Target Tracking Method, Apparatus, Base Station, and Storage Medium

Abstract

The present application relates to a target tracking method, apparatus, base station, and storage medium. The method includes: obtaining a first tracking result and a second tracking result of respectively tracking a target using synchronized first point cloud data and second point cloud data; the first point cloud data being data collected by a lidar, and the second point cloud data being data collected by a millimeter wave radar; performing target association and matching processing on the first tracking result and the second tracking result to determine a target matching result; and performing fusion processing on the first tracking result and the second tracking result according to the target matching result to determine a tracking result of the target. Using this method can improve the accuracy of the obtained target tracking result.

Description

Technical Field

[0001] This application relates to the field of computer technologies, and particularly to a target tracking method, apparatus, base station, and storage medium. Background Art

[0002] Target tracking refers to establishing the positional relationship of a target to be tracked in a continuous data sequence to obtain the complete motion trajectory of the target. Usually, given the position features of the target in the previous frame, the position and the size of the bounding box of the target in the next frame are predicted to achieve the tracking of the target.

[0003] Currently, with the rapid development of lidar technology, in order to achieve accurate tracking of a target, most often lidar is used to collect point cloud data of the target at various moments, and target detection and target tracking are performed on the collected point cloud data at various moments, so as to achieve precise tracking of the target.

[0004] However, when using the above technology to track a target, there is a problem that the obtained tracking result is not accurate enough. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a target tracking method, apparatus, base station, and storage medium that can improve the accuracy of the obtained tracking result.

[0006] In a first aspect, this application provides a target tracking method, which includes:

[0007] Obtaining a first tracking result and a second tracking result of respectively tracking a target using synchronized first point cloud data and second point cloud data; the first point cloud data is data collected by a lidar, and the second point cloud data is data collected by a millimeter-wave radar;

[0008] Performing target association and matching processing on the first tracking result and the second tracking result to determine a target matching result;

[0009] Performing fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0010] In one of the embodiments, the above performing target association and matching processing on the first tracking result and the second tracking result to determine a target matching result includes:

[0011] Performing target association and matching processing on the first tracking result and the second tracking result using the Hungarian algorithm to determine a target matching result;

[0012] Among them, the target matching result includes the successfully matched target, the first target that fails to be matched, and the second target that fails to be matched. The matched target is the same target that exists in both the first tracking result and the second tracking result. The first target is the target in the first tracking result, and the second target is the target in the second tracking result.

[0013] In one embodiment, the first tracking result includes the laser detection box information and laser identifier of the matched target, and the second tracking result includes the millimeter-wave detection box information and millimeter-wave identifier of the matched target.

[0014] The above-mentioned fusion processing of the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target includes:

[0015] If the target matching result includes the matched target, then perform fusion processing on the laser detection box information and laser identifier of the matched target, and the millimeter-wave detection box information and millimeter-wave identifier of the matched target to determine the tracking result of the matched target.

[0016] In one embodiment, the above-mentioned fusion processing of the laser detection box information and laser identifier of the matched target, and the millimeter-wave detection box information and millimeter-wave identifier of the matched target to determine the tracking result of the matched target includes:

[0017] Detect whether there are a laser identifier and a millimeter-wave identifier in the historical trajectory of the target to obtain a detection result. The historical trajectory of the target is the motion trajectory obtained by tracking the target with a lidar and a millimeter-wave radar at a historical moment.

[0018] According to the detection result, perform fusion processing on the laser detection box information and laser identifier of the matched target, and the millimeter-wave detection box information and millimeter-wave identifier of the matched target to determine the tracking result of the matched target.

[0019] In one embodiment, the first tracking result includes the first identifier and first detection box information of the first target, and the second tracking result includes the second identifier and second detection box information of the second target.

[0020] The above-mentioned fusion processing of the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target includes:

[0021] If the target matching result includes the first target, then determine the tracking result of the first target according to the first identifier, first detection box information, and the second tracking result.

[0022] If the target matching result includes the second target, then determine the tracking result of the second target according to the second identifier, second detection box information, and the first tracking result.

[0023] In one embodiment, determining the tracking result of the first target according to the first identifier, the first detection frame information, and the second tracking result includes:

[0024] Detecting whether the first identifier exists in the preset target historical trajectory;

[0025] If the first identifier exists, discard the second tracking result, and use the first identifier and the first detection frame information as the tracking result of the first target.

[0026] In one embodiment, determining the tracking result of the second target according to the second identifier, the second detection frame information, and the first tracking result includes:

[0027] Detecting whether the second identifier exists in the preset target historical trajectory;

[0028] If the second identifier exists, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result;

[0029] Determine the tracking result of the second target according to the second detection frame information and the latest distance deviation.

[0030] In one embodiment, determining the tracking result of the second target according to the second detection frame information and the latest distance deviation includes:

[0031] Subtract the latest distance deviation from the second detection frame information, and combine it with the second identifier to obtain the tracking result of the second target.

[0032] In a second aspect, the present application also provides a target tracking device, which includes:

[0033] An acquisition module, configured to acquire the first tracking result and the second tracking result of tracking the target by using the synchronized first point cloud data and second point cloud data respectively; the first point cloud data is the data collected by a lidar, and the second point cloud data is the data collected by a millimeter wave radar;

[0034] A matching module, configured to perform target association matching processing on the first tracking result and the second tracking result to determine a target matching result;

[0035] A determination module, configured to perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0036] In a third aspect, the present application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0037] Obtain the first tracking result and the second tracking result of respectively tracking the target with the first point cloud data and the second point cloud data after synchronization; the first point cloud data is the data collected by the lidar, and the second point cloud data is the data collected by the millimeter wave radar;

[0038] Perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result;

[0039] Perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0040] In a fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0041] Obtain the first tracking result and the second tracking result of respectively tracking the target with the first point cloud data and the second point cloud data after synchronization; the first point cloud data is the data collected by the lidar, and the second point cloud data is the data collected by the millimeter wave radar;

