Dynamic regulation constraint method, device and computer equipment for automatic driving

Abstract

The application relates to a dynamic regulation constraint method, device and computer equipment for automatic driving. The method comprises the following steps: acquiring self-vehicle behavior decision information of a vehicle, compliance detection information of the vehicle, environment information of a perceived environment of the vehicle and map information where the vehicle is located, and identifying relative position information of an interactive object which has a regulation constraint behavior with the vehicle based on the self-vehicle behavior decision information and the environment information; generating compliance judgment information of the interactive object through a compliance judgment strategy based on the relative position information of the interactive object, and constructing a violation driving area of the vehicle based on the compliance judgment information of the interactive object, the map information and the compliance detection information of the vehicle; and constructing a dynamic regulation element potential field of the vehicle based on the violation driving area of the vehicle, and performing constraint control on an automatic driving process of the vehicle based on the dynamic regulation element potential field. The method can effectively improve the constraint accuracy of the automatic driving state.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a method, apparatus, and computer device for dynamic regulatory constraints on autonomous driving. Background Technology

[0002] Autonomous driving technology is constantly developing and gradually being implemented. For a long time to come, autonomous driving will coexist with human driving in the existing road traffic system. A key technical issue that autonomous driving decision-making and planning needs to consider is its applicability to traffic regulations. These regulations contain many provisions that interact with other road users, such as those governing lane changing, overtaking, and yielding at intersections. Because these regulations are written for human drivers, their descriptions of the constraints are often vague. Therefore, improving the applicability and effectiveness of autonomous driving in complying with road traffic regulations is a current research focus.

[0003] Traditional autonomous driving decision-making and planning algorithms primarily consider ensuring safety and improving efficiency. Their consideration of road traffic regulations only involves complying with speed limits, not running red lights, and similar regulations that do not involve interaction with other participants. This makes it difficult to directly apply such vaguely defined interaction regulations to the autonomous driving state during actual autonomous driving control, resulting in poor accuracy in constraining the autonomous driving state. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for dynamic regulatory constraints in autonomous driving, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides a dynamic regulatory constraint method for autonomous driving, comprising:

[0006] The system acquires the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information where the vehicle is located. Based on the autonomous behavior decision information and the environmental information, it identifies the relative position information of the interactive objects that have legally constrained behaviors with the vehicle.

[0007] Based on the relative position information of the interactive object, compliance judgment information of the interactive object is generated through a compliance judgment strategy. Based on the compliance judgment information of the interactive object, the map information, and the vehicle's compliance detection information, the area where the vehicle violates driving regulations is constructed.

[0008] Based on the areas where the vehicle violates traffic regulations, a dynamic regulatory element potential field for the vehicle is constructed, and the autonomous driving process of the vehicle is constrained and controlled based on the dynamic regulatory element potential field.

[0009] Optionally, acquiring the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information where the vehicle is located includes:

[0010] The system receives vehicle interaction behavior instructions output by the decision module and arranges these instructions in the order of their generation time as vehicle behavior decision information.

[0011] By collecting sensor data from various sensors in the vehicle, the vehicle's location information, vehicle structure data, and road information of the environment in which the vehicle is located are recorded to obtain the environmental information of the environment perceived by the vehicle.

[0012] The system receives compliance threshold information between the vehicle and the vehicles interacting with the vehicle, output by the compliance detection module, and uses the compliance threshold information as the vehicle's compliance detection information.

[0013] The road route information of the road where the vehicle is located is collected, and the road route information is processed by lane marking to obtain the map information of the vehicle's location.

[0014] Optionally, the step of identifying the relative position information of interactive objects with legally constrained behaviors related to the vehicle based on the vehicle behavior decision information and the environmental information includes:

[0015] Based on the vehicle interaction behavior instructions of the vehicle, identify the target vehicle that has an interaction relationship with the vehicle, as well as the interaction mode between the target vehicle and the vehicle, and regard the target vehicle as the interaction object of the vehicle with legally constrained behavior;

[0016] Based on the interaction pattern between the interactive object and the vehicle, the relative position information between the vehicle and the interactive object is identified.

[0017] Optionally, the step of generating compliance determination information for the interactive object based on its relative position information and through a compliance determination strategy includes:

[0018] Based on the vehicle's location information, calculate the road vertex coordinates of the road where the vehicle is located;

[0019] Based on the location information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object, the coordinate information of the compliance determination starting point of the interactive object is filtered from the road vertex coordinate information of the road where the vehicle is located.

[0020] Based on the coordinate information of the compliance determination starting point corresponding to the interactive object, and the compliance determination starting point of the interactive object, the compliance determination direction corresponding to the interactive object is identified through the compliance direction determination strategy.

[0021] The compliance determination starting point of the interactive object and the corresponding compliance determination direction of the interactive object are used as the compliance determination information of the interactive object.

[0022] Optionally, constructing the area of ​​illegal driving of the vehicle based on the compliance determination information of the interactive object, the map information, and the compliance detection information of the vehicle includes:

[0023] Based on the compliance determination starting point of the interactive object and the map information, the initial illegal driving area of ​​the vehicle is constructed along the compliance determination direction using the illegal driving area construction strategy.

[0024] Based on the initial area of ​​violation of the vehicle's driving regulations, the area is adjusted using the vehicle's compliance detection information to obtain the area of ​​violation of the vehicle's driving regulations.

