle and second information of a second target vehicle, wherein the self-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; A processing unit, configured to establish a dynamic game model according to the first information and the second information, wherein the first information and the second information are used to determine a loss of the ego vehicle in the dynamic game model; A control unit is used to control the driving of the vehicle according to a driving strategy, wherein the driving strategy is determined by the dynamic game model.
15. The automatic driving device according to claim 14, characterized in that: The device also includes: The recognition unit is used to recognize that the vehicle is in a preset scenario, wherein the preset scenario includes: a driving scenario at an intersection without left-turn protection, an entrance and exit ramp driving scenario, and a lane merging driving scenario.
16. The automatic driving device according to claim 14 or 15, characterized in that: The self-vehicle can eliminate the risk of collision with the first target vehicle by accelerating to overtake or decelerating to yield.
17. The automatic driving device according to any one of claims 14 to 16, characterized in that: The ego vehicle loss refers to a measure of possible changes in one or both of the road passability or safety of the ego vehicle due to the game strategy selected by the ego vehicle during the game.
18. According to any one of claims 14 to 17, the self-driving vehicle loss includes conflict loss and safety distance loss.
19. The automatic driving device according to claim 18, characterized in that: The conflict loss indicates the loss of the ego vehicle when both the ego vehicle and the target vehicle choose to accelerate to overtake.
20. The automatic driving device according to claim 18, characterized in that: The safety distance loss indicates a loss of the ego vehicle when a spatial distance between the ego vehicle and the target vehicle changes.
21. The automatic driving device according to any one of claims 14 to 20, characterized in that: The processing unit is specifically used for: Obtain the time TTC corresponding to the ego vehicle to the trajectory conflict point; Determine a TTC corresponding to the first target vehicle according to the first information; The conflict loss in the dynamic game model is determined according to the difference between the TTC corresponding to the ego vehicle and the TTC corresponding to the first target vehicle.
22. The automatic driving device according to any one of claims 14 to 21, characterized in that: The processing unit is specifically used for: Determining a first safety distance loss or a second safety distance loss in the dynamic game model according to the position and speed of the ego vehicle and the second information; Among them, the first safety distance loss is the front safety distance loss when executing the acceleration and overtaking strategy, and the second safety distance loss is the rear safety distance loss when executing the deceleration and yielding strategy.
23. The automatic driving device according to any one of claims 14 to 22, characterized in that: The processing unit is also used for: Determining, based on the ego vehicle and the first target vehicle, that the ego vehicle has the right of way; According to the vehicle having the right of way, the profit of grabbing the right of way is determined, and the profit of grabbing the right of way is used to establish the dynamic game model.
24. An automatic driving device, characterized in that: The method comprises a processor, wherein the processor is coupled to a memory, wherein the memory stores program instructions, and when the program instructions stored in the memory are executed by the processor, the method according to any one of claims 1 to 13 is implemented.
25. A computer-readable storage medium comprising a program, which, when executed on a computer, causes the computer to execute the method according to any one of claims 1 to 13.
26. An autonomous driving vehicle, characterized in that: The autonomous vehicle comprises processing circuitry and storage circuitry configured to perform the method of any one of claims 1 to 13.
27. A computer program product, characterized in that The computer program product comprises: a computer program code, and when the computer program code is executed by a processor, the method according to any one of claims 1 to 13 is implemented.
