An object matching method and device, a terminal device, and a storage medium

Abstract

The application is suitable for the field of radar detection technology, and provides a matching method and device of objects, a terminal device and a storage medium. The matching method of objects specifically comprises: obtaining feature vectors generated by an upstream radar and a downstream radar respectively, the feature vectors being used to represent the relative relationship between objects in a target area; matching each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively; determining the start objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar respectively as paired objects, and determining the end objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar respectively as paired objects. The embodiments of the application can avoid the problem of false matching when the objects are dense, and further avoid the problem of accuracy reduction of maintaining the continuity of the track.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of radar detection, and particularly relates to a matching method and device for objects, a terminal device and a storage medium. BACKGROUND

[0002] In the field of intelligent traffic and autonomous navigation, object detection is often achieved through a radar group. The radar group is generally composed of multiple radars. When an object passes through the detection areas of two adjacent radars in the radar group, object matching is often needed to associate the detection data of the same object from different radars, so as to maintain the continuity of the trajectory of the object in the detection area of the radar group, so as to analyze the motion trajectory of the object.

[0003] The current object matching method is prone to mis-matching when the objects in the detection area of the radar are relatively dense, thereby reducing the accuracy of maintaining the trajectory continuity. SUMMARY

[0004] The embodiments of the application provide a matching method and device for objects, a terminal device and a storage medium, which can avoid the mis-matching problem when the objects are dense, thereby avoiding the problem of reducing the accuracy of maintaining the trajectory continuity.

[0005] The first aspect of the embodiments of the application provides a matching method for objects, comprising:

[0006] obtaining feature vectors respectively generated by an upstream radar and a downstream radar, the feature vectors being used to represent the relative relationship between objects located in a target area two by two;

[0007] matching each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively;

[0008] determining the start objects respectively corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that are matched with each other as paired objects, and determining the end objects respectively corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that are matched with each other as paired objects.

[0009] The second aspect of the embodiments of the application provides a matching device for objects, comprising:

[0010] a data acquisition unit configured to obtain feature vectors respectively generated by an upstream radar and a downstream radar, the feature vectors being used to represent the relative relationship between objects located in a target area two by two;

[0011] a data matching unit configured to match each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively;

[0012] The object matching unit is configured to determine the start point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as matched objects, and determine the end point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as matched objects.

[0013] The third aspect of the embodiments of the present application provides a terminal device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0014] The fourth aspect of the embodiments of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the above method.

[0015] The fifth aspect of the embodiments of the present application provides a computer program product, which, when executed on a terminal device, causes the terminal device to execute the matching method of any one of the above first aspect.

[0016] In the embodiments of the present application, by acquiring the feature vectors generated by the upstream radar and the downstream radar respectively, and matching each feature vector generated by the downstream radar with each feature vector generated by the upstream radar, since the feature vectors are used to represent the relative relationship between two objects located in the target area, compared with using the feature data of a single object for matching, the relative relationship between two objects detected by different radars does not change due to the installation angle error of the radars or the coordinate conversion error, therefore, determining the start point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as matched objects, and determining the end point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as matched objects, can avoid the mismatching problem when the objects are dense, and further improve the accuracy of maintaining the trajectory continuity. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments or prior art description will be briefly introduced below, and obviously, the drawings in the following description can only be some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Figure 1 is an implementation flow diagram of the object matching method provided by the embodiments of the present application;

[0019] Figure 2 is a schematic diagram of an overlapping region provided by an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of determining an overlapping region provided by an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of determining a feature vector provided by an embodiment of the present application;

[0022] Figure 5 is a schematic diagram of an implementation flow of a simple matching mode provided by an embodiment of the present application;

[0023] Figure 6 is a schematic diagram of calculating a relative lane position provided by an embodiment of the present application;

[0024] Figure 7 is a schematic diagram of a structure of an object matching device provided by an embodiment of the present application;

[0025] Figure 8 is a schematic diagram of a structure of a terminal device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the objects, technical solutions and advantages of the present application clearer, further detailed description will be made to the present application 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. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the present application.

