Method for determining incidence relation between road elements, related system and storage medium

Abstract

The embodiment of the invention provides a method for determining an incidence relation between road elements, a related system and a storage medium. The method comprises the following steps: acquiring road perception element information and road prior element information; processing the road perception element information and the road prior element information to obtain a first feature vector and a second feature vector; and obtaining an association relationship between the road perception elements and / or an association relationship between the road perception elements and the road prior elements according to the first feature vector and the second feature vector. According to the means, the current road element is associated with the road prior information, and the road prior information can be used as a bridge, so that the level of relation reasoning of the current road element is improved. Meanwhile, by reasoning the relation between the current road element and the road prior, the redundancy relation can be constructed, the current road element relation can be conveniently checked, and the robustness and interpretability of the whole scheme are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a method for determining the association relationship between road elements and a related system and storage medium. Background Art

[0002] Autonomous driving requires road structure information to plan and control the vehicle and assist in the prediction of road participants. Traditionally, road structure information is stored in high-precision maps. In addition to the geometric information of lanes and roads, road structure information also includes multiple associations between lanes and traffic lights, road signs, zebra crossings and other lanes. When the vehicle is driving, it is necessary to determine the rules that the vehicle needs to follow by combining the association between lanes and other rule-based information. For example, lane speed limits and turn information. For another example, which traffic light should be compared to the lane you are in? The left-turn lane may require viewing both the straight-ahead light and the left-turn light at the same time.

[0003] Traditionally, such associations can be obtained through high-precision maps. However, the production of high-precision maps is relatively complicated. First, the data collection of high-precision maps relies on relatively expensive collection equipment; second, the production process of high-precision maps is relatively complicated, and in many cases a large amount of manual participation is still required. This has led to the current slow production of high-precision maps, extremely limited coverage areas, and untimely updates to changes in road scenes. In addition, in the process of autonomous driving, the use of high-precision maps requires high-precision positioning information. When the positioning module of autonomous driving has accidental deviations, it will cause errors in loading high-precision maps, leading to serious accidents. The limited status of high-precision maps limits the application scope of high-level autonomous driving. On the one hand, this has promoted the development of high-precision map production, and on the other hand, it has promoted the development of real-time road structure cognition.

[0004] Road structure recognition can greatly expand the application boundaries of high-level autonomous driving by reconstructing the road conditions around the vehicle through real-time analysis of the collected data from on-board sensors. Currently, many solutions only focus on the reconstruction of geometric information on the road. In the past two years, some work has begun to focus on analyzing lane topology, but very few works have noticed that it is also necessary to analyze the relationship between various elements on the road. In addition, due to the limited detection accuracy of geometric information, it is difficult to reason about the relationship between road elements. Summary of the invention

[0005] The present application discloses a method for determining the association relationship between road elements and a related system and storage medium, which can improve the current level of road element relationship reasoning.

[0006] In a first aspect, an embodiment of the present application provides a method for determining an association relationship between road elements, and the method may include:

[0007] Acquire road perception element information and road prior element information;

[0008] Processing the road perception element information and the road prior element information to obtain a first eigenvector and a second eigenvector;

[0009] The association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements is obtained according to the first feature vector and the second feature vector.

[0010] In an embodiment of the present application, the first eigenvector and the second eigenvector are obtained by processing the road perception element information and the road prior element information, and then the correlation relationship between the road perception elements and / or the correlation relationship between the road perception elements and the road prior elements are obtained based on the first eigenvector and the second eigenvector. This method, by associating the current road element with the road prior information, can use the road prior information as a bridge to improve the level of reasoning about the current road element relationship. At the same time, by reasoning about the relationship between the current road element and the road prior, a redundant relationship can be constructed, which is convenient for checking the current road element relationship and improving the robustness and interpretability of the overall solution.

[0011] This solution does not rely heavily on road prior information for relationship reasoning. According to needs, the final output can be only the association of current road elements, avoiding negative impacts caused by outdated road prior information.

[0012] The road perception element information can be understood as real-time information of elements such as roads. The road perception element information can include one or more of the element information such as lanes, roads, traffic lights, road signs, zebra crossings, road boundaries, intersections, signs, etc., and of course, it can also be other element information, which is not limited in this solution.

[0013] The road prior element information can be understood as the known information of elements such as roads.

[0014] In a possible implementation, the vehicle may obtain road perception element information from roadside devices, vehicle cameras and / or radar sensors. The vehicle may obtain road prior element information from user input or maps (e.g., standard definition maps, navigation maps (SD maps)). Of course, other methods are possible, and this solution does not limit this.

[0015] For example, the vehicle obtains road perception element information from the perception module and road prior element information from the navigation map.

[0016] In a possible implementation, the processing of the road perception element information and the road priori element information to obtain a first feature vector and a second feature vector includes:

[0017] Feature extraction is performed on the road perception element information and the road prior element information respectively to obtain the first feature vector and the second feature vector, wherein the first feature vector is a road perception element feature vector, and the second feature vector is a road prior element feature vector.

[0018] For example, feature extraction is performed on the road perception element information to obtain a road perception element feature vector. Feature extraction is performed on the road prior element information to obtain a road prior element feature vector.

[0019] In another possible implementation, the processing of the road perception element information and the road a priori element information to obtain a first feature vector and a second feature vector includes:

[0020] The road perception element information and the road prior element information are subjected to feature interaction processing to obtain the first feature vector and the second feature vector, wherein the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

[0021] The fused road perception element feature vector can be understood as a road perception element feature vector obtained by implicitly considering the influence of road prior element information. Of course, it can also include the fusion of road perception element information, etc. This solution does not limit this.

[0022] Accordingly, the fused road priori element feature vector can be understood as a road priori element feature vector obtained by implicitly considering the influence of the road perception element information. Of course, it can also include the fusion of road priori element information.