[0025] Optionally, constructing the dynamic regulatory element potential field of the vehicle based on the area where the vehicle violated traffic regulations includes:

[0026] Calculate the path information of the area where the violation occurred, and based on the path information and the vehicle's location information, identify the vehicle's current violation information through a vehicle violation determination strategy;

[0027] Based on the current violation information, the dynamic regulatory element potential field of the vehicle is identified through a dynamic regulatory element potential field identification strategy.

[0028] Secondly, this application also provides a dynamic regulatory constraint device for autonomous driving, comprising:

[0029] The acquisition module is used to acquire the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information of the environment perceived by the vehicle, and the map information of the vehicle's location, and based on the autonomous behavior decision information and the environmental information, to identify the relative position information of the interactive objects that have legally constrained behaviors with the vehicle.

[0030] The construction module is used to generate compliance judgment information of the interactive object based on the relative position information of the interactive object and through a compliance judgment strategy, and to construct the illegal driving area of ​​the vehicle based on the compliance judgment information of the interactive object, the map information, and the compliance detection information of the vehicle.

[0031] The constraint module is used to construct a dynamic regulatory element potential field for the vehicle based on the areas where the vehicle violates driving regulations, and to constrain and control the autonomous driving process of the vehicle based on the dynamic regulatory element potential field.

[0032] Optionally, the acquisition module is specifically used for:

[0033] The system receives vehicle interaction behavior instructions output by the decision module and arranges these instructions in the order of their generation time as vehicle behavior decision information.

[0034] By collecting sensor data from various sensors in the vehicle, the vehicle's location information, vehicle structure data, and road information of the environment in which the vehicle is located are recorded to obtain the environmental information of the environment perceived by the vehicle.

[0035] The system receives compliance threshold information between the vehicle and the vehicles interacting with the vehicle, output by the compliance detection module, and uses the compliance threshold information as the vehicle's compliance detection information.

[0036] The road route information of the road where the vehicle is located is collected, and the road route information is processed by lane marking to obtain the map information of the vehicle's location.

[0037] Optionally, the acquisition module is specifically used for:

[0038] Based on the vehicle interaction behavior instructions of the vehicle, identify the target vehicle that has an interaction relationship with the vehicle, as well as the interaction mode between the target vehicle and the vehicle, and regard the target vehicle as the interaction object of the vehicle with legally constrained behavior;

[0039] Based on the interaction pattern between the interactive object and the vehicle, the relative position information between the vehicle and the interactive object is identified.

[0040] Optionally, the building module is specifically used for:

[0041] Based on the vehicle's location information, calculate the road vertex coordinates of the road where the vehicle is located;

[0042] Based on the location information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object, the coordinate information of the compliance determination starting point of the interactive object is filtered from the road vertex coordinate information of the road where the vehicle is located.

[0043] Based on the coordinate information of the compliance determination starting point corresponding to the interactive object, and the compliance determination starting point of the interactive object, the compliance determination direction corresponding to the interactive object is identified through the compliance direction determination strategy.

[0044] The compliance determination starting point of the interactive object and the corresponding compliance determination direction of the interactive object are used as the compliance determination information of the interactive object.

[0045] Optionally, the building module is specifically used for:

[0046] Based on the compliance determination starting point of the interactive object and the map information, the initial illegal driving area of ​​the vehicle is constructed along the compliance determination direction using the illegal driving area construction strategy.

[0047] Based on the initial area of ​​violation of the vehicle's driving regulations, the area is adjusted using the vehicle's compliance detection information to obtain the area of ​​violation of the vehicle's driving regulations.

[0048] Optionally, the constraint module is specifically used for:

[0049] Calculate the path information of the area where the violation occurred, and based on the path information and the vehicle's location information, identify the vehicle's current violation information through a vehicle violation determination strategy;

[0050] Based on the current violation information, the dynamic regulatory element potential field of the vehicle is identified through a dynamic regulatory element potential field identification strategy.

[0051] Thirdly, this application provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any one of the first aspects.

[0052] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0053] Fifthly, this application provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0054] The aforementioned dynamic regulatory constraint method, apparatus, and computer equipment for autonomous driving acquire vehicle behavior decision information, vehicle compliance detection information, environmental information of the environment perceived by the vehicle, and map information of the vehicle's location. Based on the vehicle behavior decision information and the environmental information, it identifies the relative position information of interactive objects with which the vehicle has regulatory constraint behaviors. Based on the relative position information of the interactive objects, it generates compliance judgment information of the interactive objects through a compliance judgment strategy. Based on the compliance judgment information of the interactive objects, the map information, and the vehicle's compliance detection information, it constructs the vehicle's illegal driving area. Based on the vehicle's illegal driving area, it constructs a dynamic regulatory element potential field of the vehicle, and based on the dynamic regulatory element potential field, it constrains and controls the vehicle's autonomous driving process. This solution receives autonomous vehicle behavior decision-making information, compliance detection information, environmental information, and map information. Using the relative position information between the vehicle and the interacting object as the basis for judgment, it constructs a dynamic regulatory element potential field that accurately corresponds to the regulatory constraint area. This provides dynamic regulatory constraint guidance to the decision-making generation module of autonomous driving technology. It avoids the problem of traditional hard-constraint construction struggling to flexibly balance safety and compliance conflicts, and solves the problem of poor mapping accuracy between existing compliance potential fields and actual regulatory constraint areas. This effectively provides a more reasonable and effective compliance potential field for the autonomous driving decision-making generation module, thereby significantly improving the accuracy of constraints on the autonomous driving state. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating a dynamic regulatory constraint method for autonomous driving in one embodiment.