[0027] In the field of intelligent transportation, autonomous navigation, etc., the detection of objects is often realized by radar groups. A radar group is generally composed of multiple radars. When an object passes through the detection areas of two adjacent radars in the radar group, object matching is often needed to be performed to associate the detection data of the same object from different radars, so as to maintain the continuity of the trajectory of the object in the detection area of the radar group, so as to analyze the motion trajectory of the object.

[0028] The current object matching mode generally needs to compare the same feature data detected by two adjacent radars respectively to determine whether the data detected by the two radars respectively belong to the same object.

[0029] For example, the position data detected by an upstream radar and the position data detected by a downstream radar are matched. If the deviation between the two position data is within a certain distance range, it can be confirmed that the objects corresponding to the two position data are the same object.

[0030] However, it is found in actual application that when the objects in the detection area of the upstream radar and the downstream radar are relatively dense, the installation angle error or the coordinate conversion error of the radar will cause errors in the respective features collected by the upstream radar and the downstream radar, and further cause the occurrence of the mis-matching problem.

[0031] For example, when detecting a vehicle A with actual world coordinates (1, 0) and a vehicle B with actual world coordinates (3, 0), due to the existence of the upstream radar self-error, the upstream radar detects the world coordinates of the vehicle A as (0, 0) and the world coordinates of the vehicle B as (2, 0). Due to the existence of the downstream radar self-error, the downstream radar detects the world coordinates of the vehicle A as (2, 0) and the world coordinates of the vehicle B as (4, 0). At this time, the vehicle A detected by the downstream radar and the vehicle B detected by the upstream radar will be confirmed as the same object, that is, the mis-matching problem occurs.

[0032] The occurrence of the mis-matching problem will further cause the accuracy of the trajectory continuity maintenance to decrease.

[0033] The technical solution provided in the present application is exactly to solve the above technical problem.

[0034] In order to illustrate the technical solution of the present application, the following will be illustrated by specific embodiments.

[0035] Figure 1 An implementation flowchart of a matching method of an object provided in an embodiment of the present application is shown, which can be applied to a terminal device and can be applicable to a situation in which the mis-matching problem occurring when the objects are dense needs to be reduced and the accuracy of the trajectory continuity maintenance needs to be improved.

[0036] The terminal device can be any radar in a radar group, the radar is provided with a processor and has a certain data processing capability; the terminal device can also be an upper computer capable of acquiring radar data, and the upper computer can be a computer, a smart phone or other smart devices.

[0037] It should be noted that the radar group can be installed in different application scenarios, and the detected objects can also be different according to the different application scenarios.

[0038] In some specific application scenarios, the staff can deploy a road perception system on the intelligent highway. The road perception system can be composed of an upper computer and multiple radars, and can detect the features and trajectories of vehicles on the intelligent highway. Specifically, a gantry can be arranged at a certain interval on the intelligent highway. A radar is arranged on each gantry. The radar on the gantry can be used to detect the data of objects on the two-way road within a certain range of the gantry, and the detection range can cover one or more lanes of the intelligent highway. A roadside pole can be arranged at a certain interval between two adjacent gantries. A radar is also arranged on the roadside pole. The radar on the roadside pole can be used to detect the data of objects on the road within a certain range of the roadside pole. The detection areas of the radars on the roadside poles located on the same side of the road can cover the entire road section of the intelligent highway. The matching method provided in the present application can be applied to the radars on the gantries, the radars on the roadside poles, and the upper computer in the road perception system.

[0039] It should be noted that the above is only an example of a specific application scenario. As long as there are at least two radars in the scenario, and the detection areas of the two radars overlap, the patent solution can be applied. In the embodiments of the present application, the type of radar can be selected according to actual conditions, for example, it can be a millimeter wave radar, a laser radar, etc.

[0040] Specifically, the matching method of the above object can include the following steps S101 to S103.

[0041] Step S101, obtaining the feature vectors generated by the upstream radar and the downstream radar respectively.

[0042] In the embodiments of the present application, the upstream radar and the downstream radar are two adjacent radars. In some specific embodiments, when the object passes through the two adjacent radars, the first radar passed through can be referred to as the upstream radar, and the second radar passed through can be referred to as the downstream radar.

[0043] In the embodiments of the present application, the feature vectors are used to represent the relative relationship between each two objects located in the target area.