[0023] The feature interaction processing can be understood as referring to the road priori element information while extracting the features of the road perception element information. Correspondingly, referring to the road perception element information while extracting the features of the road priori element information.

[0024] In a possible implementation, performing feature interaction processing on the road perception element information and the road prior element information to obtain the first feature vector and the second feature vector includes:

[0025] The road perception element information and the road prior element information are processed by a multi-head attention mechanism to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0026] In another possible implementation, performing feature interaction processing on the road perception element information and the road priori element information to obtain the first feature vector and the second feature vector includes:

[0027] According to at least one of the position information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information, elements whose distance is less than a preset distance are interactively fused to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0028] For example, the road surface marking features and lane features are interactively integrated to obtain a lane description with marking information. Another example is to merge lanes in the same direction to form a road-level feature description.

[0029] The distance can be understood as the degree of relevance indicated by the information, rather than just the traditional geographical distance, etc. This example uses distance as the selection condition, and of course other conditions can also be used, and this solution does not limit this.

[0030] This solution can be implemented through rule constraints, extracting key features of road elements through manually defined rules, constructing matching strategies to calculate association relationships, and completing the verification and output of the final relationship through rule transmission methods.

[0031] In a possible implementation, obtaining the association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements according to the first feature vector and the second feature vector includes:

[0032] Constructing a matching cost function according to the fused road perception element feature vector and the fused road prior element feature vector;

[0033] The association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements is obtained according to the matching cost function.

[0034] For example, the matching cost function can be a function constrained by distance, orientation, category, etc., and this solution does not impose any restrictions on this. Among them, if the constructed association relationship constraints are met, there is a relationship between the elements. If the constructed association relationship constraints are not met, there is no relationship between the elements. For example, combining the various road perception element information, road prior element information and fused road perception element feature vectors, fused road prior element feature vectors, a matching cost function is constructed, and the association between the perception elements and between the perception elements and the prior information is determined through the matching cost function. For example, in order for there to be an association between the perceived lane and the SD Map road, the geometric positions of the two need to be close and the driving directions need to be similar. For another example, in order for there to be an association between the perceived lane and the perceived traffic light, the direction of travel of the perceived lane needs to be toward the location of the traffic light, and the traffic light needs to be toward the lane.

[0035] In a possible implementation, the method further includes:

[0036] Merging and / or deleting redundant association relationships between the road perception elements to obtain processed association relationships between the road perception elements;

[0037] The redundant association relationship between the road perception element and the road a priori element is merged and / or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

[0038] This can simplify the association relationship between road perception elements and the association relationship between road perception elements and road prior elements, making the obtained association relationship results clearer, more concise and accurate.

[0039] In a second aspect, an embodiment of the present application provides a device for determining an association relationship between road elements, including:

[0040] An acquisition module, used to acquire road perception element information and road prior element information;

[0041] A processing module, used for processing the road perception element information and the road prior element information to obtain a first eigenvector and a second eigenvector;

[0042] A determination module is used to obtain the association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements according to the first feature vector and the second feature vector.

[0043] In a possible implementation, the processing module is used to:

[0044] Feature extraction is performed on the road perception element information and the road prior element information respectively to obtain the first feature vector and the second feature vector, wherein the first feature vector is a road perception element feature vector, and the second feature vector is a road prior element feature vector.

[0045] In another possible implementation, the processing module is used to:

[0046] The road perception element information and the road prior element information are subjected to feature interaction processing to obtain the first feature vector and the second feature vector, wherein the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

[0047] In a possible implementation, the processing module is further configured to:

[0048] The road perception element information and the road prior element information are processed by a multi-head attention mechanism to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0049] In another possible implementation, the processing module is further configured to:

[0050] According to at least one of the position information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information, elements whose distance is less than a preset distance are interactively fused to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0051] In a possible implementation, the processing module is further configured to:

[0052] Constructing a matching cost function according to the fused road perception element feature vector and the fused road prior element feature vector;

[0053] The association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements is obtained according to the matching cost function.

[0054] In a possible implementation manner, the determining module is further configured to:

[0055] Merging and / or deleting redundant association relationships between the road perception elements to obtain processed association relationships between the road perception elements;

[0056] The redundant association relationship between the road perception element and the road a priori element is merged and / or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

[0057] In a third aspect, an embodiment of the present application provides a device for determining the association relationship between road elements, comprising a processor and a memory; wherein the memory is used to store program code, and the processor is used to call the program code to execute a method provided in any possible implementation manner of the first aspect.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a method provided in any possible implementation manner of the first aspect.

[0059] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on a computer, the computer executes a method provided in any possible implementation manner of the first aspect.

[0060] In a sixth aspect, an embodiment of the present application provides a vehicle, comprising the device described in any one of the second aspects above.

[0061] It can be understood that the device described in the second aspect, the device described in the third aspect, the computer storage medium described in the fourth aspect, or the computer program product described in the fifth aspect, and the vehicle described in the sixth aspect are all used to execute any method provided in the first aspect. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The following is an introduction to the drawings used in the embodiments of the present application.