[0042] Perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result;

[0043] Perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0044] In a fifth aspect, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0045] Obtain the first tracking result and the second tracking result of respectively tracking the target with the first point cloud data and the second point cloud data after synchronization; the first point cloud data is the data collected by the lidar, and the second point cloud data is the data collected by the millimeter wave radar;

[0046] Perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result;

[0047] Perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0048] The above-mentioned target tracking method, device, base station, and storage medium obtain the first tracking result and the second tracking result after tracking the target through the synchronized lidar point cloud data and millimeter-wave radar point cloud data, perform target association matching processing on the first tracking result and the second tracking result, and fuse the first tracking result and the second tracking result through the obtained target matching result to determine the tracking result of the target. In this method, since the tracking results of the two can be fused through the target association matching result in the point cloud data of both the lidar and the millimeter-wave radar, the high-precision advantage of the lidar point cloud data and the long-distance advantage of the millimeter-wave radar point cloud data can be retained, realizing the complementary advantages of the two point cloud data, that is, the obtained tracking result combines the advantages of the two point cloud data, thereby making the tracking result obtained after data fusion more accurate. Description of the Drawings

[0049] Figure 1 It is a schematic structural diagram of a roadside base station in an embodiment;

[0050] Figure 2 It is a schematic flowchart of a target tracking method in an embodiment;

[0051] Figure 3 It is a schematic flowchart of a target tracking method in another embodiment;

[0052] Figure 4 It is a schematic flowchart of a target tracking method in another embodiment;

[0053] Figure 5 It is a schematic flowchart of a target tracking method in another embodiment;

[0054] Figure 6 It is a schematic detailed flowchart of a target tracking method in another embodiment;

[0055] Figure 7 It is a structural block diagram of a target tracking device in an embodiment;

[0056] Figure 8 It is an internal structural diagram of a computer device in an embodiment. Detailed Embodiments

[0057] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] The target tracking method provided by the embodiments of the present application can be applied to, for example Figure 1The roadside base station shown. Among them, the roadside base station includes a millimeter-wave radar 102, a lidar 104, and a computer device 106 connected to both the millimeter-wave radar and the lidar. The millimeter-wave radar 102 can collect target data within the field of view of the millimeter-wave radar at each moment, and transmit the obtained millimeter-wave point cloud data to the computer device 106 for processing; at the same time, the lidar 104 can collect target data within the field of view of the lidar at each moment, and transmit the obtained lidar point cloud data to the computer device 106 for processing. For the number of the millimeter-wave radars 102 and the number of the lidars 104, both can be one or more, and can be set according to the actual road conditions. The computer device 106 can be an independent terminal or server, or an integrated device integrated inside the millimeter-wave radar 102 or integrated inside the lidar 104, and no specific limitation is made here.

[0059] In one embodiment, as Figure 2 shown, a target tracking method is provided. This embodiment relates to the specific process of how to determine the tracking result of the target by combining the tracking results of the lidar and the millimeter-wave radar. Taking the method applied to Figure 1 the computer device 106 in it as an example for illustration, the method may include the following steps:

[0060] S202, obtain the first tracking result and the second tracking result of respectively tracking the target with the synchronized first point cloud data and the second point cloud data.

[0061] Among them, the first point cloud data is the data collected by the lidar, and the second point cloud data is the data collected by the millimeter-wave radar. The synchronization here includes time synchronization and space synchronization. For time synchronization, it mainly includes the following steps: Step 1, obtain the lidar data and the millimeter-wave radar data simultaneously, and mark the current system timestamps of the two frames of data as t1 and t2 respectively while obtaining the two frames of data; Step 2, calculate the time difference between t1 and t2; Step 3, if the time difference calculated in Step 2 is less than 0.1 second, it is considered that the two frames of data are synchronized, otherwise execute Step 4; Step 4, discard the current lidar data, take the timestamp of the next frame of lidar data, and calculate the time difference; Step 5, when the time difference in Step 4 is less than 0.1, it is considered that the two frames of data are time-synchronized, and then Step 4 can be cyclically executed.

[0062] For spatial synchronization, the rotation and translation matrix can be calculated by obtaining lidar data and millimeter-wave radar data at the same moment and using the corresponding points selected on two time-synchronized data frames. Specifically, it includes the following steps: Step 1, obtain multiple frames of lidar data and multiple frames of millimeter-wave radar data on time synchronization; Step 2, through the obtained multiple frames of lidar data, find a lidar data frame suitable for registration from the multiple frames of lidar data. For example, if the number of vehicles included in a certain frame of lidar data is moderate, preferably 4 to 6, then this frame of lidar data is used as the lidar data frame suitable for registration; Step 3, convert the millimeter-wave radar data to the world coordinate system, draw it into the lidar data, observe the points in the millimeter-wave radar data corresponding to the vehicles in the lidar data, and take 4-5 corresponding points in the front direction of the vehicle head; Step 4, use the corresponding data points to solve the rotation and translation matrix; Step 5, use the rotation and translation matrix to draw the millimeter-wave data into the lidar data, and observe whether the two frames of data are spatially synchronized. If not, execute Steps 3-5. If synchronized, end the process.

[0063] Specifically, the data collected by the lidar is recorded as the first point cloud data, and the data collected by the millimeter-wave radar is recorded as the second point cloud data. Through the above steps, the first point cloud data and the second point cloud data can be synchronized in time and space. After that, target detection and target tracking processing can be performed on the first point cloud data to obtain the first tracking result; at the same time, target detection and target tracking processing can be performed on the second point cloud data to obtain the second tracking result.