[0044] The target area can refer to the detection area of the corresponding radar, that is, the feature vector generated by the upstream radar is used to represent the relative relationship between each two objects located in the detection area of the upstream radar, and the feature vector generated by the downstream radar is used to represent the relative relationship between each two objects located in the detection area of the downstream radar.

[0045] The target area can also refer to the overlapping area between the upstream radar and the downstream radar. Please refer to Figure 2The two intersecting circles represent the detection areas of the upstream and downstream radars, respectively, and the overlapping area is the part where the detection areas of the upstream and downstream radars overlap. Figure 2 (As shown in the shaded area). When the target area is an overlapping area, the terminal device only needs to match objects located within the overlapping area. Correspondingly, when transmitting data, the radar can only transmit data of objects within the overlapping area. Therefore, while maintaining the continuity of the object trajectory, it can reduce the amount of data transmitted by the radar and the amount of computation required for object matching.

[0046] Specifically, after deploying the radar group, personnel can input the configuration information of each radar into a host computer, which then distributes the configuration information to each radar within the group. The configuration information may include information related to the radar waveform and the installation location of each radar. When an object passes through an overlapping area, the upstream radar can transmit the object's feature data within the overlapping area to the downstream radar, which then performs object matching. Furthermore, the host computer can also distribute connection information to each radar within the group. This connection information carries the necessary information for establishing communication connections between adjacent radars, enabling upstream radars to establish connections with downstream radars in each data frame, thereby transmitting feature data to the downstream radars.

[0047] Before data transmission, either the host computer or the radar can calculate the overlap area between upstream and downstream radars to filter out objects located within the overlap area from those detected by the upstream and downstream radars. Please refer to [link / reference]. Figure 3 The host computer or radar can obtain the installation distance D of the upstream and downstream radars, the farthest detection range L1 of the upstream radar, and the nearest detection range L2 of the downstream radar to calculate the range of the overlapping area. For the upstream radar, the range of the overlapping area is [D+L2, L1] in the upstream radar coordinate system, and for the downstream radar, the range of the overlapping area is [L2, L1-D] in the downstream radar coordinate system.

[0048] At this point, based on the feature data of objects located in the overlapping area detected by the upstream radar, a feature vector between each pair of objects corresponding to the upstream radar can be generated. Similarly, based on the feature data of objects located in the overlapping area detected by the downstream radar, a feature vector between each pair of objects corresponding to the downstream radar can be generated.

[0049] In some embodiments of this application, the aforementioned feature vector may include at least one of the position vector difference and velocity vector difference between any two objects.

[0050] Specifically, the feature data detected by the upstream and downstream radars can include absolute position, speed, relative lane position and the like of the object. Since the position of the object detected by each radar is relative to the radar coordinate system of the radar, before matching the object, a unified coordinate system needs to be established for the upstream and downstream radars, that is, the position of the object is converted from the radar coordinate system to the geodetic coordinate system by using the conversion relationship to obtain the absolute position coordinates of the object.

[0051] The conversion relationship can be obtained by calibration in advance. The calibration process is as follows: the absolute position (which can refer to latitude and longitude) of the radar installation point is obtained, the coordinate position of the calibration point in the radar coordinate system is determined according to the position of the calibration point measured by the radar, and then the conversion relationship between the radar coordinate system and the geodetic coordinate system can be calculated. The calibration point can be a mobile reference object (such as a reference vehicle in motion) in the application scenario, or a fixed reference object (such as a signboard) set in the application scenario.

[0052] Then, the terminal device can generate a feature vector using the feature data after coordinate conversion.

[0053] The position vector difference in the feature vector can represent the vector difference between the absolute positions of the two objects detected by the radar, which can be specifically decomposed into the position difference in the x direction and the position difference in the y direction; the speed vector difference in the feature vector can represent the vector difference between the speeds of the two objects detected by the radar, which can be specifically decomposed into the speed difference in the x direction and the speed difference in the y direction.

[0054] The x direction and the y direction can refer to the x axis direction and the y axis direction of the geodetic coordinate system, and in some application scenarios, the x direction can refer to the driving direction of the vehicle on the lane, and correspondingly, the y direction can refer to the direction perpendicular to the x direction in the lane plane.