[0063] Figure 1 It is a schematic diagram of the architecture of a system for determining association relationships between road elements provided in an embodiment of the present application;

[0064] Figure 2a It is a flowchart of a method for determining the association relationship between road elements provided in an embodiment of the present application;

[0065] Figure 2b This is a schematic diagram of obtaining data provided by an embodiment of the present application;

[0066] Figure 2c is another schematic diagram of acquiring data provided in an embodiment of the present application;

[0067] Figure 3is a schematic diagram of another method for determining the association relationship between road elements provided in an embodiment of the present application;

[0068] Figure 4 It is a structural schematic diagram of a device for determining association relationships between road elements provided in an embodiment of the present application;

[0069] Figure 5 It is a structural schematic diagram of another device for determining the association relationship between road elements provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] It should be noted that the prefixes such as "first" and "second" used in this application are only used to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. For example, if the described object is a "field", the ordinal number before the "field" in the "first field" and the "second field" does not limit the position or order between the "fields". "First" and "second" do not limit whether the "fields" they modify are in the same message, nor do they limit the order of the "first field" and the "second field". For another example, if the described object is a "level", the ordinal number before the "level" in the "first level" and the "second level" does not limit the priority between the "levels". For another example, the number of described objects is not limited by the prefix and can be one or more. Taking "first device" as an example, the number of "devices" can be one or more. In addition, the objects modified by different prefixes can be the same or different. For example, if the object being described is a "device", then the "first device" and the "second device" can be the same device, the same type of device, or different types of devices; for another example, if the object being described is "information", then the "first information" and the "second information" can be information of the same content or information of different contents. In short, the use of prefixes used to distinguish the described objects in the embodiments of the present application does not constitute a limitation on the described objects. For the statement of the described objects, refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes.

[0071] It should be noted that the description methods such as "at least one of a1, a2, ... and an" used in the embodiments of the present application include the situation where any one of a1, a2, ... and an exists alone, and also include any combination of any multiple of a1, a2, ... and an, and each situation can exist alone. For example, the description method of "at least one of a, b and c" includes the situation where a is alone, b is alone, c is alone, a combination of a and b, a combination of a and c, b and c, or a combination of abc.

[0072] To facilitate understanding, the following first introduces relevant terms that may be involved in the embodiments of the present application.

[0073] Autonomous driving, also known as intelligent driving or assisted driving, is an important direction for the development of vehicle intelligence. With the development of perception technology and the improvement of chip capabilities, intelligent driving provides people with more and more rich driving functions and gradually realizes different levels of driving experience. The Society of Automotive Engineers (SAE) provides a driving automation classification standard, including driving levels L0 to L5, where L0 is no automation and the human driver has full control over the vehicle. During driving, the driver can get warnings or assistance from the driving system, such as autonomous emergency braking (AEB), blind spot monitoring (BSM) or lane departure warning (LDW). Level 1 is driving assistance, and the driving operation is completed by the human driver and the driving system. The driving system can provide driving assistance for the steering wheel or acceleration and deceleration operations through the driving environment, and other driving operations are performed by the human driver, such as adaptive cruise control (ACC) or lane keep assistance / support (LKA / LKS). Level 2 is partial automation, which provides driving assistance for the steering wheel and multiple acceleration and deceleration operations through the driving environment, and other driving actions are performed by the human driver, such as a combination of adaptive cruise control (ACC) and lane keep assistance (LKA / LKS). Level 3 is conditional automation, where the driving system can complete all driving operations, but the human driver needs to respond to the request of the driving system at the appropriate time, that is, the human driver needs to be prepared to take over the driving system; Level 4 is highly automated, where the driving system can complete all driving operations, and the human driver does not necessarily need to respond to the request of the driving system. For example, when the road and environmental conditions permit (such as closed parks, highways, urban roads or fixed driving routes, etc.), the human driver may not take over the driving; Level 5 is fully automated, where the driving system can autonomously complete driving operations under various road and environmental conditions that human drivers can cope with. It can be seen that at levels L0 to L2, the driving system mainly provides support for the driver, and the driver still needs to supervise the driving and steer, brake or accelerate as needed to ensure safety. At levels L3 to L5, the driving system can complete all driving operations on behalf of the driver. At level L3, the driver must be prepared to take over the driving. At levels L4 and L5, the driving system can achieve full driving under some conditions and all conditions, and the driver can choose whether to take over.

[0074] The above classification is an example. With the evolution of technology or different regulations in different countries or regions, the above classification may change. For example, the vehicle automation classification proposed by the Ministry of Industry and Information Technology of China includes 6 levels of vehicle driving automation, among which Levels 0-2 are driving assistance, the system assists humans in performing dynamic driving tasks, and the driving subject is still the driver; Levels 3-5 are automatic driving, the system replaces humans in performing dynamic driving tasks under the designed operating conditions, and when the function is activated, the driving subject is the system. The names and definitions of each level are as follows: Level 0 driving automation (emergency assistance) The system cannot continuously perform the lateral or longitudinal motion control of the vehicle in dynamic driving tasks, but has the ability to continuously perform partial target and event detection and response in dynamic driving tasks. Level 1 driving automation (partial driver assistance) The system continuously performs the lateral or longitudinal motion control of the vehicle in dynamic driving tasks under its designed operating conditions (or called the design operating range ODD), and has the ability to detect and respond to partial targets and events that are compatible with the lateral or longitudinal motion control of the vehicle being performed. Level 2 driving automation (combined driver assistance) systems continuously perform lateral and longitudinal motion control of the vehicle in dynamic driving tasks under their designed operating conditions, and have the ability to detect and respond to some targets and events that are compatible with the lateral and longitudinal motion control of the vehicle being performed. Level 3 driving automation (conditionally automated driving) systems continuously perform all dynamic driving tasks under their designed operating conditions. Level 4 driving automation (highly automated driving) systems continuously perform all dynamic driving tasks under their designed operating conditions and automatically implement minimum risk strategies. Level 5 driving automation (fully automated driving) systems continuously perform all dynamic driving tasks under any drivable conditions and automatically implement minimum risk strategies. Among them, lateral control is mainly used for vehicle steering control, for example, controlling the steering wheel torque or angle to control the direction of the vehicle; longitudinal control is mainly used for vehicle speed control, for example, controlling the brake pedal, accelerator pedal, or gear position to control the acceleration / deceleration, braking, etc. of the vehicle.

[0075] Regardless of the classification method adopted, the description of the embodiments of the present application can be applied to the above-mentioned automatic driving systems that require partial or full participation in vehicle driving.