[0064] Among them, the first tracking result can include the detection box information and the identification of each detected target. Of course, it can also include the running trajectories of each target at historical moments, the speeds at each moment, etc.; similarly, the second tracking result can include the detection box information and the identification of each detected target. Of course, it can also include the running trajectories of each target at historical moments, the speeds at each moment, etc.

[0065] S204, perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result.

[0066] It should be noted that in this embodiment, both the lidar and the millimeter-wave radar perform global tracking, and each has its own advantages when performing global tracking. Generally, when the lidar detects and tracks targets, the detection accuracy is relatively high, while when the millimeter-wave radar detects and tracks targets, the detection and tracking distance is relatively far, that is, it can detect and track targets at a long distance.

[0067] The target association and matching here refers to the process of determining whether the targets in the synchronized lidar data and millimeter-wave radar data come from the same target.

[0068] In this step, after obtaining the first tracking result corresponding to the lidar and the second tracking result corresponding to the millimeter-wave radar, a relevant association algorithm can be used to perform association matching on the targets in the first tracking result and the second tracking result to obtain the matching result of the targets. The association algorithms here can include, for example, the nearest neighbor algorithm, the Hungarian algorithm, the probabilistic data association algorithm (PDA), the joint probabilistic data association algorithm (JPDA), etc.

[0069] S206. Perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0070] In this step, after performing association matching on the target in the above step, the tracking result in the lidar and the tracking result in the millimeter-wave radar of the same target can be fused through the association matching result of the target, that is, the superior data of the two data is taken, so that the obtained tracking result of the target is relatively accurate.

[0071] In the above target tracking method, the first tracking result and the second tracking result obtained by tracking the target through the synchronized lidar point cloud data and millimeter-wave radar point cloud data are obtained, and the target association matching process is performed on the first tracking result and the second tracking result, and the first tracking result and the second tracking result are fused through the obtained target matching result to determine the tracking result of the target. In this method, since the tracking results of the two can be fused through the target association matching result in the point cloud data of the lidar and the millimeter-wave radar, the high-precision advantage of the lidar point cloud data and the long-distance advantage of the millimeter-wave radar point cloud data can be retained, realizing the complementary advantages of the two point cloud data, that is, the obtained tracking result combines the advantages of the two point cloud data, so that the accuracy of the tracking result obtained after data fusion is higher.

[0072] In the above embodiments, it is mentioned that the association matching process can be performed on two types of tracking results. The following embodiments will elaborate on the specific association matching process.

[0073] In another embodiment, another target tracking method is provided. On the basis of the above embodiments, the above S204 may include the following step A:

[0074] Step A. Use the Hungarian algorithm to perform target association matching processing on the first tracking result and the second tracking result to determine the target matching result.

[0075] Among them, the target matching result includes the successfully matched target, the first target that fails to be matched, and the second target that fails to be matched. The matched target is the same target that exists in both the first tracking result and the second tracking result. The first target is the target in the first tracking result, and the second target is the target in the second tracking result.

[0076] In this step, the Hungarian algorithm is mainly used to associate and match the targets in the tracking results of lidar and millimeter-wave radar, and determine whether the targets in the two types of tracking results are the same target.

[0077] The Hungarian algorithm is mainly used to solve the combinatorial optimization problem of task assignment. For example, a target with the highest matching degree can be found for the target in the first tracking result of lidar in the second tracking result of millimeter-wave radar. Of course, there may also be a situation where the target in the first tracking result of lidar has no corresponding target in the second tracking result of millimeter-wave radar, or the target in the second tracking result of millimeter-wave radar has no corresponding target in the first tracking result of lidar.

[0078] That is to say, in this step, after associating and matching the targets in the tracking results of lidar and millimeter-wave radar, the matched target (denoted as the matched target), the target that only exists in the first lidar point cloud data (denoted as the first target), and the target that only exists in the second millimeter-wave radar point cloud data (denoted as the second target) can be obtained.

[0079] Exemplarily, assume that the first tracking result includes targets m1, m2, m3, m4, and the second tracking result includes targets n1, n2, n3. Assume that after matching by the Hungarian algorithm, it is known that m1 and n1 are the same target, for example, from the same vehicle, and m2 and n3 are the same target. Then m1 and n1 can be recorded as a group of matched targets here, and m2 and n3 can be recorded as another group of matched targets; it can be seen that m3 and m4 have no matched same targets in the second tracking result, so m3 and m4 are both recorded as the first target; n2 also has no matched same targets in the first tracking result, and n2 is recorded as the second target.

[0080] In this embodiment, by using the Hungarian algorithm to perform associative matching processing on the two types of tracking results, a target matching result is obtained, which can improve the accuracy of the obtained target matching result.

[0081] In the above embodiment, it is mentioned that the target matching result includes the matched targets that are successfully matched in the two types of tracking results. In the following embodiments, when the lidar detection box information and lidar identifier including the matched target are included in the above first tracking result, and the millimeter-wave detection box information and millimeter-wave identifier including the matched target are included in the above second tracking result, the process of how the matched target specifically fuses the two types of tracking results will be described.

[0082] In another embodiment, another target tracking method is provided. Based on the above embodiment, step S206 may include the following step B:

[0083] Step B: If the target matching result includes a matching target, perform fusion processing on the laser detection box information and laser identifier of the matching target, and the millimeter-wave detection box information and millimeter-wave identifier of the matching target to determine the tracking result of the matching target.

[0084] In this step, for the target with a successful target matching result, the lidar data and millimeter-wave radar data corresponding to the successfully matched target can be fused. For the specific data fusion process, optionally, reference can be made to Figure 3 As shown, the above step B may include the following steps:

[0085] S302: Detect whether there are laser identifiers and millimeter-wave identifiers in the historical trajectory of the target to obtain a detection result.