[0055] Please refer to Figure 4 , the coordinate vector can be represented as (Δx, Δy, Δv x , Δv y ), wherein Δx represents the absolute position difference between the two objects in the x direction, Δy represents the absolute position difference between the two objects in the y direction, Δv x represents the speed difference in the x direction, and Δv y represents the speed difference in the y direction.

[0056] It should be noted that the above object can be selected according to the actual scene, which can refer to a vehicle, a pedestrian, a robot or other objects.

[0057] Step S102, match each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively.

[0058] In the embodiments of the present application, the terminal device can match the feature vector formed between the two objects detected by the downstream radar with the feature vector formed between the two objects detected by the upstream radar.

[0059] When performing the matching, the terminal device can sequentially take any one of the feature vectors generated by the downstream radar which has not been matched as a current vector, and each time the current vector is updated, the current vector is matched with each of the feature vectors generated by the upstream radar to obtain the matching result between the feature vectors generated by the upstream radar and the feature vectors generated by the downstream radar.

[0060] In some embodiments of the present application, the terminal device can calculate the difference degree between each of the feature vectors generated by the downstream radar and each of the feature vectors generated by the upstream radar, and take the two feature vectors with the difference degree less than a degree threshold as the mutually matched feature vectors.

[0061] The difference degree between the feature vectors can be represented as the Euclidean distance between the feature vectors, the Mahalanobis distance or a value calculated based on other difference degree algorithms or similarity algorithms. The smaller the difference degree between the feature vectors, the more similar the two feature vectors, that is, the two objects corresponding to one of the feature vectors and the two objects corresponding to the other feature vector are more likely to be the same objects.

[0062] Therefore, the terminal device can take the two feature vectors with the difference degree less than the degree threshold as the matched feature vectors, wherein the degree threshold can be set according to actual conditions.

[0063] Specifically, taking the feature vectors represented as and as examples, the terminal device can calculate the Euclidean distance d between the two feature vectors If the Euclidean distance d between the two feature vectors is less than a Euclidean distance threshold Thre1, it can be confirmed that the feature vector and the feature vector are mutually matched.

[0064] In step S103, the terminal device determines the start point objects corresponding to the mutually matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the paired objects, and determines the end point objects corresponding to the mutually matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the paired objects.

[0065] In the embodiments of the present application, when calculating the feature vectors, the terminal device can generate the feature vectors in the order of the small coordinate pointing to the large coordinate or the large coordinate pointing to the small coordinate in the geodetic coordinate system. Correspondingly, the start point object is the object corresponding to the start point of the vector among the two objects corresponding to the feature vector, and the end point object is the object corresponding to the end point of the vector among the two objects corresponding to the feature vector.

[0066] Correspondingly, the terminal device can determine the start point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the pairing objects, and determine the end point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the pairing objects.

[0067] The pairing objects can refer to the same object.

[0068] For ease of understanding, specific examples are used for illustration. It is assumed that there are object A and object B in the target area. The upstream radar detects feature data of object A1, detects feature data of object B1, and further generates feature vector A1B1; the downstream radar detects feature data of object A2, detects feature data of object B2, and further generates feature vector A2B2. After step S102, the terminal device confirms that the feature vector A1B1 and the feature vector A2B2 are matched with each other, and then can determine the object A1 detected by the upstream radar (i.e., the start point object of the feature vector A1B1) and the object A2 detected by the downstream radar (i.e., the start point object of the feature vector A2B2) as the pairing objects, and determine the object B1 detected by the upstream radar (i.e., the end point object of the feature vector A1B1) and the object B2 detected by the downstream radar (i.e., the end point object of the feature vector A2B2) as the pairing objects.

[0069] In the embodiments of the present application, by acquiring the feature vectors generated by the upstream radar and the downstream radar respectively, and matching each feature vector generated by the downstream radar with each feature vector generated by the upstream radar, since the feature vectors are used to represent the relative relationship between two objects in the target area, compared with using the feature data of a single object for matching, the relative relationship between two objects collected by different radars does not change due to the installation angle error of the radars or the coordinate conversion error. Therefore, determining the start point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the pairing objects, and determining the end point objects corresponding to the matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as the pairing objects, can avoid the mismatching problem when the objects are dense, and further improve the accuracy of maintaining the trajectory continuity.

