Vehicle association matching method and device, electronic equipment and storage medium

Abstract

The application provides a vehicle association matching method and device, electronic equipment and a storage medium, and belongs to the technical field of automatic driving. The method comprises the following steps: obtaining monitoring data obtained by a plurality of sensors for monitoring the motion state of a target vehicle; matching preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; assigning the target vehicle matching list based on a preset matching threshold to obtain a value corresponding to the target vehicle matching list; and obtaining an optimal matching target vehicle based on the value corresponding to the target vehicle matching list. The application solves the technical problems of low association matching precision and limited application scenarios in the prior art.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and more specifically to a vehicle association and matching method, device, electronic device, and storage medium. Background Technology

[0002] Prior art CN 117150433 A provides a target association matching method, which inputs data into a trained Siamese neural network, calculates the matching degree between each fused target and multiple sensing targets through the Siamese neural network, and matches the fused target with the sensing targets based on multiple matching degrees to obtain a first matching relationship. Prior art CN 114170270 A discloses a single-sensor multi-target association matching method, which preprocesses the acquired target information, generates a first detection and tracking SIFT (Scale-Invariant Feature Transform) feature map based on the position and velocity of the detected and tracked targets, performs target association pairing based on the SIFT feature map, evaluates the optimal matching pair by combining tracking pairing information, and outputs the optimal matching pair as the sensor's detected target.

[0003] In comparison, CN 117150433 A requires Siamese neural network training. Siamese neural networks have complex structures and high computational requirements, limiting their application in certain scenarios and making them unsuitable for engineering applications. CN114170270 A requires generating SIFT feature maps, which has high computational complexity and limits its use in real-time applications. Although the SIFT algorithm performs well under rotation and scale changes, it is not invariant to other types of transformations such as image distortion and projection transformations, which may lead to feature point extraction failures in these cases. Furthermore, in some situations, the SIFT algorithm may encounter the problem of feature point scarcity, which also affects its performance in specific application scenarios. Therefore, existing technologies suffer from low correlation matching accuracy and limited application scenarios. Summary of the Invention

[0004] In view of this, it is necessary to provide a vehicle association matching method, device, electronic device and storage medium to solve the technical problems of low association matching accuracy and limited application scenarios in the prior art.

[0005] To address the aforementioned technical problems, this invention provides a vehicle association and matching method, comprising: Acquire monitoring data obtained from multiple sensors monitoring the motion state of the target vehicle; The preset sensor type parameters and target vehicle motion parameters are matched with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data; The target vehicle matching list is assigned a value based on the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list. Based on the cost value corresponding to the target vehicle matching list, the optimal matching target vehicle is obtained.

[0006] In one possible implementation, the step of matching preset sensor type parameters and target vehicle motion parameters with monitoring data to obtain a target vehicle matching list includes: The monitoring data is matched with preset sensor type parameters to obtain a target vehicle tracking list; the target vehicle tracking list includes: multiple target vehicles and their corresponding tracking data; The target vehicle tracking list is matched sequentially with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list.

[0007] In one possible implementation, the step of sequentially matching the target vehicle tracking list with preset target vehicle motion parameters, and adding the target vehicle tracking list to the target vehicle matching list when all matches are found, includes: Update the target vehicle tracking list to obtain the updated list; The updated list is matched sequentially with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list.

[0008] In one possible implementation, the step of updating the target vehicle tracking list to obtain an updated list includes: Obtain updated monitoring data; Within the created container, an update operation is performed based on the updated monitoring data and the target vehicle tracking list to obtain the updated list.

[0009] In one possible implementation, the step of performing an update operation within the created container based on the updated monitoring data and the target vehicle tracking list to obtain an updated list includes: Within the created container, the updated monitoring data and target vehicle tracking list are updated using the kdtree algorithm to obtain the updated list.

[0010] In one possible implementation, the updated monitoring data and target vehicle tracking list are updated based on the kdtree algorithm within the created container to obtain an updated list, including: Within the created container, a preset search radius and the number of times the target tracking list is set are used. Based on the preset search radius, the updated monitoring data and the target vehicle tracking list are updated by incrementing the tracking count by 1 using the kdtree algorithm, resulting in the updated list.

[0011] In one possible implementation, obtaining the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list includes: Sort the value of each vehicle in the target vehicle matching list by size, and select the vehicle with the highest value as the optimal matching target vehicle.