[0057] Figure 2 This is a schematic diagram of an interaction scenario between a vehicle and an interactive object in one embodiment;

[0058] Figure 3 This is a flowchart illustrating an example of dynamic regulatory constraints for autonomous driving in one embodiment.

[0059] Figure 4 This is a structural block diagram of a dynamic regulatory constraint device for autonomous driving in one embodiment;

[0060] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] The dynamic regulatory constraint method for autonomous driving provided in this application embodiment can be applied to a dynamic regulatory constraint system for autonomous driving. This system can be applied to a terminal, which can be, but is not limited to, various personal computers, laptops, mid-range computers, etc. The terminal receives vehicle behavior decision information, compliance detection information, environmental information, and map information. Using the relative position information between the vehicle and the interacting object as the basis for judgment, it constructs a dynamic regulatory element potential field that accurately corresponds to the regulatory constraint area, based on the regulations interacting with traffic participants. This provides dynamic regulatory constraint guidance to the decision-making generation module of autonomous driving technology, avoiding the problem of traditional hard-constraint construction struggling to flexibly balance safety and compliance conflicts. It also solves the problem of poor mapping accuracy between existing compliance potential fields and actual regulatory constraint areas. This effectively provides a more reasonable and effective compliance potential field substitution value for the decision-making generation module of autonomous driving, thereby effectively improving the accuracy of constraints on the autonomous driving state.

[0063] In one exemplary embodiment, such as Figure 1 As shown, a dynamic regulatory constraint method for autonomous driving is provided. Taking the application of this method to a terminal as an example, the method includes the following steps S101 to S103. Wherein:

[0064] Step S101: Obtain the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information where the vehicle is located. Based on the autonomous behavior decision information and the environmental information, identify the relative position information of the interactive objects with which the vehicle has legally constrained behavior.

[0065] In this embodiment, the terminal acquires the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information where the vehicle is located. The autonomous behavior decision information refers to the interaction behavior instructions generated by the vehicle's autonomous driving decision generation module, which interact with target vehicles. These interaction behavior instructions, Ego.Intention, include eight enumeration values ​​from 1 to 8. Enumeration values ​​1-7 represent interaction-related behavior instructions, namely following, lane changing, overtaking, meeting oncoming traffic, left turn, right turn, and U-turn (e.g., following is Ego.Intention==1). Enumeration value 8 refers to behaviors other than the above seven (Ego.Intention==8), collectively referred to as "others." The environmental information is the ordinate of other vehicles in the autonomous vehicle's coordinate system, with the vehicle as the origin. x-axis Heading angle Vehicle length Vehicle width and the lane ID The compliance detection information is the compliance threshold for the autonomous vehicle and the interactive target output by the compliance monitoring module. This map information includes the IDs of each lane. And the left and right lane alignment point array information, lane ID is The left and right lane alignment point arrays are respectively and The arrays of lane line points for the left and right lanes within the same lane must have the same length. Then, based on the vehicle's behavior decision information and environmental information, the terminal identifies the relative position information of interactive objects that are subject to regulatory constraints on the vehicle. This relative position information characterizes the relative position between the interactive object and the vehicle; for example, the left lane is in front of the vehicle; the right lane is behind the vehicle; the current lane is in front of the vehicle, etc. The specific identification process will be explained in detail later.

[0066] Step S102: Based on the relative position information of the interactive object, a compliance judgment information of the interactive object is generated through a compliance judgment strategy. Based on the compliance judgment information of the interactive object, map information, and vehicle compliance detection information, the area where the vehicle violates driving regulations is constructed.

[0067] In this embodiment, the terminal generates compliance judgment information for the interactive object based on its relative position information and a compliance judgment strategy. Then, based on this compliance judgment information, map information, and vehicle compliance detection information, it constructs the vehicle's illegal driving area. The compliance judgment strategy identifies the starting point and direction of compliance judgment for the vehicle's interactive object. This strategy involves constructing a rectangular box within the vehicle's two-dimensional road plane, identifying the vertices of the interactive object within the box, and then identifying the direction from the interactive object to the vehicle. The specific identification process will be explained in detail later. The illegal driving area is defined as the area where the vehicle will commit a traffic violation. The specific construction process will be explained in detail later.

[0068] Step S103: Based on the areas where the vehicle violates driving regulations, construct a dynamic regulatory element potential field for the vehicle, and constrain and control the autonomous driving process of the vehicle based on the dynamic regulatory element potential field.

[0069] In this embodiment, the terminal constructs a dynamic regulatory element potential field for the vehicle based on the areas where the vehicle violates traffic rules, and uses this dynamic regulatory element potential field to constrain and control the vehicle's autonomous driving process. This dynamic regulatory element potential field represents the distribution information of the violation potential field cost values ​​of the vehicle during its dynamic movement. The terminal transmits the vehicle's current position and the dynamic regulatory element potential field to the autonomous driving decision generation module, thereby regulating the downstream decision generation module to automatically adjust the vehicle's decision-making behavior based on the violation potential field cost values ​​when generating behavioral decisions.