[0070] Continuing to take the vehicle A detecting the actual world coordinate (1, 0) and the vehicle B detecting the actual world coordinate (3, 0) as an example, due to the existence of the upstream radar self-error, the world coordinate of the vehicle A detected by the upstream radar is (0, 0), and the world coordinate of the vehicle B detected by the upstream radar is (2, 0), and the upstream radar calculates that the position vector difference between the vehicle A and the vehicle B in the x direction is 2 and the position vector difference in the y direction is 0. Due to the existence of the downstream radar self-error, the world coordinate of the vehicle A detected by the downstream radar is (2, 0), and the world coordinate of the vehicle B detected by the downstream radar is (4, 0), and the downstream radar calculates that the position vector difference between the vehicle A and the vehicle B in the x direction is also 2 and the position vector difference in the y direction is also 0. Through the feature vector, the vehicle A detected by the downstream radar and the vehicle A detected by the upstream radar are confirmed as the pairing objects, and the vehicle B detected by the downstream radar and the vehicle B detected by the upstream radar are confirmed as the pairing objects, and the object matching is successful.

[0071] In actual application, when the object matching is performed by using the feature vector, the situation that a single object is paired with multiple objects can occur. For example, the terminal device confirms that the feature vector A1B1 generated by the upstream radar and the feature vector A2B2 generated by the downstream radar match each other, and the feature vector A1C1 generated by the upstream radar and the feature vector A2B2 generated by the downstream radar match each other. At this time, the object B2 detected by the downstream radar will be confirmed as the pairing object of the B1 and the C1 detected by the upstream radar respectively. This phenomenon is caused by the environmental error when the radar collects the data of a data frame. In view of the above problem, the terminal device can obtain the pairing objects corresponding to multiple data frames respectively, and confirm two objects as the same object if the number of times that the two objects form the pairing objects in the multiple data frames is greater than or equal to a number threshold.

[0072] Each data frame can correspond to a sampling moment of the radar.

[0073] That is, the terminal device can perform steps S101 to S103 on the data of each data frame to obtain the pairing objects corresponding to each data frame respectively. At this time, according to the pairing objects corresponding to multiple data frames respectively, if the number of times that two objects form the pairing objects is greater, it means that the two objects are more likely to be the same object. Based on this, the terminal device can set the number threshold Thre2 according to the actual situation, and then confirm two objects as the same object if the number of times that the two objects form the pairing objects in the multiple data frames is greater than or equal to the number threshold Thre2, thereby avoiding the influence of the environmental error occurring in a data frame on the accuracy of the object matching.

[0074] It should be noted that the terminal device can perform the steps of obtaining the pairing objects corresponding to a plurality of data frames respectively and determining two objects as the same object if the number of times of forming the pairing objects in the plurality of data frames is greater than or equal to the number threshold, by using the pairing objects corresponding to the N data frames each time after obtaining the pairing objects corresponding to the N consecutive data frames; or perform the steps of obtaining the pairing objects corresponding to a plurality of data frames respectively and determining two objects as the same object if the number of times of forming the pairing objects in the plurality of data frames is greater than or equal to the number threshold, by using the pairing objects corresponding to the data frame and the pairing objects corresponding to the M data frames before the data frame when the single object and the multiple objects pairing in the data frame is determined.

[0075] Considering that the object matching is performed by using the feature vectors, the feature vectors between the two objects need to be generated by using the feature data between the two objects, and compared with directly performing the object matching by using the feature data of the single object, the operation amount required is increased, and since the object matching is performed by using the feature data of the single object, the mis-matching problem is less likely to occur in the case that the objects in the target region are sparse, therefore, the terminal device can determine the total number of the objects in the target region, and when the total number is greater than or equal to the number threshold, the accurate matching manner is adopted, and when the total number is less than the number threshold, the simple matching manner is adopted.

[0076] The accurate matching manner is that the steps S101 to S103 are performed. The simple matching manner is that the object matching is performed by using the feature data of the single object. According to the total number of the objects in the target region, different manners are adopted to perform the object matching, so that the accuracy and the rate of the object matching can be balanced.

[0077] It should be noted that the number threshold can be set according to the actual situation, and when the target region refers to the overlapping region, the number threshold can be set to 5.

[0078] Please refer to Figure 5 In some embodiments of the present application, the simple matching manner described above can specifically include the following steps S501 to S503.