[0086] Among them, the historical trajectory of the target is the motion trajectory obtained by tracking the target using a lidar and a millimeter-wave radar at historical moments. The laser identifier refers to the identifier obtained by marking the target in the first tracking result corresponding to the lidar, for example, it can be denoted as lid; the millimeter-wave radar identifier refers to the identifier obtained by marking the target in the second tracking result corresponding to the millimeter-wave radar, for example, it can be denoted as rid.

[0087] Correspondingly, the laser detection box information corresponds to the target in the first tracking result corresponding to the lidar. The laser detection box information may include the three-dimensional position information of the target (including the length, width, and height of the laser detection box, and the coordinates of the target in the x, y, and z directions), speed, heading angle, etc.; the millimeter-wave detection box information corresponds to the target in the second tracking result corresponding to the millimeter-wave radar. The millimeter-wave detection box information may include the two-dimensional position information of the target (including the length and width of the millimeter-wave detection box, and the coordinates of the target in the x and y directions), speed, heading angle, etc.

[0088] The historical trajectory here may be the trajectory after fusing the motion trajectories of the targets in the above first tracking result and second tracking result at historical moments. Taking the motion trajectory of the target in the first tracking result at historical moments as an example, it may be to perform curve fitting on the position coordinates of a target at each moment to obtain the motion trajectory, and at the same time combine the laser identifier of the target as the historical trajectory of the target. When fusing the historical trajectory in the lidar and the historical trajectory in the millimeter-wave radar of the same target, finally, the laser identifier and millimeter-wave identifier of the target can be retained, and then a historical trajectory is obtained through the fusion of the two historical trajectories.

[0089] After obtaining the matching target in the matching result, it is possible to detect whether there are the same laser identifier and millimeter-wave identifier in the fused historical trajectory corresponding to the matching target to obtain a detection result.

[0090] S304. According to the detection result, perform fusion processing on the laser detection box information and laser identifier of the matching target, and the millimeter-wave detection box information and millimeter-wave identifier of the matching target to determine the tracking result of the matching target.

[0091] In this step, after obtaining the detection result above, if both the laser identifier and millimeter-wave identifier of the matching target are in the historical trajectory of the matching target, obtain the three-dimensional position information from the laser detection box information of the matching target, and obtain information such as speed and heading angle from the millimeter-wave detection box information of the matching target, and update this information together with the laser identifier and millimeter-wave identifier of the matching target to the database. The database may include information such as the motion trajectory, three-dimensional position information, laser identifier, millimeter-wave identifier, and confidence of the matching target to obtain the tracking result of the matching target.

[0092] If the laser identifier of the matching target is in the historical trajectory of the matching target, but the millimeter-wave identifier is not in the historical trajectory of the matching target, it means that the millimeter-wave identifier of the matching target has jumped. Then, the current millimeter-wave identifier can be used to replace the old millimeter-wave identifier in the historical trajectory, and other information can be normally fused and updated to the database. For example, in the historical trajectory, the laser identifier of the matching target is lid = 1, and the millimeter-wave identifier is rid = 1. However, at the current moment, the lid = 1 and rid = 2 of the matching target. Then, rid = 2 can be used to replace the previous rid = 1, and lid = 1 remains unchanged.

[0093] If the millimeter-wave identifier of the matching target is in the historical trajectory of the matching target, but the laser identifier is not in the historical trajectory of the matching target, it means that the laser identifier of the matching target has jumped. Then, the current laser identifier can be used to replace the old laser identifier in the historical trajectory, and other information can be normally fused and updated to the database. For example, in the historical trajectory, the laser identifier of the matching target is lid = 1, and the millimeter-wave identifier is rid = 1. However, at the current moment, the lid = 2 and rid = 1 of the matching target. Then, lid = 2 can be used to replace the previous lid = 1, and rid = 1 remains unchanged.

[0094] If both the laser identifier and millimeter-wave identifier of the matching target are not in the historical trajectory of the matching target, the matching target is regarded as a new target, and a new trajectory is created. At the same time, the three-dimensional position information in the laser detection box information of the matching target at the current moment, the speed in the millimeter-wave detection box information, and the laser identifier and millimeter-wave identifier are used as the tracking result of the matching target, and the tracking result is updated to the database.

[0095] Of course, for the database of the matching target mentioned above, the distance deviation of the matching target at each moment can also be stored therein, that is, the distance deviation between the laser detection box information and the millimeter-wave detection box information of the matching target can be calculated. For example, the distance deviation can be equivalent to the Euclidean distance. Then, when updating the database above, the distance deviation of the matching target can also be updated. Generally, a matching target only corresponds to one distance deviation, which can be continuously updated.

[0096] In this embodiment, by performing fusion processing on the laser detection box information and laser identifier of the matching target, and the millimeter-wave detection box information and millimeter-wave identifier of the matching target, the tracking result of the matching target is determined. In this way, the corresponding tracking result of the matching target can be obtained by combining the information of the matching target in the two types of tracking results, and the respective advantages of the two types of tracking results can be combined, thereby improving the accuracy of the obtained tracking result.

[0097] In the above embodiment, it is mentioned that the target matching result includes the first target and the second target that fail to match in the two types of tracking results. In the following embodiments, when the first tracking result includes the first identifier and the first detection box information of the first target, and the second tracking result includes the second identifier and the second detection box information of the second target, the specific process of determining the tracking results of the first target and the second target respectively will be described.

[0098] In another embodiment, another target tracking method is provided. On the basis of the above embodiment, S206 may include the following steps C and D:

[0099] Step C, if the target matching result includes the first target, determine the tracking result of the first target according to the first identifier, the first detection box information, and the second tracking result.

[0100] In this step, for the first target that fails to match in the first tracking result of the lidar, the tracking result of the first target can be obtained by combining its corresponding first identifier, first detection box information, and second tracking result.

[0101] Step D, if the target matching result includes the second target, determine the tracking result of the second target according to the second identifier, the second detection box information, and the first tracking result.