[0070] Based on the above scheme, by receiving autonomous vehicle behavior decision information, compliance detection information, environmental information, and map information, and using the relative position information between the vehicle and the interactive object as the judgment basis, the regulations interacting with traffic participants are constructed into a dynamic regulatory element potential field that accurately corresponds to the regulatory constraint area. This provides dynamic regulatory constraint guidance to the decision generation module of autonomous driving technology, avoiding the problem of traditional hard constraint construction making it difficult to flexibly balance the conflict between safety and compliance. It also solves the problem of poor mapping accuracy between the existing compliance potential field and the actual regulatory constraint area. This can effectively provide a more reasonable and effective compliance potential field substitution value for the decision generation module of autonomous driving, thereby effectively improving the accuracy of constraints on the autonomous driving state.

[0071] Optionally, acquiring vehicle behavior decision information, vehicle compliance detection information, environmental information of the environment perceived by the vehicle, and map information of the vehicle's location includes: receiving vehicle interaction behavior commands output by the decision module, and arranging the vehicle interaction behavior commands in the order of command generation time as vehicle behavior decision information; recording the vehicle's position information, vehicle structure data, and road information of the environment in which the vehicle is located through sensor data collected by the vehicle's various sensors to obtain environmental information of the environment perceived by the vehicle; receiving compliance threshold information between the vehicle and the vehicles interacting with it output by the compliance detection module, and using the compliance threshold information as vehicle compliance detection information; collecting road route information of the road where the vehicle is located, and performing lane marking processing on the road route information to obtain map information of the vehicle's location.

[0072] In this embodiment, the terminal receives vehicle interaction behavior commands output by the decision module and arranges these commands in chronological order of their generation as vehicle behavior decision information. When only one interaction behavior command exists, no arranging is required. This command is represented by a number: when the command is 1 (Ego.Intention == 1), the vehicle intends to follow the car in front; when it is 2 (Ego.Intention == 2), the vehicle intends to change lanes; when it is 3 (Ego.Intention == 3), the vehicle intends to overtake; when it is 4 (Ego.Intention == 4), the vehicle intends to meet oncoming traffic; when it is 5 (Ego.Intention == 5), the vehicle intends to turn left; when it is 6 (Ego.Intention == 6), the vehicle intends to turn right; and when it is 7 (Ego.Intention == 7), the vehicle intends to make a U-turn.

[0073] By collecting sensor data from various sensors on the vehicle, the system records the location information of other vehicles around the vehicle, their vehicle structure data, and the road information of the surrounding environment, thus obtaining the environmental information perceived by the vehicle. Specifically, when recording the location information of other vehicles, the terminal uses the vehicle as the origin of a coordinate system to identify the coordinate information of other vehicles, thereby obtaining their location information.

[0074] Then, the terminal receives the compliance threshold information for the vehicle and the vehicles interacting with it from the compliance detection module, and uses this compliance threshold information as the vehicle's compliance detection information. Finally, the terminal collects the road route information of the road where the vehicle is located, and performs lane marking processing on the road route information to obtain the map information of the vehicle's location.

[0075] Based on the above scheme, by collecting information on the vehicle's autonomous behavior decision-making, compliance monitoring module, environmental perception, and map information, it is possible to improve the comprehensive analysis of the vehicle itself, the vehicles it interacts with, and the related information of the vehicle, thereby enhancing the comprehensiveness of the analysis of the actual environment in which the vehicle is located.

[0076] Optionally, based on vehicle behavior decision information and environmental information, the relative position information of the interactive object with which the vehicle has legally constrained behavior is identified, including: based on the vehicle's vehicle interaction behavior instructions, identifying the target vehicle with which the vehicle has an interactive relationship, and the interaction mode between the target vehicle and the vehicle, and taking the target vehicle as the interactive object with which the vehicle has legally constrained behavior; and based on the interaction mode between the interactive object and the vehicle, identifying the relative position information between the vehicle and the interactive object.

[0077] In this embodiment, the terminal identifies target vehicles with which it interacts, as well as the interaction patterns between these target vehicles, based on the vehicle's interaction behavior commands. The target vehicles are then designated as the interaction objects with which the vehicle exhibits legally regulated behavior. Next, the terminal identifies the relative position information between the vehicle and the interaction object based on the interaction pattern between the interaction object and the vehicle. Specifically, when the behavior command output by the autonomous vehicle behavior decision module is 1 (Ego.Intention == 1, indicating the autonomous vehicle intends to follow the vehicle ahead), the interaction object is the vehicle in the same lane as the autonomous vehicle. When the behavior instruction output by the autonomous vehicle behavior decision module is 2, that is... The vehicle intends to change lanes, and the interaction target is the vehicle in front of it in the same lane. and the vehicle behind in the target lane When the behavior instruction output by the autonomous vehicle behavior decision module is 3, that is... The vehicle intends to overtake another vehicle, and the interaction target is the vehicle in front of it in the same lane. and vehicles behind in the target lane When the behavior instruction output by the autonomous vehicle behavior decision module is 4, that is... The vehicle intends to meet oncoming traffic; the interaction target is the vehicle in the adjacent oncoming lane traveling in the opposite direction. When the behavior instruction output by the autonomous vehicle behavior decision module is 5, i.e., Ego.Intention == 5, the autonomous vehicle intends to turn left, and the interaction object is the oncoming straight-going vehicle. When the behavior instruction output by the autonomous vehicle behavior decision module is 6, i.e., Ego.Intention == 6, the autonomous vehicle intends to turn right, and the interaction object is the vehicle going straight on the left. and vehicles turning left from the opposite direction When the behavior instruction output by the autonomous vehicle behavior decision module is 7, that is, Ego.Intention==7, the autonomous vehicle intends to make a U-turn, and the interaction object is the oncoming straight-going vehicle. When the behavior instruction output by the vehicle behavior decision module is 1-7, proceed to step three; when the behavior instruction output by the vehicle behavior decision module is 8, that is, Ego.Intention==8, the vehicle behavior does not involve interaction or the vehicle has the highest right-of-way, and there is no interaction object at this time, so the method is directly abandoned.