[0079] In step S501, if the total number is less than the number threshold, each feature data detected by the upstream radar and the downstream radar is obtained.

[0080] Each feature data can be associated with an object located in the target region.

[0081] In some embodiments of the present application, the feature data described above can specifically include at least one of the absolute position, the speed, and the relative lane position of the object.

[0082] Specifically, the relative lane position can be calculated based on the relative position between the object and the reference lane line, the lane identification number of the reference lane line, and the width of each lane in the road. The reference lane line can refer to any lane line in the road where the object is located, for example, the lane line closest to the object.

[0083] In some embodiments of the present application, the relative lane position L can be calculated by the following formula: wherein d is the relative distance between the object and the reference lane line, W is the width of each lane, and I is the lane identification number of the reference lane line.

[0084] Please refer to Figure 6 When the vehicle is driving in lane 3 (between the two lane lines with lane identification numbers 2 and 3), assuming the width of each lane is 2m, if the lane with lane identification number 2 is taken as the reference lane line, then according to the relative distance d = 1m between the vehicle and the reference lane line 2, the relative lane position L can be calculated as Similarly, if the lane with lane identification number 3 is taken as the reference lane line, then according to the relative distance d = 1m between the vehicle and the reference lane line 3, the relative lane position L can be calculated as

[0085] It should be understood that in the prior art, the absolute lane (such as the lane identification number) is often directly used as feature data for matching of the object. However, the lane line is not an obstacle, and the object may frequently change lanes during driving. For example, when the vehicle is driving near the lane line with lane identification number 3, the upstream radar may detect that the lane where the vehicle is located has lane identification number 3, i.e., the vehicle is driving in lane 3, and the downstream radar may detect that the lane where the vehicle is located has lane identification number 4, i.e., the vehicle is driving in lane 4. At this time, if the lane identification number is used for matching, the problem of matching failure will occur. Therefore, the present application uses the relative lane position instead of the absolute lane, which can avoid the influence of frequent lane changing of the object when the object is driving near the lane line on the matching of the object.

[0086] Step S502, each feature data associated with each object detected by the downstream radar is matched with the corresponding feature data associated with each object detected by the upstream radar, respectively.

[0087] Step S503, the objects whose feature data are matched with each other are determined as the paired objects.

[0088] In some embodiments of the present application, the terminal device can sequentially take each object detected by the downstream radar as a current object. When updating the current object each time, the terminal device calculates the difference between each feature data of the current object and the corresponding feature data respectively associated with each object detected by the upstream radar, and the object whose difference between each corresponding feature data associated and each feature data associated with the current object is less than the difference threshold is confirmed as the paired object of the current object.

[0089] Specifically, the terminal device can traverse the absolute position of each object of the upstream radar, calculate the position difference between the absolute position of the current object, and when the position difference is less than the threshold Thre3, set the position matching success flag bit to 1, that is, confirm that the two objects whose position difference is less than the threshold Thre3 match each other in the absolute position. The terminal device can traverse the speed of each object of the upstream radar, calculate the speed difference between the speed of the current object, and when the speed difference is less than the threshold Thre4, set the speed matching success flag bit to 1, that is, confirm that the two objects whose speed difference is less than the threshold Thre4 match each other in the speed. The terminal device can traverse the relative lane position of each object of the upstream radar, calculate the relative lane difference between the relative lane position of the current object, and when the relative lane difference is less than the threshold Thre5, set the relative lane matching success flag bit to 1, that is, confirm that the two objects whose relative lane difference is less than the threshold Thre5 match each other in the relative lane position.

[0090] When the absolute position, speed, and relative lane position of a certain object and the current object are matched, it can be confirmed that the object and the current object match each other. By analogy, after traversing all the objects detected by the downstream radar, the objects detected by the upstream radar and the objects detected by the downstream radar can be formed into paired objects.

[0091] Similarly, the terminal device can obtain the paired objects corresponding to a plurality of data frames respectively, and confirm the two objects whose number of times of being formed into paired objects in a plurality of data frames is greater than or equal to the number threshold as the same object, thereby avoiding the problem of decreased object matching accuracy caused by environmental errors in a certain data frame.