[0012] On the other hand, the present invention also provides a vehicle association matching device, comprising: The monitoring data acquisition module is used to acquire monitoring data obtained from multiple sensors monitoring the motion state of the target vehicle. The matching module is used to match preset sensor type parameters and target vehicle motion parameters with monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data; The cost value module is used to assign values ​​based on the target vehicle matching list and the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list. The optimal matching target vehicle module is used to obtain the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list.

[0013] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the vehicle association matching method in any of the above implementations.

[0014] On the other hand, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instructions, which, when executed by a processor, can implement the steps in the vehicle association matching method described in any of the above implementations.

[0015] The beneficial effects of this invention are as follows: The method provided by this invention includes: acquiring monitoring data obtained by multiple sensors monitoring the motion state of a target vehicle. This is done to coordinate multiple sensors to meet detection requirements in different environments, thus making it applicable to multiple scenarios; matching preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes multiple matching target vehicles and their corresponding matching data. This is done to fuse multiple sensors, thereby performing sensor correlation matching and effectively integrating data collected by different sensors; assigning values ​​based on the target vehicle matching list and a preset matching threshold to obtain the cost value corresponding to the target vehicle matching list; and obtaining the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list. This is done to sort the matching correlation based on matching cost, thereby obtaining the best matching result. This process has the advantages of low computational load, high real-time performance, and high correlation matching accuracy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic flowchart of an embodiment of the vehicle association matching method provided by the present invention; Figure 2 A schematic diagram of the matching process in one embodiment of the vehicle association matching method provided by the present invention; Figure 3 A flowchart illustrating the process of updating the matching list in one embodiment of the vehicle association matching method provided by the present invention; Figure 4 A flowchart illustrating the association condition matching process in one embodiment of the vehicle association matching method provided by the present invention; Figure 5 A schematic diagram of an embodiment of the vehicle association matching device provided by the present invention; Figure 6 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] In the description of the embodiments of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In this embodiment of the invention, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product or device.

[0021] The naming or numbering of steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical order indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effect can be achieved.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] This invention provides a vehicle association matching method, device, electronic device, and storage medium, which are described below.

[0024] Figure 1 A schematic flowchart of an embodiment of the vehicle association matching method provided by the present invention and Figure 2 A schematic diagram of the matching process in one embodiment of the vehicle association matching method provided by the present invention includes: S101. Acquire monitoring data obtained from multiple sensors monitoring the motion state of the target vehicle; S102. Match the preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data; S103. Assign values ​​based on the target vehicle matching list and the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list. S104. Based on the cost value corresponding to the target vehicle matching list, obtain the optimal matching target vehicle.

[0025] It is understood that this application proposes a vehicle association matching method, which obtains the best matching result by setting relevant matching parameters and ranking the matching association degree based on the matching cost. The method proposed in this application has the advantages of low computational load, high real-time performance and high association matching accuracy.

[0026] It can be further understood that the method provided by the present invention includes: acquiring monitoring data obtained by multiple sensors monitoring the motion state of a target vehicle. The purpose of this is to coordinate multiple sensors to meet the detection requirements in different environments, thus making it applicable to multiple scenarios; matching preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data. The purpose of this is to fuse multiple sensors, thereby associating and matching the sensors and effectively integrating the data collected by different sensors; assigning values ​​to the target vehicle matching list and preset matching thresholds to obtain the cost value corresponding to the target vehicle matching list; and obtaining the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list. The purpose of this is to sort the matching correlation based on the matching cost, thereby obtaining the best matching result. This process has the advantages of low computational load, high real-time performance, and high correlation matching accuracy.

[0027] It should be further explained that in step S101, the monitoring data obtained is obtained by acquiring data from multiple sensors (camera / millimeter-wave radar) (including information such as ID, horizontal and vertical position, horizontal and vertical velocity, etc.), and the data from different sensors can be distinguished according to the sensor type sensor_type.

[0028] In step S102, the sensor type parameter is set as a preliminary matching parameter according to different sensor types. The matching parameters include whether it matches the sensor, the number of times it has been tracked, the most recent number of times it has not matched, the maximum speed difference of obstacles, and the set search radius.