[0076] Among them, road structure recognition can greatly expand the application boundaries of high-level autonomous driving by reconstructing the road conditions around the vehicle through real-time analysis of the collected data from the on-board sensors. Currently, many solutions only focus on the reconstruction of geometric information on the road. In the past two years, some work has begun to focus on the analysis of lane topology, and very few works have noticed the need to parse the relationship between various elements on the road. In addition, due to the limitations of the detection accuracy of geometric information, it is difficult to reason about the relationship between road elements. Based on this, an embodiment of the present application provides a method for determining the relationship between road elements, which can be processed based on road perception element information and road prior element information to obtain the relationship between road elements and support road structure reconstruction. By introducing road prior information, the effect of relationship reasoning can be effectively enhanced.

[0077] The following will describe the system architecture of the embodiment of the present application in detail with reference to the accompanying drawings. Figure 1 , Figure 1 1 is a schematic diagram of the architecture of a system for determining the association relationship between road elements applicable to an embodiment of the present application, the system includes a vehicle 101 and a server 102. Among them:

[0078] The vehicle 101 is a device with communication and computing capabilities, capable of providing mobile travel services to users. The vehicle 101 can provide an environment for deploying software, hardware, or a combination of software and hardware modules.

[0079] The vehicle 101 in the embodiment of the present application is a device that moves by power drive. The vehicle 101 generally includes a variety of subsystems, such as but not limited to a travel system, a sensor system, a control system, one or more peripheral devices, a power supply or a user interface, etc. Optionally, the vehicle 101 may also include more or fewer subsystems, and each subsystem may include multiple elements. In addition, each subsystem and element of the vehicle 101 may be interconnected by wire or wirelessly.

[0080] It should be noted that the vehicle 101 in the embodiment of the present application can be a car or an electric car, a rail-running vehicle, an intelligent vehicle (such as an unmanned vehicle) or an intelligent mobile robot, etc.

[0081] Among them, intelligent vehicles support the functions of sensing the road environment through the on-board sensor system, automatically planning the driving route and controlling the vehicle to reach the predetermined target location. Intelligent vehicles use computers, sensors, information fusion, communications, artificial intelligence machines or automatic control technologies, and are a high-tech complex that integrates environmental perception, planning and decision-making, and multi-level assisted driving. For example, intelligent vehicles can be cars or wheeled mobile robots with assisted driving systems or fully automatic driving systems.

[0082] The server 102 is a device with centralized computing capabilities. Exemplarily, the server 102 can be implemented by a server, a virtual machine, a cloud, a roadside device, or a robot.

[0083] When the service end 102 includes a server, the type of the server includes but is not limited to a general-purpose computer, a dedicated server computer, a blade server, etc. The present application does not impose a strict limit on the number of servers included in the service end 102, and the number may be one or more (such as a server cluster, etc.). As a possible implementation, the service end 102 may be in communication with the vehicle 101.

[0084] A virtual machine refers to a computing module that has complete hardware system functions and runs in a completely isolated environment through software simulation. Of course, in addition to virtual machines, the server 102 can also be implemented through other computing instances, such as containers.

[0085] The cloud is a software platform that uses application virtualization technology, which enables one or more software and applications to be developed and run in an independent virtualized environment. Optionally, when the server 102 is implemented through the cloud, the cloud can be deployed on a public cloud, a private cloud, or a hybrid cloud.

[0086] A roadside device is a device installed on the side of a road (or at an intersection, or on the side of a road, etc.). The road can be an outdoor road (such as a main road, a secondary road, an elevated road, or a temporary road, etc.), or an indoor road (such as a road in an indoor parking lot, etc.). A roadside device can provide services for vehicles. It should be noted that a roadside device can be an independent device or integrated into other devices. For example, a roadside device can be integrated into devices such as smart gas stations, charging piles, smart traffic lights, street lights, telephone poles, or traffic signs.

[0087] It should be noted that the system for determining the association relationship between road elements of the present application may also include only vehicle 101, or only server 102, etc., and this solution does not limit this.

[0088] The embodiments of the present application can be applied to the field of autonomous driving, for example, for autonomous driving vehicle-side systems. It can be installed in an autonomous driving system that does not rely on a high-precision map (HD map), and can be used between an autonomous driving perception module and a regulation and control module. It can also be used for autonomous driving map construction. In the construction of autonomous driving maps, by applying this solution, it can help self-packaging and crowdsourcing mapping to associate new data with old data. At the same time, since it avoids association based only on old data, it can better cope with the impact of outdated map versions and real changes, and help to quickly iterate map versions. Of course, it can also be applied to other systems, and this solution does not limit this.

[0089] The architecture of the embodiment of the present application is described above, and the method of the embodiment of the present application is described in detail below.

[0090] Reference Figure 2a FIG. 1 is a flow chart of a method for determining the association relationship between road elements provided in an embodiment of the present application. Optionally, the method can be applied to the aforementioned system for determining the association relationship between road elements, for example Figure 1 A system for determining the relationship between road elements shown in FIG. Figure 2a The method for determining the association relationship between road elements shown may include steps 201-203. It should be understood that for the convenience of description, this application is described in the order of 201-203, and is not intended to limit the execution to the above order. The embodiment of the present application does not limit the execution order, execution time, execution number, etc. of the above one or more steps. The following description takes the execution subject of steps 201-203 of the method for determining the association relationship between road elements as an example of a vehicle, and this application is also applicable to other execution subjects such as servers. Steps 201-203 are as follows:

[0091] 201. Obtain road perception element information and road prior element information.

[0092] The road perception element information can be understood as real-time information of elements such as roads. The road perception element information can include one or more of the element information such as lanes, roads, traffic lights, road signs, zebra crossings, road boundaries, intersections, signs, etc., and of course, it can also be other element information, which is not limited in this solution.

[0093] The road prior element information can be understood as the known information of elements such as roads.

[0094] In a possible implementation, the vehicle may obtain road perception element information from roadside devices, vehicle cameras and / or radar sensors. The vehicle may obtain road prior element information from user input or maps (e.g., standard definition maps, navigation maps (SD maps)). Of course, other methods are possible, and this solution does not limit this.