[0092] In some embodiments of the present application, after determining the paired objects, the terminal device can also assign a uniform identification number to the paired objects. Since each radar assigns an identification number to each object detected by itself, when a certain object passes through two adjacent radars, the terminal device can confirm that one object detected by the upstream radar and another object detected by the downstream radar are essentially the same object through matching of the passing object. At this time, the identification number assigned to the object by the upstream radar can be continuously assigned to the same object detected by the downstream radar, thereby making the identification number of the same object uniform when passing through the detection area of the radar group, and realizing the maintenance of the trajectory.

[0093] In addition, when the terminal device is a downstream radar, the terminal device can further send the result of maintaining the trajectories to an upper computer, and the upper computer displays the trajectories of the objects, facilitating the staff to check. In addition, the upper computer can use a trajectory analysis algorithm to analyze the trajectories of the objects, so as to alarm the abnormal trajectories.

[0094] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, certain steps can be performed in other sequences.

[0095] As shown in Figure 7 Fig. 7 is a structural schematic diagram of an object matching device 700 provided by an embodiment of the present application, which is configured on a terminal device.

[0096] Specifically, the object matching device 700 can include:

[0097] A data acquisition unit 701, configured to acquire feature vectors generated by an upstream radar and a downstream radar respectively, the feature vectors being used to represent the relative relationship between two objects in a target area;

[0098] A data matching unit 702, configured to match each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively;

[0099] An object matching unit 703, configured to determine the start objects respectively corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as paired objects, and determine the end objects respectively corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as paired objects.

[0100] In some embodiments of the present application, the object matching unit 703 can be specifically configured to: acquire paired objects respectively corresponding to a plurality of data frames; and determine two objects as the same object if the number of times of forming paired objects in the plurality of data frames is greater than or equal to a number threshold.

[0101] In some embodiments of the present application, the data matching unit 702 can be specifically configured to: calculate the difference degree between each feature vector generated by the downstream radar and each feature vector generated by the upstream radar; and determine two feature vectors as mutually matched feature vectors if the difference degree is less than a degree threshold.

[0102] In some embodiments of the present application, the data acquisition unit 701 can be specifically configured to: determine the total number of objects in the target area; and acquire the feature vectors generated by the upstream radar and the downstream radar respectively if the total number is greater than or equal to a number threshold.

[0103] In some embodiments of the present application, the object matching unit 703 can be specifically configured to: acquire each feature data associated with each object in the target area if the total number is less than the number threshold; match each feature data associated with each object detected by the downstream radar with corresponding feature data associated with each object detected by the upstream radar respectively; and determine the objects with matched feature data as paired objects.

[0104] In some embodiments of the present application, the object matching unit 703 can be specifically configured to: sequentially take each object detected by the downstream radar as a current object; calculate the difference between each feature data associated with the current object and corresponding feature data associated with each object detected by the upstream radar respectively each time the current object is updated; and determine the object with corresponding feature data associated with each feature data associated with the current object and having a difference less than a difference threshold as the paired object of the current object each time the current object is updated.

[0105] In some embodiments of the present application, the feature data includes the relative lane position of each object in the target area, which is calculated based on the relative position between the object and a reference lane line, the lane identification number of the reference lane line, and the width of each lane in the road.

[0106] It should be noted that, for the convenience and brevity of description, the specific working process of the object matching apparatus 700 can be referred to Figures 1 to 6 the corresponding process of the method, which will not be repeated here.

[0107] As Figure 8 shown, a schematic diagram of a terminal device provided by an embodiment of the present application is shown. The terminal device 8 can include a processor 80, a memory 81, and a computer program 82 stored in the memory 81 and executable on the processor 80, such as an object matching program. The processor 80 implements the steps in each of the object matching method embodiments described above when executing the computer program 82, such as steps S101-S103 shown in Figure 1 Alternatively, the processor 80 implements the functions of each module / unit in each of the apparatus embodiments described above when executing the computer program 82, such as the functions of the object matching unit 703 shown in Figure 7The illustrated data acquisition unit 701, data matching unit 702 and object matching unit 703.

[0108] The computer program can be divided into one or more modules / units stored in the memory 81 and executed by the processor 80 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the terminal device.