[0029] In step S102, the monitoring data is matched sequentially with preset sensor type parameters. When all matches are found, the data is added to the target vehicle tracking list. The specific matching process is as follows: For each target within the search range, perform the following checks in sequence: a. Determine if the target index is valid; if invalid, proceed to the next target. b. Determine if the target is newly added; if so, proceed to the next target. c. Determine if the target contains valid data; if not, proceed to the next target. d. Determine if the target matches the current sensor; if not, proceed to the next target. e. Determine if the target has been tracked too few times; if so, move on to the next target. f. Determine if the target has not been tracked recently; if so, proceed to the next target. g. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0030] If a target meets the above conditions, it will be added to the target vehicle tracking list.

[0031] In steps S102 and S103, the target vehicle tracking list is matched with preset target vehicle motion parameters to obtain a target vehicle matching list. The target vehicle matching list is then assigned a value based on a preset matching threshold to obtain the corresponding cost value. The specific process is as follows: Matching costs are calculated based on the following conditions. Matching costs are calculated based on the following factors: Positional deviation: If the positional difference between two objects is small, the cost is low.

[0032] Vertical distance: If an object is closer to the source object in terms of vertical distance, the cost is lower.

[0033] Speed ​​deviation: If the speed difference between two objects is small, the cost is low.

[0034] Number of tracking attempts: The more times an object is tracked, the lower the cost.

[0035] If an object has just matched another object, the cost is lower.

[0036] The system checks if the conditions for adding a partnership are met based on the parameters. If not, it continues processing the next object. The filtered results are as follows: Figure 2 As shown, the camera target IDs within the search radius (dashed circle in the figure) are 1, 2, and 6. The millimeter-wave radar IDs are 2 and 3.

[0037] In step S104, the optimal matching target vehicle is obtained based on the cost value corresponding to the target vehicle matching list. The specific process of sorting the cost values ​​corresponding to the target matching list and selecting the one with the largest cost value as the optimal matching target is as follows: 1) Sort the associated target information according to the matching cost.

[0038] 2) Obtain a reference to the object that matches the current object, and obtain a pointer to the first association. Then, check if the index of the first association is equal to the index of the current object.

[0039] If it is not equal, it means that the current object can be better matched with another object, and it is necessary to continue searching for other matches. If the index of the first association is equal to the index of the current object, it means that the current object matches itself. Therefore, the best match has been found.

[0040] 3) Finally, the optimal matching target.

[0041] In some embodiments of the present invention, preset sensor type parameters and target vehicle motion parameters are matched with monitoring data to obtain a target vehicle matching list, including: The monitoring data is matched with preset sensor type parameters to obtain a target vehicle tracking list; the target vehicle tracking list includes: multiple target vehicles and their corresponding tracking data; The target vehicle tracking list is matched sequentially with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list.

[0042] Understandably, the specific steps are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished by sensor_type. Each sensor's data is added to the target vehicle tracking list. Newly added targets (camera IDs: 1, 2, 3, 4, 5; millimeter-wave radar IDs: 1, 2, 3, 4, 5, 6) are set to new tracking targets. The target tracking count, matching count, and loop count are initialized to 0.

[0043] The second step is to set the sensor type parameters. These parameters vary depending on the sensor type and include whether the target is matched, the number of times it has been tracked, the most recent number of unmatched targets, the maximum target speed difference, and the set search radius.

[0044] In this embodiment, the camera type parameters are set as follows: number of times the camera has been matched with the millimeter-wave radar and matched with the camera: 8 times, the most recent number of times the camera has not been matched: 6 times, the maximum speed difference of the target: 8 km / h, and the search radius: 5 m.

[0045] The third step is to update the target vehicle tracking list. A container is created to store the indexes and information of matching targets. A kd-tree (K-Dimensional Tree, a spatially distinct data structure) is used to find matching targets within the search radius and add them to the target vehicle tracking list.

[0046] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list and save their indexes to the index container. 3) For each target within the search range, perform the following checks in sequence: h. Determine if the target index is valid; if invalid, proceed to the next target. i. Determine if the target is newly added; if so, proceed to the next target. j. Determine if the target contains valid data; if not, proceed to the next target. k. Determine if the target matches the current sensor; if not, proceed to the next target. l. Determine if the target has been tracked too few times; if so, proceed to the next target. m. Determine if the target has not been tracked recently; if so, proceed to the next target. n. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0047] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0048] Figure 3 A flowchart illustrating the process of updating the matching list in one embodiment of the vehicle association matching method provided by the present invention includes: Update the target vehicle tracking list to obtain the updated list; The updated list is matched sequentially with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list.