[0095] like Figure 2b As shown, the vehicle 101 receives data from the user device 103, which data may be obtained according to information input by the user, for example. Then, the vehicle 101 obtains road perception element information and / or road prior element information according to the data.

[0096] User equipment refers to equipment that can interact with a user. The user equipment of the embodiment of the present application includes a handheld terminal, a wearable device, an entertainment device, etc.

[0097] For example, refer to Figure 2c As shown, the vehicle 101 receives data from a device associated with the vehicle 101. Then, the vehicle 101 obtains road perception element information and road prior element information based on the data.

[0098] The device associated with the vehicle refers to a device that can establish a communication connection with the vehicle 101. For example, the roadside device 104, the intersection radar, the base station communicating with the vehicle, the server providing the vehicle with an intelligent driving strategy, etc.

[0099] For example, the road perception element information and road a priori element information obtained from the roadside device 104 may be obtained through other devices, and this solution does not specifically limit this.

[0100] Optionally, the device associated with the vehicle may also include an on-board sensing device. The on-board sensing device may be, for example, one or more of an on-board camera, a millimeter wave radar, a laser radar, or an ultrasonic radar. The vehicle may obtain data through the on-board sensing device. Then, the vehicle obtains road sensing element information based on the data. For example, the vehicle may sense road sensing element information through an on-board camera.

[0101] 202. Process the road perception element information and the road priori element information to obtain a first eigenvector and a second eigenvector.

[0102] It is understandable that the number of the first eigenvector and the second eigenvector can be one or more. Preferably, the number of the first eigenvector and the second eigenvector is multiple.

[0103] In a possible implementation, the first feature vector is a road perception element feature vector, and the second feature vector is a road priori element feature vector.

[0104] For example, feature extraction is performed on the road perception element information to obtain a road perception element feature vector. Feature extraction is performed on the road priori element information to obtain a road priori element feature vector.

[0105] Optionally, the vehicle obtains road perception element information from the perception module and obtains road prior element information from the navigation map. Then, the vehicle inputs the road perception element information and the road prior element information into the feature extraction model for processing, and obtains a road perception element feature vector and a road prior element feature vector. Exemplarily, the feature extraction model may be a convolutional neural network, a fully connected neural network, etc. It is understandable that the road perception element feature vector may include feature vectors corresponding to a plurality of different road perception element information. Accordingly, the road prior element feature vector may include feature vectors corresponding to a plurality of different road prior element information.

[0106] Of course, this processing can also be performed on the server side, and this solution does not limit this.

[0107] In another possible implementation, the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

[0108] The fused road perception element feature vector can be understood as a road perception element feature vector obtained by implicitly considering the influence of road prior element information. Of course, it can also include the fusion of road perception element information, such as fusion of lanes and road arrows to obtain lane information with turning attributes.

[0109] Accordingly, the fused road prior element feature vector can be understood as a road prior element feature vector obtained by implicitly considering the influence of the road perception element information. Of course, it can also include the fusion of road prior element information, etc. This solution does not limit this.

[0110] For example, the road perception element features and the road a priori element features are obtained by performing feature interaction processing on the road perception element information and the road a priori element information.

[0111] The feature interaction processing can be understood as referring to the road priori element information while extracting the features of the road perception element information. Correspondingly, referring to the road perception element information while extracting the features of the road priori element information.

[0112] Optionally, both the road perception element information and the road prior element information are input into a neural network for multi-head attention mechanism processing, and a fused road perception element feature vector and a fused road prior element feature vector can be obtained.

[0113] In this way, a more stable and comprehensive lane feature can be obtained.

[0114] Among them, this scheme can also be implemented in a rule-constrained manner, by extracting the key features of road elements through artificially defined rules, constructing matching strategies to calculate the association relationship, and completing the final relationship verification and output through the rule transmission method. For example, by obtaining one or more of the location information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information. Among them, for the perception element, the attribute information can refer to its perception confidence; for the map data, the attribute information can refer to the time of its construction, indicating its information update status, etc., and this scheme does not limit this. Then, according to one or more of the location information, category information, direction information, and attribute information of the different elements, the elements with a distance less than the preset distance are interactively fused to obtain a fused road perception element feature vector and a fused road prior element feature vector. For example, the road surface sign feature and the lane feature are interactively fused to obtain a lane description with sign information. For another example, the same-direction lanes are merged to form a road-level feature description.

[0115] The distance can be understood as the degree of relevance indicated by the information, rather than just the traditional geographical distance. This example uses distance as the selection criterion, and of course it can also be based on other information for interactive merging, such as interactive fusion based on the orientation information of the elements. For example, for the association between lanes and traffic lights, in addition to judging by the distance, the orientation information of the two can also be referred to for judgment. Of course, other methods can also be used, and this solution does not limit this.

[0116] The above is only an example, and other processing methods may also be used, which are not limited in this solution.

[0117] 203. Obtain, according to the first feature vector and the second feature vector, an association relationship between road perception elements and / or an association relationship between road perception elements and road a priori elements.

[0118] The association between the road perception elements may be, for example, an association between lanes, or an association between lanes and traffic lights, etc. The association between the road perception elements and the road prior elements may be, for example, an association between lanes and roads SDRoads from road prior SD Maps, etc.

[0119] The association relationship may include having a relationship, not having a relationship, etc.

[0120] The first eigenvector is the road perception element eigenvector, and the second eigenvector is the road prior element eigenvector:

[0121] For example, by performing a dot product on any two road perception element feature vectors, the association matrix (or other forms that can describe the association relationship, such as triples, etc.) between the two road perception elements can be calculated, that is, the association relationship between the two road perception elements can be obtained. If the corresponding relationship position of the association matrix meets the threshold setting, the two road perception elements have an association relationship. If the corresponding relationship position of the association matrix does not meet the threshold setting, the two road perception elements do not have an association relationship.