[0102] In this step, for the second target that fails to match in the second tracking result of the millimeter-wave radar, the tracking result of the second target can be obtained by combining its corresponding second identifier, second detection box information, and first tracking result.

[0103] In this embodiment, by fusing the lidar data and millimeter-wave radar data of the target with a failed match and determining its corresponding tracking result, it is also possible to fuse the tracking results of the target with a failed match, improving the accuracy of the obtained tracking result.

[0104] The following embodiments illustrate the process of determining the tracking result of the first target. In another embodiment, another target tracking method is provided. On the basis of the above embodiment, as Figure 4 shown, the above step C may include the following steps:

[0105] S402, Detect whether there is a first identifier in the preset target historical trajectory.

[0106] In this step, similar to S302 above, it is possible to detect whether there is a corresponding first identifier (i.e., the identifier corresponding to the lidar data) in the fused historical trajectory corresponding to the first target, and obtain a detection result.

[0107] S404, If there is a first identifier, discard the second tracking result, and use the first identifier and the first detection box information as the tracking result of the first target.

[0108] In this step, if there is a corresponding first identifier in the historical trajectory of the first target detected for the first time, then in order to improve the accuracy of determination, it is possible to detect multiple times whether there is a corresponding first identifier in the historical trajectory of the first target. The number of detections can set a threshold according to the actual situation. When the number of detected existences is greater than the threshold, it can be determined that there is indeed a corresponding first identifier in the historical trajectory of the first target. Then, the first detection box information, the first identifier corresponding to the first target, and the trajectory of the first target at the historical moment can be directly used as the tracking result of the first target. At the same time, discard the corresponding second tracking result, and set the lifespan corresponding to the first target to 0. Here, the lifespan refers to the time when the first target is away from the last successful match, and a lifespan of 0 means that the first target is relatively close to the time of the last successful match.

[0109] In this embodiment, by detecting the historical trajectory of the first target in the lidar data that fails to match successfully and using the corresponding first identifier and first detection box information as its tracking result when it exists, it is possible to retain the high-precision advantage of the lidar data and improve the accuracy of the obtained tracking result.

[0110] The following embodiments illustrate the process of determining the tracking result of the first target. In another embodiment, another target tracking method is provided. On the basis of the above embodiment, as Figure 5 shown, the above step D may include the following steps:

[0111] S502, detect whether there is a second identifier in the preset target historical trajectory.

[0112] In this step, similar to S402 above, it is possible to detect whether there is a corresponding second identifier (i.e., the identifier corresponding to the millimeter-wave radar data) in the fused historical trajectory corresponding to the second target, and obtain a detection result.

[0113] S504, if there is a second identifier, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result.

[0114] In this step, if there is a corresponding second identifier in the historical trajectory of the second target detected for the first time, then to improve the accuracy of determination, it is possible to detect multiple times whether there is a corresponding second identifier in the historical trajectory of the second target. The number of detections can set a number threshold according to the actual situation. When the number of detections of existence is greater than the number threshold, it can be determined that there is indeed a corresponding second identifier in the historical trajectory of the second target. After that, the latest distance deviation of the matching target can be obtained through the tracking result of the matching target in S304 above. Here, it is assumed that there is only one distance deviation of the matching target, then the distance deviation of this one matching target can be directly used as the latest distance deviation here; assuming that the distance deviations of multiple matching targets are obtained, then after averaging the distance deviations of these multiple matching targets, the obtained average value can be used as the latest distance deviation here.

[0115] S506, determine the tracking result of the second target according to the second detection box information and the latest distance deviation.

[0116] In this step, after obtaining the latest distance deviation above, optionally, the second detection box information can be subtracted by the latest distance deviation, and combined with the second identifier to obtain the tracking result of the second target.

[0117] That is to say, the millimeter-wave detection box information of the second target can be subtracted by the latest distance deviation to obtain new millimeter-wave detection box information, and the second identifier of the second target, the new millimeter-wave detection box information, and the trajectory of the second target at the historical moment are used together as the tracking result of the second target. At the same time, the corresponding first tracking result is discarded, and the lifespan corresponding to the second target is set to 0. Here, the lifespan refers to the time when the second target was last successfully matched. A lifespan of 0 means that the time when the second target was last successfully matched is relatively short.

[0118] In this embodiment, by detecting the historical trajectory of the second target in the millimeter-wave radar data that fails to match successfully, and when it exists, taking the corresponding millimeter-wave identifier and detection box information as its tracking result, the long-distance measurement advantage of the millimeter-wave radar data can be retained, and the accuracy of the obtained tracking result can be improved. Further, by subtracting the distance deviation from the detection box information of the second target to obtain new detection box information, the accuracy of the tracking result of the second target obtained can be further improved.

[0119] To facilitate a more detailed description of the technical solution of this application, a detailed embodiment is given below for further illustration. On the basis of the above embodiment, see Figure 6 As shown, the method may include the following steps:

[0120] S1, Perform spatio-temporal synchronization on the first point cloud data collected by the lidar and the second point cloud data collected by the millimeter-wave radar.

[0121] S2, Perform target tracking processing on the synchronized first point cloud data and second point cloud data respectively to obtain a first tracking result and a second tracking result.

[0122] S3, Use the Hungarian algorithm to perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result; the target matching result includes the matching targets that match successfully, the first targets in the first tracking result that do not match successfully, and the second targets in the second tracking result that do not match successfully.

[0123] S4, If the target matching result includes matching targets, detect whether there are laser identifiers and millimeter-wave identifiers of the matching targets in the historical trajectory of the targets to obtain a detection result.

[0124] S5, According to the detection result, perform fusion processing on the laser detection box information and laser identifier of the matching targets, and the millimeter-wave detection box information and millimeter-wave identifier of the matching targets to determine the tracking result of the matching targets.

[0125] S6, If the target matching result includes the first target, detect whether there is a first identifier in the preset target historical trajectory.