[0049] Understandably, the main steps of the update operation are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished according to the sensor type (sensor_type). Each sensor's data is added to the target vehicle tracking list. The status of newly added targets is set to "new tracking target". The target tracking count, matching count, and loop count are initialized to 0.

[0050] The second step is to set the initial sensor type parameters. These parameters vary depending on the sensor type and include whether the target is matched, the number of times it has been tracked, the most recent number of unmatched targets, the maximum target velocity difference, and the set search radius.

[0051] The third step is to update the target vehicle tracking list. A container is created to store the indexes of matching obstacles and target information. A kdtree is used to find matching targets within the search radius and add them to the matching list.

[0052] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list, increment the tracking count by 1, and save its index to the index container. 3) For each target within the search range, perform the following checks in sequence: a. Determine if the target index is valid; if invalid, proceed to the next target. b. Determine if the target is newly added; if so, proceed to the next target. c. Determine if the target contains valid data; if not, proceed to the next target. d. Determine if the target matches the current sensor; if not, proceed to the next target. e. Determine if the target has been tracked too few times; if so, move on to the next target. f. Determine if the target has not been tracked recently; if so, proceed to the next target. g. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0053] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0054] In some embodiments of the present invention, the step of updating the target vehicle tracking list to obtain an updated list includes: Obtain updated monitoring data; Within the created container, an update operation is performed based on the updated monitoring data and the target vehicle tracking list to obtain the updated list.

[0055] Understandably, the main steps of the update operation are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished according to the sensor type (sensor_type). Each sensor's data is added to the target vehicle tracking list. The status of newly added targets is set to "new tracking target". The target tracking count, matching count, and loop count are initialized to 0.

[0056] The second step is to set the sensor type parameters. These parameters vary depending on the sensor type and include whether the target is matched, the number of times it has been tracked, the most recent number of unmatched targets, the maximum target speed difference, and the set search radius.

[0057] The third step is to update the target vehicle tracking list. A container is created to store the indexes of matching obstacles and target information. A kdtree is used to find matching targets within the search radius and add them to the matching list.

[0058] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list, increment the tracking count by 1, and save its index to the index container. 3) For each target within the search range, perform the following checks in sequence: a. Determine if the target index is valid; if invalid, proceed to the next target. b. Determine if the target is newly added; if so, proceed to the next target. c. Determine if the target contains valid data; if not, proceed to the next target. d. Determine if the target matches the current sensor; if not, proceed to the next target. e. Determine if the target has been tracked too few times; if so, move on to the next target. f. Determine if the target has not been tracked recently; if so, proceed to the next target. g. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0059] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0060] In some embodiments of the present invention, the step of performing an update operation in the created container based on the updated monitoring data and the target vehicle tracking list to obtain an updated list includes: Within the created container, the updated monitoring data and target vehicle tracking list are updated using the kdtree algorithm to obtain the updated list.

[0061] Understandably, the main steps of the update operation are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished according to the sensor type (sensor_type). Each sensor's data is added to the target vehicle tracking list. The status of newly added targets is set to "new tracking target". The target tracking count, matching count, and loop count are initialized to 0.

[0062] The second step is to set the sensor type parameters. These parameters vary depending on the sensor type and include whether the target is matched, the number of times it has been tracked, the most recent number of unmatched targets, the maximum target speed difference, and the set search radius.

[0063] The third step is to update the target vehicle tracking list. A container is created to store the indexes of matching obstacles and target information. A kdtree is used to find matching targets within the search radius and add them to the matching list.

[0064] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list, increment the tracking count by 1, and save its index to the index container. 3) For each target within the search range, perform the following checks in sequence: a. Determine if the target index is valid; if invalid, proceed to the next target. b. Determine if the target is newly added; if so, proceed to the next target. c. Determine if the target contains valid data; if not, proceed to the next target. d. Determine if the target matches the current sensor; if not, proceed to the next target. e. Determine if the target has been tracked too few times; if so, move on to the next target. f. Determine if the target has not been tracked recently; if so, proceed to the next target. g. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0065] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0066] In some embodiments of the present invention, the step of updating the updated monitoring data and the target vehicle tracking list based on the kdtree algorithm in the created container to obtain the updated list includes: Within the created container, a preset search radius and tracking count for the tracking list are established. Based on the preset search radius, the tracking count for the updated monitoring data and the target vehicle tracking list is incremented by 1 using the kdtree algorithm to obtain the updated list.