[0078] Based on the above scheme, by combining different vehicle interaction commands and the location information of the interactive vehicle, the relative position information between interactive vehicles can be identified. This can efficiently and quickly locate the relative position information of interactive vehicles, improving the accuracy and efficiency of identifying the relative position information of interactive vehicles.

[0079] Optionally, based on the relative position information of the interactive object, compliance judgment information of the interactive object is generated through a compliance judgment strategy, including: calculating the road vertex coordinates of the road where the vehicle is located based on the vehicle's position information; filtering the coordinates of the starting point of the compliance judgment of the interactive object from the road vertex coordinates of the road where the vehicle is located based on the position information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object; identifying the compliance judgment direction corresponding to the interactive object through a compliance direction judgment strategy based on the coordinates of the starting point of the compliance judgment of the interactive object and the starting point of the compliance judgment of the interactive object; and using the starting point of the compliance judgment of the interactive object and the compliance judgment direction corresponding to the interactive object as the compliance judgment information of the interactive object.

[0080] In this embodiment, the terminal calculates the road vertex coordinates of the road where the vehicle is located based on the vehicle's location information. Specifically, the terminal views the interactive object as a rectangular box within the two-dimensional plane of the road, with its left and front...

[0081] , right front Left rear Right rear The coordinates of the four vertices are calculated as follows:

[0082]

[0083] Based on the vehicle's behavioral commands and the interaction objects, the specified interaction object endpoints are selected to form the starting point for compliance determination. .

[0084] Based on the location information between the vehicle and the interactive object, and the corresponding sub-interaction commands of the interactive object, the terminal filters the coordinates of the starting point of the compliance determination of the interactive object from the road vertex coordinates of the road where the vehicle is located. Then, based on the coordinates of the starting point of the compliance determination of the interactive object, and the starting point of the compliance determination of the interactive object, the terminal identifies the compliance determination direction corresponding to the interactive object through a compliance direction determination strategy. Finally, the terminal uses the starting point of the compliance determination of the interactive object and the corresponding compliance determination direction as the compliance determination information of the interactive object. Specifically:

[0085] When the behavior instruction output by the autonomous vehicle behavior decision module is 1, the vehicle in front in the same lane... The compliance determination begins at its backend, i.e. and The direction for determining compliance is the opposite of the direction of travel in the lane in which the vehicle is located.

[0086] When the behavior command output by the autonomous vehicle behavior decision module is 2 or 3, the vehicle in front in the same lane... The compliance determination begins at its backend, i.e. and The direction for compliance determination is the opposite of the direction of travel in its own lane; the vehicle behind in the target lane The compliance determination begins at its front end, i.e. and The compliance determination direction is the direction of travel in the lane in which the vehicle is located.

[0087] When the behavior instruction output by the vehicle behavior decision module is 4, vehicles in the adjacent oncoming lane traveling in the opposite direction to the vehicle's direction of travel... The compliance determination begins at its front end, i.e. and The compliance determination direction is the direction of travel in the lane in which the vehicle is located.

[0088] When the behavior instruction output by the autonomous vehicle behavior decision module is 5, the oncoming straight-going vehicle The compliance determination begins at its backend, i.e. and The compliance determination direction is the direction of travel in the lane in which the vehicle is located.

[0089] When the behavior command output by the autonomous vehicle behavior decision module is 6, vehicles traveling straight on the left... and vehicles turning left from the opposite direction The starting point for compliance determination is its front end, that is... ,as well as The compliance determination direction is the direction of travel in the lane in which the vehicle is located.

[0090] When the behavior instruction output by the autonomous vehicle behavior decision module is 7, oncoming straight-going vehicles The compliance determination begins at its front end, i.e. The compliance determination direction is the direction of travel in the lane in which the vehicle is located.

[0091] Based on the above scheme, by using different behavioral commands and rectangular boxes within the constructed two-dimensional road plane, the starting point for compliance determination of interactive objects and the corresponding compliance determination direction of interactive objects can be identified, thereby effectively improving recognition efficiency while ensuring recognition accuracy.

[0092] Optionally, based on the compliance judgment information of the interactive object, map information, and vehicle compliance detection information, the vehicle's illegal driving area is constructed, including: based on the compliance judgment starting point of the interactive object and map information, along the compliance judgment direction, and through an illegal driving area construction strategy, an initial illegal driving area of ​​the vehicle is constructed; based on the initial illegal driving area of ​​the vehicle, the area is adjusted using the vehicle's compliance detection information to obtain the vehicle's illegal driving area.