[0122] like Figure 3 As shown in the figure, a neural network is used to extract features from the perceived lane and traffic light information, as well as the road SDRoad information from the road prior SD Map, to obtain the road perception element feature vector and the road prior element feature vector. Then, by performing dot multiplication on the feature vectors, the association matrix between lanes, the association matrix between lanes and traffic lights, and the association matrix between road SDRoad and traffic lights are calculated, that is, the corresponding association relationships are obtained.

[0123] For another example, the feature vectors of different elements can be concatenated, and then the relationship categories of different elements can be decoded (for example, a multi-layer perceptron (MLP) is used on the concatenated vectors, etc.), to obtain the association relationship between road perception elements, the association relationship between road perception elements and road prior elements, etc. For example, if the decoding result meets the threshold, the two road perception elements have an association relationship. If the decoding result does not meet the threshold, the two road perception elements do not have an association relationship.

[0124] The first eigenvector is the fused road perception element eigenvector, and the second eigenvector is the fused road prior element eigenvector:

[0125] In a possible implementation, a matching cost function is constructed based on the fused road perception element feature vector and the fused road prior element feature vector. The association relationship between the road perception elements and the association relationship between the road perception elements and the road prior elements are obtained based on the matching cost function. For example, the matching cost function may be a function constrained by distance, orientation, category, etc., which is not limited in this solution. If the constructed association constraint is satisfied, there is a relationship between the elements. If the constructed association constraint is not satisfied, there is no relationship between the elements. For example, a matching cost function is constructed by combining various road perception element information, road prior element information, fused road perception element feature vector, and fused road prior element feature vector, and the matching cost function is used to determine the association relationship between the perception elements and between the perception elements and the prior information. For example, in order to have an association between the perception lane and the SD Map road, the geometric positions of the two need to be close and the driving directions need to be similar. For another example, in order to have an association between the perception lane and the perception traffic light, the traveling direction of the perception lane needs to be toward the location of the traffic light, and the traffic light needs to be toward the lane. Furthermore, the above-mentioned multiple relationships can be checked or integrated according to the downstream task requirements and finally output.

[0126] In another possible implementation, the fused feature vectors of different elements may be concatenated, and then the relationship categories of different elements may be decoded to obtain the correlation between road perception elements, the correlation between road perception elements and road prior elements, etc. For the introduction of this part, please refer to the above record, which will not be repeated here.

[0127] Based on the above method, the correlation relationship between road perception elements, the correlation relationship between road perception elements and road prior elements, and the correlation relationship between road prior elements can be obtained.

[0128] Based on the association relationship between the elements, the redundant relationship can be verified or screened.

[0129] For example, a verification function can be constructed with the help of multi-frame time series information to verify the association relationship. For example, for multi-frame data, the relationship of the same elements is tracked. If more frames have an association relationship within the time window, it can be considered that there is an association relationship.

[0130] In a possible implementation, according to the association relationship between the road perception elements, redundant association relationships between the road perception elements are merged or deleted to obtain processed association relationships between the road perception elements.

[0131] According to the association relationship between the road perception element and the road a priori element, the redundant association relationship between the road perception element and the road a priori element is merged or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

[0132] The above-mentioned merging may, for example, be transferring and merging the relationship between related elements. For example, the relationship between perceived traffic lights and perceived roads, perceived traffic lights and prior roads, and perceived roads and prior roads may be comprehensively considered to obtain the final relationship between traffic lights and perceived roads. The above-mentioned deletion may be deleting repeated association relationships, such as analogy merging, deleting auxiliary relationships, etc. For example, if the focus is on the relationship between traffic lights and perceived roads, the relationship between perceived roads and prior roads, and the relationship between perceived traffic lights and prior roads may be deleted.

[0133] Of course, the above multiple relationships can also be verified or integrated in combination with downstream task requirements and finally output. For example, through the association between perceived lanes and SD Map roads, and the association between perceived traffic lights and SD Maps, we can infer the partial association between perceived lanes and perceived traffic lights, which can be verified or supplemented with the association between perceived lanes and perceived traffic lights obtained by direct inference.

[0134] This can simplify the association relationship between road perception elements and the association relationship between road perception elements and road prior elements, making the obtained association relationship results clearer, more concise and accurate.

[0135] Among them, the association relationship between the elements can be displayed through the interface of the display device in the vehicle or the human-computer interaction interface. Of course, other methods can also be used, and this solution does not limit this.

[0136] In an embodiment of the present application, the first eigenvector and the second eigenvector are obtained by processing the road perception element information and the road prior element information, and then the correlation relationship between the road perception elements and / or the correlation relationship between the road perception elements and the road prior elements are obtained based on the first eigenvector and the second eigenvector. This method, by associating the current road element with the road prior information, can use the road prior information as a bridge to improve the level of reasoning about the current road element relationship. At the same time, by reasoning about the relationship between the current road element and the road prior, a redundant relationship can be constructed, which is convenient for checking the current road element relationship and improving the robustness and interpretability of the overall solution.

[0137] This solution does not rely heavily on road prior information for relationship reasoning. According to needs, the final output can be only the association of current road elements, avoiding negative impacts caused by outdated road prior information.

[0138] It should be noted that in the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between the various embodiments are consistent and can be referenced mutually, and the technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0139] The above detailed description of the method of the embodiment of the present application, the following provides a device of the embodiment of the present application. It can be understood that in the various device embodiments of the present application, the division of multiple units or modules is only a logical division according to function, and is not used as a limitation on the specific structure of the device. In a specific implementation, some functional modules may be subdivided into more small functional modules, and some functional modules may also be combined into one functional module, but no matter whether these functional modules are subdivided or combined, the general process performed by the device is the same. For example, some devices contain a receiving unit and a sending unit. In some designs, the sending unit and the receiving unit can also be integrated into a communication unit, which can implement the functions implemented by the receiving unit and the sending unit. Usually, each unit corresponds to its own program code (or program instruction), and when the program code corresponding to each of these units is run on the processor, the unit is controlled by the processing unit to execute the corresponding process to implement the corresponding function.