[0067] Understandably, the main steps of the update operation are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished according to the sensor type (sensor_type). Each sensor's data is added to the target vehicle tracking list. The status of newly added targets is set to "new tracking target". The target tracking count, matching count, and loop count are initialized to 0.

[0068] The second step is to set the sensor type parameters. Depending on the sensor type, matching parameters include whether the target is matched with that sensor, the number of times it has been tracked, the most recent number of unmatched matches, the maximum target velocity difference, and the set search radius.

[0069] The third step is to update the target vehicle tracking list. A container is created to store the indexes of matching obstacles and target information. A kdtree is used to find matching targets within the search radius and add them to the matching list.

[0070] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list, increment the tracking count by 1, and save its index to the index container. 3) For each target within the search range, perform the following checks in sequence: a. Determine if the target index is valid; if invalid, proceed to the next target. b. Determine if the target is newly added; if so, proceed to the next target. c. Determine if the target contains valid data; if not, proceed to the next target. d. Determine if the target matches the current sensor; if not, proceed to the next target. e. Determine if the target has been tracked too few times; if so, move on to the next target. f. Determine if the target has not been tracked recently; if so, proceed to the next target. g. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0071] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0072] Figure 4 A flowchart of an embodiment of the vehicle association matching method provided by the present invention, comprising association condition matching, including: Sort the target vehicle matching list by their corresponding cost values ​​and select the vehicle with the highest cost value as the optimal matching target vehicle. The specific steps are as follows: The first step is to read sensor data. Data from multiple sensors (camera / millimeter-wave radar) is acquired (including ID, lateral and longitudinal position, lateral and longitudinal velocity, etc.). Data from different sensors can be distinguished by sensor_type. Each sensor's data is added to the target vehicle tracking list. Newly added targets (camera IDs: 1, 2, 3, 4, 5; millimeter-wave radar IDs: 1, 2, 3, 4, 5, 6) are set to new tracking targets. The target tracking count, matching count, and loop count are initialized to 0.

[0073] The second step is to set the sensor type parameters. These parameters vary depending on the sensor type and include whether the target is matched, the number of times it has been tracked, the most recent number of unmatched targets, the maximum target speed difference, and the set search radius.

[0074] Preferably, in this embodiment, the camera type parameters are set as follows: matched with millimeter-wave radar, matched with camera, number of times already tracked: 8 times, most recent number of times not matched: 6 times, maximum target speed difference: 8 km / h, and search radius: 5 m.

[0075] The third step is to update the target vehicle tracking list. A container is created to store the indexes and information of matching targets. A kd-tree is used to find matching targets within the search radius and add them to the matching list.

[0076] 1) Calculate the coordinates of the search center point based on the input source target location information. 2) Use the kdtree algorithm to find targets within the search radius of the source target in the target vehicle tracking list and save their indexes to the index container. 3) For each target within the search range, perform the following checks in sequence: h. Determine if the target index is valid; if invalid, proceed to the next target. i. Determine if the target is newly added; if so, proceed to the next target. j. Determine if the target contains valid data; if not, proceed to the next target. k. Determine if the target matches the current sensor; if not, proceed to the next target. l. Determine if the target has been tracked too few times; if so, proceed to the next target. m. Determine if the target has not been tracked recently; if so, proceed to the next target. n. Determine if the relative speed deviation of the target is too large; if so, proceed to the next target.

[0077] 4) If the target meets the above conditions, add it to the target vehicle tracking list.

[0078] The fourth step involves adding partnerships. This involves searching the target vehicle tracking list for vehicles with matching attributes and determining whether to add them as partners based on certain conditions. The resulting value corresponds to the target vehicle matching list. The specific process is as follows: 1) First, determine if the index of the source object is valid. If it is invalid, output an error message and return. 2) Iterate through the list of matched objects. For each matched object, perform the following operations: f. Determine if the index of the matched object is valid. If invalid, output an error message and continue processing the next object. g. Obtain a reference to the matched object and its corresponding sensor reference. h. Calculate the positional difference between the source object and the matched object, and find the index of the point with the smallest positional difference. i. Calculate the matching cost based on the following conditions. The matching cost is calculated based on the following factors: Positional deviation: If the positional difference between two objects is small, the cost is low.