[0093] In this embodiment, the terminal constructs the initial illegal driving area of ​​the vehicle based on the compliance determination starting point of the interactive object and map information, along the compliance determination direction, using an illegal driving area construction strategy. Specifically, the terminal, based on the compliance determination starting point and compliance determination direction of the interactive object, starts from the two endpoints of the compliance determination starting point of the interactive object and proceeds along the compliance determination direction, starting from the lane where the interactive object is located.

[0094] Actual or virtual lane alignment points and As a horizontal boundary, with compliance threshold As a vertical boundary, a zone for illegal driving is constructed, as shown in the diagram below. Figure 2 As shown.

[0095] Then, based on the vehicle's initial violation area, the terminal performs area adjustment processing using the vehicle's compliance detection information to obtain the vehicle's violation area. The adjustment method is as follows: Figure 2 As shown: the starting points of the left and right lane lines in the lateral boundary. They are respectively taken from the next shape point along the compliance determination direction from the starting point of the compliance determination. The indexes in are all End points of left and right lane lines They are respectively taken from the next point on the compliance threshold distance extended along the compliance determination direction from the starting point of the compliance determination. The indexes in are all The extracted array of left and right lane alignment points. as follows:

[0096]

[0097] Ultimately, the compliance determination will begin at the starting point. and the extracted array of left and right lane alignment points Connecting them sequentially forms a closed polygon, which is the illegal driving area (IDA).

[0098]

[0099] in, Flip it.

[0100] Based on the above scheme, by combining compliance judgment thresholds, the illegal driving area is optimized and adjusted to ensure that the adjusted illegal driving area is dynamically applicable to the current road conditions of the vehicle, thus ensuring the accuracy of identifying the vehicle's dynamic illegal driving area.

[0101] Optionally, based on the area where the vehicle violates traffic rules, a dynamic regulatory element potential field for the vehicle is constructed, including: calculating the path information of the area where the vehicle violates traffic rules, and based on the path information and the vehicle's location information, identifying the vehicle's current violation information through a vehicle violation judgment strategy; and based on the current violation information, identifying the vehicle's dynamic regulatory element potential field through a dynamic regulatory element potential field identification strategy.

[0102] In this embodiment, the terminal calculates the path information of the illegal driving area, and based on the path information and the vehicle's location information, identifies the vehicle's current violation information through a vehicle violation determination strategy. The path information is used to characterize the location range of the vehicle and the illegal driving area within the road where the vehicle is located, and to determine whether the vehicle is within the illegal driving area. Specifically:

[0103] The terminal uses the extracted left and right lane alignment point arrays Calculate the Frenet path in the area of ​​illegal driving. :

[0104]

[0105] The length of the Frenet path in the area of ​​the traffic violation In the equation, the sum of the lengths of the line segments formed by two adjacent points is:

[0106]

[0107] Where n is Array length, Return vector The model.

[0108] Finally, based on the current violation information, the terminal identifies the vehicle's dynamic regulatory element potential field using a dynamic regulatory element potential field recognition strategy. This dynamic regulatory element potential field recognition strategy is specifically as follows:

[0109] Determine if the vehicle's origin (0,0) is within the illegal driving area. If the vehicle's origin is within the illegal driving area, then the distance of the vehicle from the illegal driving area is... The value is 0, and the origin of the vehicle is at... The projection point on the edge The distance from the starting point of the compliance determination of the interactive object is (like Figure 2 As shown), the Frenet normalized distance n is calculated according to the following formula. :

[0110]

[0111] If the vehicle's origin point is outside the illegal driving area, the distance from the vehicle to the illegal driving area will be calculated based on the nearest point principle. The nearest point in the area of ​​illegal driving is , exist The projection point on the edge The distance from the starting point of the compliance determination of the interaction object is Then, calculate the Frenet normalized distance n according to the above formula. .

[0112] Then, the dynamic regulatory element potential field at the vehicle's location is calculated according to the following formula:

[0113]

[0114] Where h is the potential field strength parameter; a and b are the potential field shape parameters, used to adjust the boundary smoothness; The activation function is calculated using the following formula: .

[0115] Based on the above scheme, the current regulatory element potential field of the vehicle can be generated efficiently, accurately and dynamically, improving the accuracy of dynamically obtaining the value of the vehicle's violation potential field.

[0116] This application also provides an example of dynamic regulatory constraints for autonomous driving, such as... Figure 3 As shown, the specific processing procedure includes the following steps:

[0117] Step S301: Receive the vehicle interaction behavior instructions output by the decision module, and arrange the vehicle interaction behavior instructions in the order of instruction generation time as the vehicle behavior decision information.

[0118] Step S302: The vehicle's location information, vehicle structure data, and road information of the vehicle's environment are recorded by the sensor data collected by the vehicle's various sensors, thereby obtaining the environmental information of the environment perceived by the vehicle.

[0119] Step S303: Receive the compliance threshold information of the vehicle and the vehicles interacting with the vehicle output by the compliance detection module, and use the compliance threshold information as the vehicle's compliance detection information.

[0120] Step S304: Collect road route information of the road where the vehicle is located, and perform lane marking processing on the road route information to obtain map information of the vehicle's location.

[0121] Step S305: Based on the vehicle's vehicle interaction behavior instructions, identify the target vehicle that has an interaction relationship with the vehicle, as well as the interaction mode between the target vehicle and the vehicle, and regard the target vehicle as the interaction object of the vehicle's legally constrained behavior.