[0140] The embodiments of the present application also provide a device for implementing any of the above methods. For example, a device for determining the association relationship between road elements includes a module (or means) for implementing each step performed by a vehicle in any of the above methods.

[0141] For example, refer to Figure 4 FIG. 1 is a schematic diagram of a structure of a device for determining associations between road elements provided in an embodiment of the present application. The device for determining associations between road elements is used to implement the aforementioned method for determining associations between road elements, for example Figure 2a , Figure 3 A method for determining the association relationship between road elements shown.

[0142] like Figure 4 As shown, the device may include an acquisition module 401, a processing module 402 and a determination module 403, which are specifically as follows:

[0143] An acquisition module 401 is used to acquire road perception element information and road prior element information;

[0144] A processing module 402 is used to process the road perception element information and the road prior element information to obtain a first feature vector and a second feature vector;

[0145] The determination module 403 is used to obtain the association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements according to the first feature vector and the second feature vector.

[0146] In a possible implementation, the processing module 402 is configured to:

[0147] Feature extraction is performed on the road perception element information and the road prior element information respectively to obtain the first feature vector and the second feature vector, wherein the first feature vector is a road perception element feature vector, and the second feature vector is a road prior element feature vector.

[0148] In another possible implementation, the processing module 402 is configured to:

[0149] The road perception element information and the road prior element information are subjected to feature interaction processing to obtain the first feature vector and the second feature vector, wherein the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

[0150] In a possible implementation, the processing module 402 is further configured to:

[0151] The road perception element information and the road prior element information are processed by a multi-head attention mechanism to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0152] In another possible implementation, the processing module 402 is further configured to:

[0153] According to at least one of the position information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information, elements whose distance is less than a preset distance are interactively fused to obtain the fused road perception element feature vector and the fused road prior element feature vector.

[0154] In a possible implementation, the processing module 402 is further configured to:

[0155] Constructing a matching cost function according to the fused road perception element feature vector and the fused road prior element feature vector;

[0156] The association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements is obtained according to the matching cost function.

[0157] In a possible implementation, the determining module 403 is further configured to:

[0158] Merging and / or deleting redundant association relationships between the road perception elements to obtain processed association relationships between the road perception elements;

[0159] The redundant association relationship between the road perception element and the road a priori element is merged and / or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

[0160] For the introduction of the above modules, please refer to the records of the above methods, which will not be repeated here.

[0161] It should be understood that the division of each module in each of the above devices is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. In addition, the modules in the device for determining the association relationship between road elements can be implemented in the form of a processor calling software; for example, the device for determining the association relationship between road elements includes a processor, the processor is connected to a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or implement the functions of each module of the device, wherein the processor is, for example, a general-purpose processor, such as a central processing unit (CPU) or a microprocessor, and the memory is a memory inside the device or a memory outside the device. Alternatively, the modules in the device can be implemented in the form of hardware circuits, and the functions of some or all units can be realized by designing the hardware circuits, and the hardware circuits can be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of some or all of the above units are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be realized by a programmable logic device (PLD), taking a field programmable gate array (FPGA) as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of some or all of the above units. All modules of the above devices can be realized in the form of a processor calling software, or in the form of hardware circuits, or in part by a processor calling software, and the rest by hardware circuits.

[0162] Reference Figure 5 , which is a schematic diagram of the hardware structure of another device for determining the association relationship between road elements provided in an embodiment of the present application. Figure 5 The device 500 for determining the association relationship between road elements shown (the device 500 may be a computer device) includes a memory 501, a processor 502, a communication interface 503 and a bus 504. The memory 501, the processor 502 and the communication interface 503 are connected to each other through the bus 504.

[0163] The memory 501 may be a read only memory (ROM), a static storage device, a dynamic storage device or a random access memory (RAM).

[0164] The memory 501 can store programs. When the program stored in the memory 501 is executed by the processor 502, the processor 502 and the communication interface 503 are used to execute the various steps of the method for determining the association relationship between road elements in the embodiment of the present application.

[0165] The processor 502 is a circuit with signal processing capability. In one implementation, the processor 502 may be a circuit with instruction reading and running capability, such as a central processing unit CPU, a microprocessor, a graphics processing unit (GPU) (which may be understood as a microprocessor), or a digital signal processor (DSP); in another implementation, the processor 502 may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit may be fixed or reconfigurable, such as the processor 502 being a hardware circuit implemented by an ASIC or a programmable logic device PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the hardware circuit configuration may be understood as the process of the processor loading instructions to implement the functions of some or all of the above modules. In addition, it may also be a hardware circuit designed for artificial intelligence, which may be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc. The processor 502 is used to execute relevant programs to implement the functions required to be executed by the units in the device for determining the association relationship between road elements in the embodiment of the present application, or to execute the method for determining the association relationship between road elements in the method embodiment of the present application.

[0166] It can be seen that each module in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0167] In addition, the modules in the above device can be fully or partially integrated together, or can be implemented independently. In one implementation, these modules are integrated together and implemented in the form of a system-on-a-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or implementing the functions of the modules of the device. The type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0168] The communication interface 503 uses a transceiver device such as, but not limited to, a transceiver to implement communication between the apparatus 500 and other devices or a communication network. For example, data can be acquired through the communication interface 503 .

[0169] The bus 504 may include a path for transmitting information between various components of the device 500 (eg, the memory 501 , the processor 502 , and the communication interface 503 ).