[0079] Vertical distance: If an object is closer to the source object in terms of vertical distance, the cost is lower.

[0080] Speed ​​deviation: If the speed difference between two objects is small, the cost is low.

[0081] Number of tracking attempts: The more times an object is tracked, the lower the cost.

[0082] Recently tracked: If an object has just been tracked, the cost is lower.

[0083] If an object has just matched another object, the cost is lower.

[0084] The system determines whether the conditions for adding a partnership are met based on the parameters. If not, it continues processing the next object. j. Get the list index of the source object and check if it is valid. k. Add associations to the matched objects and add the associated object information to the source object. 3) The function ends after all objects have been processed.

[0085] After the above steps, the targets added to the partnership are: Camera target IDs: 1, 2; Millimeter-wave radar IDs: 2, 3.

[0086] Step 5: Set the best matching object. The specific process is as follows: 1) Sort the associated target information according to the matching cost from lowest to highest. The sorted results are shown in Table 1 below: Table 1. Results sorted by matching cost

[0087] 2) Obtain a reference to the object that matches the current object, and obtain a pointer to the first association. Then, check if the index of the first association is equal to the index of the current object.

[0088] If it is not equal, it means that the current object can be better matched with another object, and it is necessary to continue searching for other matches. If the index of the first association is equal to the index of the current object, it means that the current object matches itself. Therefore, the best match has been found.

[0089] Finally, the target camera ID2, which is the best match for the millimeter-wave radar ID1, was obtained.

[0090] The method provided by this invention includes: acquiring monitoring data obtained by multiple sensors monitoring the motion state of a target vehicle, the purpose of which is to coordinate multiple sensors to meet the detection requirements in different environments, thus making it applicable to multiple scenarios; matching preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data, the purpose of which is to fuse multiple sensors, thereby associating and matching the sensors and effectively integrating the data collected by different sensors; assigning values ​​to the target vehicle matching list and preset matching thresholds to obtain the cost value corresponding to the target vehicle matching list; and obtaining the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list, the purpose of which is to sort the matching correlation based on the matching cost, thereby obtaining the best matching result. This process has the advantages of low computational load, high real-time performance, and high correlation matching accuracy.

[0091] Figure 5 A schematic diagram of an embodiment of the vehicle association matching device provided by the present invention includes: The monitoring data acquisition module 501 is used to acquire monitoring data obtained by multiple sensors monitoring the motion state of the target vehicle; The matching module 502 is used to match the preset sensor type parameters and target vehicle motion parameters with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data; The cost value module 503 is used to assign values ​​based on the target vehicle matching list and the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list. The optimal matching target vehicle module 504 is used to obtain the optimal matching target vehicle based on the cost value corresponding to the target vehicle matching list. The vehicle association matching device 500 provided in the above embodiments can implement the technical solutions described in the above vehicle association matching method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above vehicle association matching method embodiments, and will not be repeated here.

[0092] like Figure 6 As shown, the present invention also provides an electronic device 600. The electronic device 600 includes a processor 601, a memory 602, and a display 603. Figure 6 Only some components of the electronic device 600 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0093] In some embodiments, processor 601 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 602 or process data, such as the vehicle association matching method of the present invention.

[0094] In some embodiments, processor 601 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 601 may be local or remote. In some embodiments, processor 601 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, etc., or any combination thereof.

[0095] In some embodiments, memory 602 may be an internal storage unit of electronic device 600, such as a hard disk or memory of electronic device 600. In other embodiments, memory 602 may also be an external storage device of electronic device 600, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 600.

[0096] Furthermore, the memory 602 may include both internal storage units of the electronic device 600 and external storage devices. The memory 602 is used to store application software and various types of data installed on the electronic device 600.

[0097] In some embodiments, display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 603 is used to display information from electronic device 600 and to display a visual user interface. Components 601-603 of electronic device 600 communicate with each other via a system bus.

[0098] In one embodiment, when processor 601 executes the vehicle association matching program in memory 602, the following steps may be implemented: Acquire monitoring data obtained from multiple sensors monitoring the motion state of the target vehicle; The preset sensor type parameters and target vehicle motion parameters are matched with the monitoring data to obtain a target vehicle matching list; the target vehicle matching list includes: multiple matching target vehicles and their corresponding matching data; The target vehicle matching list is assigned a value based on the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list. Based on the cost value corresponding to the target vehicle matching list, the optimal matching target vehicle is obtained. It should be understood that when the processor 601 executes the vehicle association matching program in the memory 602, in addition to the above functions, it can also implement other functions, as detailed in the description of the corresponding method embodiments above.