[0122] Step S306: Based on the interaction mode between the interactive object and the vehicle, identify the relative position information between the vehicle and the interactive object.

[0123] Step S307: Based on the vehicle's location information, calculate the coordinates of the road vertices where the vehicle is located.

[0124] Step S308: Based on the location information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object, filter the coordinate information of the starting point of the compliance determination of the interactive object from the road vertex coordinate information of the road where the vehicle is located.

[0125] Step S309: Based on the coordinate information of the compliance judgment starting point corresponding to the interactive object and the compliance judgment starting point of the interactive object, the compliance judgment direction corresponding to the interactive object is identified through the compliance direction judgment strategy.

[0126] Step S310: The compliance determination starting point of the interactive object and the corresponding compliance determination direction of the interactive object are used as the compliance determination information of the interactive object.

[0127] Step S311: Based on the compliance determination starting point of the interactive object and the map information, construct the initial illegal driving area of ​​the vehicle along the compliance determination direction through the illegal driving area construction strategy.

[0128] Step S312: Based on the initial illegal driving area of ​​the vehicle, the area is adjusted using the vehicle's compliance detection information to obtain the illegal driving area of ​​the vehicle.

[0129] Step S313: Calculate the path information of the area where the vehicle violated traffic rules, and based on the path information and the vehicle's location information, identify the vehicle's current violation information through a vehicle violation judgment strategy.

[0130] Step S314: Based on the current violation information, identify the dynamic regulatory element potential field of the vehicle through a dynamic regulatory element potential field identification strategy.

[0131] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0132] Based on the same inventive concept, this application also provides a dynamic regulatory constraint device for autonomous driving to implement the dynamic regulatory constraint method for autonomous driving described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more embodiments of the dynamic regulatory constraint device for autonomous driving provided below can be found in the limitations of the dynamic regulatory constraint method for autonomous driving described above, and will not be repeated here.

[0133] In one exemplary embodiment, such as Figure 4 As shown, a dynamic regulatory constraint device for autonomous driving is provided, comprising: an acquisition module 410, a construction module 420, and a constraint module 430, wherein:

[0134] The acquisition module 410 is used to acquire the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information of the environment perceived by the vehicle, and the map information of the vehicle's location, and based on the autonomous behavior decision information and the environmental information, to identify the relative position information of the interactive object with which the vehicle has legally constrained behavior.

[0135] The construction module 420 is used to generate compliance judgment information of the interactive object based on the relative position information of the interactive object and through a compliance judgment strategy, and to construct the illegal driving area of ​​the vehicle based on the compliance judgment information of the interactive object, the map information, and the compliance detection information of the vehicle.

[0136] The constraint module 430 is used to construct a dynamic regulatory element potential field of the vehicle based on the area where the vehicle violates driving regulations, and to constrain and control the autonomous driving process of the vehicle based on the dynamic regulatory element potential field.

[0137] Optionally, the acquisition module 410 is specifically used for:

[0138] The system receives vehicle interaction behavior instructions output by the decision module and arranges these instructions in the order of their generation time as vehicle behavior decision information.

[0139] By collecting sensor data from various sensors in the vehicle, the vehicle's location information, vehicle structure data, and road information of the environment in which the vehicle is located are recorded to obtain the environmental information of the environment perceived by the vehicle.

[0140] The system receives compliance threshold information between the vehicle and the vehicles interacting with the vehicle, output by the compliance detection module, and uses the compliance threshold information as the vehicle's compliance detection information.

[0141] The road route information of the road where the vehicle is located is collected, and the road route information is processed by lane marking to obtain the map information of the vehicle's location.

[0142] Optionally, the acquisition module 410 is specifically used for:

[0143] Based on the vehicle interaction behavior instructions of the vehicle, identify the target vehicle that has an interaction relationship with the vehicle, as well as the interaction mode between the target vehicle and the vehicle, and regard the target vehicle as the interaction object of the vehicle with legally constrained behavior;

[0144] Based on the interaction pattern between the interactive object and the vehicle, the relative position information between the vehicle and the interactive object is identified.

[0145] Optionally, the building module 420 is specifically used for:

[0146] Based on the vehicle's location information, calculate the road vertex coordinates of the road where the vehicle is located;

[0147] Based on the location information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object, the coordinate information of the compliance determination starting point of the interactive object is filtered from the road vertex coordinate information of the road where the vehicle is located.

[0148] Based on the coordinate information of the compliance determination starting point corresponding to the interactive object, and the compliance determination starting point of the interactive object, the compliance determination direction corresponding to the interactive object is identified through the compliance direction determination strategy.

[0149] The compliance determination starting point of the interactive object and the corresponding compliance determination direction of the interactive object are used as the compliance determination information of the interactive object.

[0150] Optionally, the building module 420 is specifically used for:

[0151] Based on the compliance determination starting point of the interactive object and the map information, the initial illegal driving area of ​​the vehicle is constructed along the compliance determination direction using the illegal driving area construction strategy.

[0152] Based on the initial area of ​​violation of the vehicle's driving regulations, the area is adjusted using the vehicle's compliance detection information to obtain the area of ​​violation of the vehicle's driving regulations.

[0153] Optionally, the constraint module 430 is specifically used for:

[0154] Calculate the path information of the area where the violation occurred, and based on the path information and the vehicle's location information, identify the vehicle's current violation information through a vehicle violation determination strategy;

[0155] Based on the current violation information, the dynamic regulatory element potential field of the vehicle is identified through a dynamic regulatory element potential field identification strategy.