[0126] S7, If the first identifier exists, discard the second tracking result, and take the first identifier and the first detection box information as the tracking result of the first target.

[0127] S8, If the target matching result includes the second target, detect whether there is a second identifier in the preset target historical trajectory.

[0128] S9. If there is a second identifier, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result.

[0129] S10. Subtract the latest distance deviation from the second detection box information, and combine it with the second identifier to obtain the tracking result of the second target.

[0130] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps in other steps.

[0131] Based on the same inventive concept, an embodiment of the present application also provides a target tracking device for implementing the above-mentioned target tracking method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the target tracking device provided below can refer to the limitations on the target tracking method in the above text, and will not be repeated here.

[0132] In one embodiment, as Figure 7 shown, a target tracking device is provided, including: an acquisition module 11, a matching module 12, and a determination module 13, where:

[0133] The acquisition module 11 is configured to obtain the first tracking result and the second tracking result of tracking the target by using the synchronized first point cloud data and the second point cloud data respectively; the first point cloud data is the data collected by the lidar, and the second point cloud data is the data collected by the millimeter wave radar;

[0134] The matching module 12 is configured to perform target association matching processing on the first tracking result and the second tracking result to determine the target matching result;

[0135] The determination module 13 is configured to perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0136] In another embodiment, another target tracking device is provided. On the basis of the above embodiment, the matching module 12 may include a matching unit, which is configured to perform target association matching processing on the first tracking result and the second tracking result by using the Hungarian algorithm to determine a target matching result. The target matching result includes matching targets that are successfully matched, first targets that are not successfully matched, and second targets that are not successfully matched. The matching targets are the same targets that exist in both the first tracking result and the second tracking result. The first targets are the targets in the first tracking result, and the second targets are the targets in the second tracking result.

[0137] In another embodiment, another target tracking device is provided. On the basis of the above embodiment, the first tracking result includes laser detection box information and laser identification of the matching target, and the second tracking result includes millimeter-wave detection box information and millimeter-wave identification of the matching target. The determination module 13 may include a first determination unit, which is configured to, if the target matching result includes a matching target, perform fusion processing on the laser detection box information and laser identification of the matching target, and the millimeter-wave detection box information and millimeter-wave identification of the matching target to determine the tracking result of the matching target.

[0138] Optionally, the first determination unit may include a first detection subunit and a first determination subunit, where:

[0139] The first detection subunit is configured to detect whether there are laser identification and millimeter-wave identification in the historical trajectory of the target to obtain a detection result. The historical trajectory of the target is a motion trajectory obtained by tracking the target with a lidar and a millimeter-wave radar at a historical moment.

[0140] The first determination subunit is configured to, according to the detection result, perform fusion processing on the laser detection box information and laser identification of the matching target, and the millimeter-wave detection box information and millimeter-wave identification of the matching target to determine the tracking result of the matching target.

[0141] In another embodiment, another target tracking device is provided. On the basis of the above embodiment, the first tracking result includes a first identification and first detection box information of the first target, and the second tracking result includes a second identification and second detection box information of the second target. The determination module 13 may include a second determination unit and a third determination unit, where:

[0142] The second determination unit is configured to, if the target matching result includes the first target, determine the tracking result of the first target according to the first identification, the first detection box information, and the second tracking result.

[0143] A third determination unit, configured to, if the target matching result includes a second target, determine the tracking result of the second target according to the second identifier, the second detection box information, and the first tracking result.

[0144] Optionally, the above-mentioned second determination unit may include a second detection subunit and a second determination subunit, where:

[0145] The second detection subunit is configured to detect whether a first identifier exists in a preset target historical trajectory;

[0146] The second determination subunit is configured to, if the first identifier exists, discard the second tracking result and use the first identifier and the first detection box information as the tracking result of the first target.

[0147] Optionally, the above-mentioned third determination unit may include a third detection subunit, a deviation determination subunit, and a third determination subunit, where:

[0148] The third detection subunit is configured to detect whether a second identifier exists in a preset target historical trajectory;

[0149] The deviation determination subunit is configured to, if the second identifier exists, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result;

[0150] The third determination subunit is configured to determine the tracking result of the second target according to the second detection box information and the latest distance deviation.

[0151] Optionally, the above-mentioned third determination subunit is specifically configured to subtract the latest distance deviation from the second detection box information and combine it with the second identifier to obtain the tracking result of the second target.

[0152] Each module in the above-mentioned target tracking device can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0153] In one embodiment, a computer device is provided. Taking this computer device as a terminal as an example, its internal structure diagram can be as Figure 8As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a target tracking method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0154] Those skilled in the art can understand that Figure 8 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0155] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0156] Obtain the first tracking result and the second tracking result of respectively tracking the target with the synchronized first point cloud data and the second point cloud data; the first point cloud data is the data collected by a lidar, and the second point cloud data is the data collected by a millimeter-wave radar; perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result; perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0157] In one embodiment, when the processor executes the computer program, the following steps are also implemented:

[0158] Use the Hungarian algorithm to perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result; wherein, the target matching result includes the matching targets that are successfully matched, the first targets that are not successfully matched, and the second targets that are not successfully matched. The matching targets are the same targets that exist in both the first tracking result and the second tracking result. The first targets are the targets in the first tracking result, and the second targets are the targets in the second tracking result.

[0159] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0160] If the target matching result includes a matching target, perform fusion processing on the laser detection box information and laser identifier of the matching target, and the millimeter-wave detection box information and millimeter-wave identifier of the matching target to determine the tracking result of the matching target.

[0161] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0162] Detect whether there are laser identifiers and millimeter-wave identifiers in the historical trajectory of the target to obtain a detection result; the historical trajectory of the target is the motion trajectory obtained by tracking the target using a lidar and a millimeter-wave radar at a historical moment; according to the detection result, perform fusion processing on the laser detection box information and laser identifier of the matching target, and the millimeter-wave detection box information and millimeter-wave identifier of the matching target to determine the tracking result of the matching target.