[0099] Furthermore, this embodiment of the invention does not specifically limit the type of electronic device 600 mentioned. Electronic device 600 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the invention, electronic device 600 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0100] Accordingly, this application also provides a computer-readable storage medium for storing a computer-readable program or instruction. When the program or instruction is executed by a processor, it can implement the steps or functions of the vehicle association matching method provided in the above-described method embodiments.

[0101] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0102] The vehicle association matching method, device, electronic device, and storage medium provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for vehicle association and matching, characterized in that, include: The system acquires monitoring data obtained from multiple sensors monitoring the motion state of a target vehicle; the monitoring data includes vehicle ID, lateral and longitudinal position, and lateral and longitudinal speed; the sensors include cameras and millimeter-wave radar. The monitoring data is matched with preset sensor type parameters to obtain a target vehicle tracking list; the sensor type parameters include: whether it matches the sensor, the number of times it has been tracked, the most recent number of times it has not matched, the maximum speed difference of obstacles, and the set search radius; the target vehicle tracking list includes: multiple vehicles of the tracked targets and their corresponding tracking data; Obtain the updated monitoring data; in the created container, preset the search radius, and based on the preset search radius and the tracking count of the tracking list, use the kdtree algorithm to update the updated monitoring data and the target vehicle tracking list by incrementing the tracking count by 1 to obtain the updated list; The updated list is matched sequentially with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list. The target vehicle matching list includes multiple matching target vehicles and their corresponding matching data. The target vehicle matching list is assigned a value based on a preset matching threshold to obtain the cost value corresponding to the target vehicle matching list; the cost value is determined comprehensively based on position deviation, longitudinal distance, speed deviation, number of tracking attempts, and whether a match has just been made. Sort the value of each vehicle in the target vehicle matching list by size, and select the vehicle with the highest value as the optimal matching target vehicle.

2. The vehicle association matching method according to claim 1, characterized in that, Before sequentially matching the updated list with the preset target vehicle motion parameters, the following judgments are made on the targets within the search range: The system determines whether the target index is valid, whether the target is newly added, whether the target contains valid data, whether the target matches the current sensor, whether the target has been tracked too few times, whether the target has not been tracked recently, and whether the target's relative vehicle speed deviation is too large. Targets that meet all the determination conditions are included in the matching range.

3. A vehicle association and matching device, characterized in that, include: The monitoring data acquisition module is used to acquire monitoring data obtained from multiple sensors monitoring the motion state of the target vehicle; the monitoring data includes vehicle ID, lateral and longitudinal position, and lateral and longitudinal speed; the sensors include cameras and millimeter-wave radar; The tracking list generation module is used to match monitoring data with preset sensor type parameters to obtain a target vehicle tracking list; the sensor type parameters include: whether it matches the sensor, the number of times it has been tracked, the most recent number of times it has not been matched, the maximum speed difference of obstacles, and the set search radius; The target vehicle tracking list includes: multiple target vehicles and their corresponding tracking data; The list update module is used to obtain the updated monitoring data. In the created container, a preset search radius is set, and based on the preset search radius and the number of tracking times in the tracking list, the updated monitoring data and the target vehicle tracking list are updated by incrementing the tracking count by 1 using the kdtree algorithm to obtain the updated list. The matching list generation module sequentially matches the updated list with the preset target vehicle motion parameters. When all matches are found, the target vehicle tracking list is added to the target vehicle matching list. The target vehicle matching list includes multiple matching target vehicles and their corresponding matching data. The cost value calculation module is used to assign values ​​based on the target vehicle matching list and the preset matching threshold to obtain the cost value corresponding to the target vehicle matching list; the cost value is determined comprehensively based on position deviation, longitudinal distance, speed deviation, number of tracking times and whether a match has just been made. The optimal matching target vehicle module is used to sort the cost values ​​corresponding to the target vehicle matching list by size and select the one with the largest cost value as the optimal matching target vehicle.

4. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is configured to execute the program stored in the memory to implement the steps in the vehicle association matching method as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the vehicle association matching method according to any one of claims 1 to 2.

Images