[0156] The modules in the aforementioned dynamic regulatory constraint device for autonomous driving can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can invoke and execute the corresponding operations of each module.

[0157] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a dynamic regulatory constraint method for autonomous driving. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0158] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0159] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program that, when executed by the processor, implements steps of a dynamic regulatory constraint method for autonomous driving.

[0160] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements steps of a dynamic regulatory constraint method for autonomous driving.

[0161] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements steps of a dynamic regulatory constraint method for autonomous driving.

[0162] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0163] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0164] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0165] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A dynamic regulation constraint method for automatic driving, characterized in that, The method includes: The system acquires the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information where the vehicle is located. Based on the autonomous behavior decision information and the environmental information, it identifies the relative position information of the interactive objects that have legally constrained behaviors with the vehicle. Based on the relative position information of the interactive object, compliance judgment information of the interactive object is generated through a compliance judgment strategy. Based on the compliance judgment information of the interactive object, the map information, and the vehicle's compliance detection information, the area where the vehicle violates driving regulations is constructed. Based on the areas where the vehicle violates traffic regulations, a dynamic regulatory element potential field for the vehicle is constructed, and the autonomous driving process of the vehicle is constrained and controlled based on the dynamic regulatory element potential field.

2. The method according to claim 1, characterized in that, The acquisition of the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information perceived by the vehicle, and the map information of the vehicle's location includes: The system receives vehicle interaction behavior instructions output by the decision module and arranges these instructions in the order of their generation time as vehicle behavior decision information. By collecting sensor data from various sensors in the vehicle, the vehicle's location information, vehicle structure data, and road information of the environment in which the vehicle is located are recorded to obtain the environmental information of the environment perceived by the vehicle. The system receives compliance threshold information between the vehicle and the vehicles interacting with the vehicle, output by the compliance detection module, and uses the compliance threshold information as the vehicle's compliance detection information. The road route information of the road where the vehicle is located is collected, and the road route information is processed by lane marking to obtain the map information of the vehicle's location.

3. The method according to claim 2, characterized in that, The step of identifying the relative position information of interactive objects with legally constrained behaviors related to the vehicle based on the vehicle behavior decision information and the environmental information includes: Based on the vehicle interaction behavior instructions of the vehicle, identify the target vehicle that has an interaction relationship with the vehicle, as well as the interaction mode between the target vehicle and the vehicle, and regard the target vehicle as the interaction object of the vehicle with legally constrained behavior; Based on the interaction pattern between the interactive object and the vehicle, the relative position information between the vehicle and the interactive object is identified.

4. The method according to claim 3, characterized in that, The process of generating compliance determination information for the interactive object based on its relative position information and through a compliance determination strategy includes: Based on the vehicle's location information, calculate the road vertex coordinates of the road where the vehicle is located; Based on the location information between the vehicle and the interactive object, and the sub-interaction behavior instructions corresponding to the interactive object, the coordinate information of the compliance determination starting point of the interactive object is filtered from the road vertex coordinate information of the road where the vehicle is located. Based on the coordinate information of the compliance determination starting point corresponding to the interactive object, and the compliance determination starting point of the interactive object, the compliance determination direction corresponding to the interactive object is identified through the compliance direction determination strategy. The compliance determination starting point of the interactive object and the corresponding compliance determination direction of the interactive object are used as the compliance determination information of the interactive object.

5. The method according to claim 4, characterized in that, The process of constructing the vehicle's illegal driving area based on the compliance determination information of the interactive object, the map information, and the vehicle's compliance detection information includes: Based on the compliance determination starting point of the interactive object and the map information, the initial illegal driving area of ​​the vehicle is constructed along the compliance determination direction using the illegal driving area construction strategy. Based on the initial area of ​​violation of the vehicle's driving regulations, the area is adjusted using the vehicle's compliance detection information to obtain the area of ​​violation of the vehicle's driving regulations.

6. The method according to claim 1, characterized in that, The process of constructing a dynamic regulatory element potential field for the vehicle based on the area where the vehicle violated traffic regulations includes: Calculate the path information of the area where the violation occurred, and based on the path information and the vehicle's location information, identify the vehicle's current violation information through a vehicle violation determination strategy; Based on the current violation information, the dynamic regulatory element potential field of the vehicle is identified through a dynamic regulatory element potential field identification strategy.

7. A dynamic regulatory constraint device for autonomous driving, characterized in that, The device includes: The acquisition module is used to acquire the vehicle's autonomous behavior decision information, the vehicle's compliance detection information, the environmental information of the environment perceived by the vehicle, and the map information of the vehicle's location, and based on the autonomous behavior decision information and the environmental information, to identify the relative position information of the interactive objects that have legally constrained behaviors with the vehicle. The construction module is used to generate compliance judgment information of the interactive object based on the relative position information of the interactive object and through a compliance judgment strategy, and to construct the illegal driving area of ​​the vehicle based on the compliance judgment information of the interactive object, the map information, and the compliance detection information of the vehicle. The constraint module is used to construct a dynamic regulatory element potential field for the vehicle based on the areas where the vehicle violates driving regulations, and to constrain and control the autonomous driving process of the vehicle based on the dynamic regulatory element potential field.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images