[0170] It should be noted that although Figure 5 The device 500 shown only shows a memory, a processor, and a communication interface, but in the specific implementation process, those skilled in the art should understand that the device 500 also includes other devices necessary for normal operation. At the same time, according to specific needs, those skilled in the art should understand that the device 500 may also include hardware devices for implementing other additional functions. In addition, those skilled in the art should understand that the device 500 may also only include the devices necessary to implement the embodiments of the present application, and does not necessarily include Figure 5 All devices shown in .

[0171] An embodiment of the present application also provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is executed on a computer or a processor, the computer or the processor executes one or more steps in any of the above methods.

[0172] The embodiment of the present application further provides a computer program product including instructions. When the computer program product is executed on a computer or a processor, the computer or the processor executes one or more steps in any of the above methods.

[0173] It should be understood that in the description of the present application, unless otherwise specified, " / " indicates that the objects associated before and after are in an "or" relationship, for example, A / B can represent A or B; wherein A and B can be singular or plural. Also, in the description of the present application, unless otherwise specified, "multiple" refers to two or more than two. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, wherein a, b, c can be single or multiple. In addition, in order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first", "second", etc. are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not limit them to be necessarily different. Meanwhile, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete manner for ease of understanding.

[0174] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the division of the unit is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The mutual coupling, direct coupling, or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0175] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a 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 through the computer-readable storage medium. The computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media integrations. The available medium may be a read-only memory (ROM), or a random access memory (RAM), or a magnetic medium, such as a floppy disk, a hard disk, a tape, a magnetic disk, or an optical medium, such as a digital versatile disc (DVD), or a semiconductor medium, such as a solid state disk (SSD), etc.

[0177] The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto, and any changes or replacements within the technical scope disclosed in the embodiment of the present application should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.

Claims

1. A method for determining the association relationship between road elements, characterized in that: include: Acquire road perception element information and road prior element information; Processing the road perception element information and the road prior element information to obtain a first eigenvector and a second eigenvector; The association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements is obtained according to the first feature vector and the second feature vector.

2. The method according to claim 1, characterized in that The processing of the road perception element information and the road prior element information to obtain a first feature vector and a second feature vector includes: Feature extraction is performed on the road perception element information and the road prior element information respectively to obtain the first feature vector and the second feature vector, wherein the first feature vector is a road perception element feature vector, and the second feature vector is a road prior element feature vector.

3. The method according to claim 1, characterized in that The processing of the road perception element information and the road prior element information to obtain a first feature vector and a second feature vector includes: The road perception element information and the road prior element information are subjected to feature interaction processing to obtain the first feature vector and the second feature vector, wherein the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

4. The method according to claim 3, characterized in that: The performing feature interaction processing on the road perception element information and the road prior element information to obtain the first feature vector and the second feature vector includes: The road perception element information and the road prior element information are processed by a multi-head attention mechanism to obtain the fused road perception element feature vector and the fused road prior element feature vector.

5. The method according to claim 3, characterized in that: Performing feature interaction processing on the road perception element information and the road prior element information to obtain the first feature vector and the second feature vector includes: According to at least one of the position information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information, elements whose distance is less than a preset distance are interactively fused to obtain the fused road perception element feature vector and the fused road prior element feature vector.

6. The method according to claim 5, characterized in that The obtaining, according to the first feature vector and the second feature vector, the association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements includes: Constructing a matching cost function according to the fused road perception element feature vector and the fused road prior element feature vector; The association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements is obtained according to the matching cost function.

7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Merging and / or deleting redundant association relationships between the road perception elements to obtain processed association relationships between the road perception elements; The redundant association relationship between the road perception element and the road a priori element is merged and / or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

8. A device for determining the association relationship between road elements, characterized in that: include: An acquisition module, used to acquire road perception element information and road prior element information; A processing module, used for processing the road perception element information and the road prior element information to obtain a first eigenvector and a second eigenvector; A determination module is used to obtain the association relationship between road perception elements and / or the association relationship between road perception elements and road a priori elements according to the first feature vector and the second feature vector.

9. The device according to claim 8, characterized in that The processing module is used to: Feature extraction is performed on the road perception element information and the road prior element information respectively to obtain the first feature vector and the second feature vector, wherein the first feature vector is a road perception element feature vector, and the second feature vector is a road prior element feature vector.

10. The device according to claim 8, characterized in that The processing module is used to: The road perception element information and the road prior element information are subjected to feature interaction processing to obtain the first feature vector and the second feature vector, wherein the first feature vector is a fused road perception element feature vector, and the second feature vector is a fused road prior element feature vector.

11. The device according to claim 10, characterized in that The processing module is further used for: The road perception element information and the road prior element information are processed by a multi-head attention mechanism to obtain the fused road perception element feature vector and the fused road prior element feature vector.

12. The device according to claim 10, characterized in that The processing module is further used for: According to at least one of the position information, category information, direction information, and attribute information of different elements in the road perception element information and the road prior element information, elements whose distance is less than a preset distance are interactively fused to obtain the fused road perception element feature vector and the fused road prior element feature vector.

13. The device according to claim 12, characterized in that The processing module is further used for: Constructing a matching cost function according to the fused road perception element feature vector and the fused road prior element feature vector; The association relationship between the road perception elements and / or the association relationship between the road perception elements and the road a priori elements is obtained according to the matching cost function.

14. The device according to any one of claims 8 to 13, characterized in that The determining module is further used for: Merging and / or deleting redundant association relationships between the road perception elements to obtain processed association relationships between the road perception elements; The redundant association relationship between the road perception element and the road a priori element is merged and / or deleted to obtain a processed association relationship between the road perception element and the road a priori element.

15. A device for determining the association relationship between road elements, characterized in that: It comprises a processor and a memory; wherein the memory is used to store program code, and the processor is used to call the program code to execute the method according to any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 7.

17. A computer program product, characterized in that When the computer program product runs on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

18. A vehicle, characterized in that: The vehicle comprises the device as described in any one of claims 8-14.

Images