[0163] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0164] If the target matching result includes a first target, determine the tracking result of the first target according to the first identifier, the first detection box information, and the second tracking result; if the target matching result includes a second target, determine the tracking result of the second target according to the second identifier, the second detection box information, and the first tracking result.

[0165] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0166] Detect whether there is a first identifier in the preset target historical trajectory; if there is a first identifier, discard the second tracking result and use the first identifier and the first detection box information as the tracking result of the first target.

[0167] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0168] Detect whether there is a second identifier in the preset target historical trajectory; if there is a second identifier, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result; according to the second detection box information and the latest distance deviation, determine the tracking result of the second target.

[0169] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0170] Subtract the latest distance deviation from the second detection box information and combine it with the second identifier to obtain the tracking result of the second target.

[0171] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0172] Obtain the first tracking result and the second tracking result of respectively tracking a target using the synchronized first point cloud data and second point cloud data; the first point cloud data is data collected by a lidar, and the second point cloud data is data collected by a millimeter-wave radar; perform target association and matching processing on the first tracking result and the second tracking result to determine a target matching result; perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0173] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0174] Use the Hungarian algorithm to perform target association and matching processing on the first tracking result and the second tracking result to determine a target matching result; wherein, the target matching result includes matching targets that are successfully matched, first targets that are not successfully matched, and second targets that are not successfully matched. The matching targets are the same targets that exist in both the first tracking result and the second tracking result. The first targets are the targets in the first tracking result, and the second targets are the targets in the second tracking result.

[0175] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0176] If the target matching result includes matching targets, perform fusion processing on the laser detection box information and laser identifier of the matching targets, and the millimeter-wave detection box information and millimeter-wave identifier of the matching targets to determine the tracking result of the matching targets.

[0177] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0178] Detect whether there are laser identifiers and millimeter-wave identifiers in the historical trajectory of the target to obtain a detection result; the historical trajectory of the target is a motion trajectory obtained by tracking the target using a lidar and a millimeter-wave radar at a historical moment; according to the detection result, perform fusion processing on the laser detection box information and laser identifier of the matching targets, and the millimeter-wave detection box information and millimeter-wave identifier of the matching targets to determine the tracking result of the matching targets.

[0179] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0180] If the target matching result includes a first target, determine the tracking result of the first target according to the first identifier, the first detection box information, and the second tracking result; if the target matching result includes a second target, determine the tracking result of the second target according to the second identifier, the second detection box information, and the first tracking result.

[0181] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0182] Detect whether a first identifier exists in a preset target historical trajectory; if the first identifier exists, discard the second tracking result, and use the first identifier and the first detection box information as the tracking result of the first target.

[0183] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0184] Detect whether a second identifier exists in a preset target historical trajectory; if the second identifier exists, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result; according to the second detection box information and the latest distance deviation, determine the tracking result of the second target.

[0185] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0186] Subtract the latest distance deviation from the second detection box information, and combine it with the second identifier to obtain the tracking result of the second target.

[0187] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0188] Obtain the first tracking result and the second tracking result of respectively tracking a target with the synchronized first point cloud data and second point cloud data; the first point cloud data is data collected by a lidar, and the second point cloud data is data collected by a millimeter wave radar; perform target association and matching processing on the first tracking result and the second tracking result to determine a target matching result; perform fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target.

[0189] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0190] The Hungarian algorithm is used to perform target association and matching processing on the first tracking result and the second tracking result to determine the target matching result. Among them, the target matching result includes the matching targets with successful matching, the first targets with unsuccessful matching, and the second targets with unsuccessful matching. The matching targets are the same targets existing in both the first tracking result and the second tracking result. The first targets are the targets in the first tracking result, and the second targets are the targets in the second tracking result.

[0191] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0192] If the target matching result includes matching targets, then fusion processing is performed on the laser detection box information and laser identification of the matching targets, and the millimeter-wave detection box information and millimeter-wave identification of the matching targets to determine the tracking result of the matching targets.

[0193] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0194] Detect whether there are laser identification and millimeter-wave identification in the historical trajectory of the target to obtain a detection result. The historical trajectory of the target is the motion trajectory obtained by tracking the target with a lidar and a millimeter-wave radar at a historical moment. According to the detection result, fusion processing is performed on the laser detection box information and laser identification of the matching targets, and the millimeter-wave detection box information and millimeter-wave identification of the matching targets to determine the tracking result of the matching targets.

[0195] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0196] If the target matching result includes the first targets, then the tracking result of the first targets is determined according to the first identification, the first detection box information, and the second tracking result. If the target matching result includes the second targets, then the tracking result of the second targets is determined according to the second identification, the second detection box information, and the first tracking result.

[0197] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0198] Detect whether there is a first identification in the preset target historical trajectory. If there is a first identification, then discard the second tracking result and use the first identification and the first detection box information as the tracking result of the first targets.

[0199] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0200] Check whether a second identifier exists in the preset target historical trajectory; if the second identifier exists, determine the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result; determine the tracking result of the second target according to the second detection box information and the latest distance deviation.

[0201] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0202] Subtract the latest distance deviation from the second detection box information, and combine it with the second identifier to obtain the tracking result of the second target.

[0203] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0204] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0205] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0206] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A target tracking method, characterized in that, The method includes: Obtaining a first tracking result and a second tracking result of respectively tracking a target with the synchronized first point cloud data and second point cloud data; the first point cloud data is data collected by a lidar, and the second point cloud data is data collected by a millimeter-wave radar; Performing target association and matching processing on the first tracking result and the second tracking result to determine a target matching result; Performing fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target; The target matching result includes a matching target with successful matching, a first target with unsuccessful matching, and a second target with unsuccessful matching. The matching target is the same target existing in both the first tracking result and the second tracking result. The first target is a target in the first tracking result, and the second target is a target in the second tracking result. The first tracking result includes laser detection box information and laser identification of the matching target, a first identification and first detection box information of the first target. The second tracking result includes millimeter-wave detection box information and millimeter-wave identification of the matching target, a second identification and second detection box information of the second target; The performing fusion processing on the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target includes: if the target matching result includes the matching target, detecting whether the laser identification and the millimeter-wave identification exist in the historical trajectory of the target to obtain a detection result; the historical trajectory of the target is a motion trajectory obtained by tracking the target with a lidar and a millimeter-wave radar at a historical moment; according to the detection result, performing fusion processing on the laser detection box information and laser identification of the matching target, and the millimeter-wave detection box information and millimeter-wave identification of the matching target to determine the tracking result of the matching target; The method further includes: in the case where the detection result is that the laser identification in the first tracking result exists in the historical trajectory of the target and the millimeter-wave identification in the second tracking result does not exist, replacing the millimeter-wave identification in the historical trajectory of the target with the millimeter-wave identification in the second tracking result; In the case where the detection result is that the millimeter-wave identification in the second tracking result exists in the historical trajectory of the target and the laser identification in the first tracking result does not exist, replacing the laser identification in the historical trajectory of the target with the laser identification in the first tracking result; In the case where the detection result is that neither the laser identification in the first tracking result nor the millimeter-wave identification in the second tracking result exists in the historical trajectory of the target, taking the matching target as a new target and creating a new trajectory.

2. The method according to claim 1, wherein The performing target association and matching processing on the first tracking result and the second tracking result to determine a target matching result includes: Using the Hungarian algorithm to perform target association and matching processing on the first tracking result and the second tracking result to determine a target matching result.

3. The method according to claim 2, wherein The first tracking result includes the laser detection box information and laser identification of the matching target, the first identification and first detection box information of the first target, and the second tracking result includes the millimeter-wave detection box information and millimeter-wave identification of the matching target, the second identification and second detection box information of the second target; The fusing the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target includes: If the target matching result includes the first target, determining the tracking result of the first target according to the first identification, the first detection box information, and the second tracking result; If the target matching result includes the second target, determining the tracking result of the second target according to the second identification, the second detection box information, and the first tracking result.

4. The method according to claim 3, wherein The determining the tracking result of the first target according to the first identification, the first detection box information, and the second tracking result includes: Detecting whether the first identification exists in a preset target historical trajectory; If the first identification exists, discarding the second tracking result and using the first identification and the first detection box information as the tracking result of the first target.

5. The method according to claim 3, characterized in that The determining the tracking result of the second target according to the second identification, the second detection box information, and the first tracking result includes: Detecting whether the second identification exists in a preset target historical trajectory; If the second identification exists, determining the latest distance deviation according to the target historical trajectory; the latest distance deviation represents the distance between the matching targets in the first tracking result and the second tracking result; Determining the tracking result of the second target according to the second detection box information and the latest distance deviation.

6. The method according to claim 5, characterized in that, The determining the tracking result of the second target according to the second detection box information and the latest distance deviation includes: Subtracting the latest distance deviation from the second detection box information and combining it with the second identification to obtain the tracking result of the second target.

7. A target tracking device, characterized in that, The device includes: An obtaining module, configured to obtain the first tracking result and the second tracking result of respectively tracking a target from the synchronized first point cloud data and second point cloud data; the first point cloud data is data collected by a lidar, and the second point cloud data is data collected by a millimeter-wave radar; A matching module, configured to perform target association matching processing on the first tracking result and the second tracking result to determine the target matching result; A determining module, configured to fuse the first tracking result and the second tracking result according to the target matching result to determine the tracking result of the target; The target matching result includes a matching target with successful matching, a first target with unsuccessful matching, and a second target with unsuccessful matching. The matching target is the same target that exists in both the first tracking result and the second tracking result. The first target is a target in the first tracking result, and the second target is a target in the second tracking result. The first tracking result includes laser detection box information and a laser identifier of the matching target, a first identifier and first detection box information of the first target. The second tracking result includes millimeter-wave detection box information and a millimeter-wave identifier of the matching target, a second identifier and second detection box information of the second target. The determination module includes a first determination unit. The first determination unit includes a first detection subunit. The first detection subunit is configured to detect whether the laser identifier and the millimeter-wave identifier exist in the historical trajectory of the target when the target matching result includes the matching target, and obtain a detection result. The historical trajectory of the target is a motion trajectory obtained by tracking the target using a lidar and a millimeter-wave radar at a historical moment. According to the detection result, perform fusion processing on the laser detection box information and the laser identifier of the matching target, and the millimeter-wave detection box information and the millimeter-wave identifier of the matching target, and determine the tracking result of the matching target. The determination module is further configured to, when the detection result is that the laser identifier in the first tracking result exists in the historical trajectory of the target and the millimeter-wave identifier in the second tracking result does not exist, replace the millimeter-wave identifier in the historical trajectory of the target with the millimeter-wave identifier in the second tracking result; when the detection result is that the millimeter-wave identifier in the second tracking result exists in the historical trajectory of the target and the laser identifier in the first tracking result does not exist, replace the laser identifier in the historical trajectory of the target with the laser identifier in the first tracking result; when the detection result is that neither the laser identifier in the first tracking result nor the millimeter-wave identifier in the second tracking result exists in the historical trajectory of the target, regard the matching target as a new target and create a new trajectory.

8. The device according to claim 7, wherein, The matching module is configured to perform target association and matching processing on the first tracking result and the second tracking result by using the Hungarian algorithm to determine the target matching result.

9. A roadside base station, comprising a lidar and a millimeter-wave radar, and a computer device connected to both the lidar and the millimeter-wave radar, the computer device including a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

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

Images