[0109] For example, the computer program can be divided into a data acquisition unit, a data matching unit and an object matching unit, and the specific functions of each unit are as follows: the data acquisition unit is configured to acquire feature vectors generated by an upstream radar and a downstream radar respectively, the feature vectors being used to represent the relative relationship between objects located in a target area; the data matching unit is configured to match each feature vector generated by the downstream radar with each feature vector generated by the upstream radar respectively; and the object matching unit is configured to determine the start objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar respectively as paired objects, and determine the end objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar respectively as paired objects.

[0110] The terminal device can include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art can understand that, Figure 8 The terminal device is only an example and does not constitute a limitation on the terminal device, and can include more or fewer components than those shown, or combine certain components, or different components, for example, the terminal device can also include an input / output device, a network access device, a bus, etc.

[0111] The processor 80 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0112] The memory 81 can be an internal storage unit of the terminal device, for example, a hard disk or a memory of the terminal device. The memory 81 can also be an external storage device of the terminal device, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device. Further, the memory 81 can also include both the internal storage unit and the external storage device of the terminal device. The memory 81 is used to store the computer program and other programs and data required by the terminal device. The memory 81 can also be used to temporarily store data that has been output or is to be output.

[0113] It should be noted that, for the convenience and brevity of description, the structure of the terminal device can also refer to the specific description of the structure in the method embodiments, which will not be repeated here.

[0114] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific name of each functional unit and module is only for easy distinction, and does not limit the protection scope of the present application. The specific working process of the unit and module in the system can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0115] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the related description of other embodiments.

[0116] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0117] In the embodiments of the present application, it should be understood that the disclosed apparatus / terminal device and method can be implemented in other manners. For example, the described apparatus / terminal device embodiments are merely schematic. For example, the division of the modules or units is merely logical function division. There can be another division manner for actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units.

[0118] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0119] In addition, each functional unit in the various embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0120] The integrated module / unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, all or part of the flow of the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer readable medium can include appropriate contents according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signal and telecommunication signal.

[0121] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those ordinarily skilled in the art should understand: the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for matching objects, characterized in that, include: The feature vectors generated by the upstream and downstream radars are obtained respectively, and the feature vectors are used to characterize the relative relationships between objects located within the target area; Each feature vector generated by the downstream radar is matched with each feature vector generated by the upstream radar. The starting objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that match each other are determined as pairing objects, and the ending objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that match each other are determined as pairing objects. The matching method further includes: sequentially taking each object detected by the downstream radar as the current object; and calculating the difference between each feature data of the current object and the corresponding feature data associated with each object detected by the upstream radar each time the current object is updated, wherein the feature data includes the relative lane position of each object in the target area, and the relative lane position is calculated based on the relative position between the object and the reference lane line, the lane identification number of the reference lane line, and the width of each lane in the road; Each time the current object is updated, objects whose differences between each associated corresponding feature data and each feature data associated with the current object are less than a difference threshold are identified as the paired objects of the current object.

2. The object matching method as described in claim 1, characterized in that, After determining the starting objects corresponding to the mutually matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as paired objects, and determining the ending objects corresponding to the mutually matched feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar as paired objects, the matching method further includes: Retrieve the paired objects corresponding to multiple data frames; Two objects that form a pair in the plurality of data frames more than or equal to a number threshold are identified as the same object.

3. The object matching method as described in claim 1 or 2, characterized in that, The step of matching each feature vector generated by the downstream radar with each feature vector generated by the upstream radar includes: Calculate the degree of difference between each feature vector generated by the downstream radar and each feature vector generated by the upstream radar; Two feature vectors whose difference is less than a threshold are considered as mutually matched feature vectors.

4. The object matching method as described in claim 1 or 2, characterized in that, The acquisition of feature vectors generated by the upstream and downstream radars respectively includes: Determine the total number of objects within the target area; If the total number is greater than or equal to the number threshold, then the feature vectors generated by the upstream radar and the downstream radar are obtained respectively.

5. The object matching method as described in claim 4, characterized in that, After determining the total number of objects within the target area, the matching method further includes: If the total number is less than the number threshold, then each feature data detected by the upstream radar and the downstream radar is acquired, and each feature data is associated with an object located within the target area; Each feature data associated with each object detected by the downstream radar is matched with the corresponding feature data associated with each object detected by the upstream radar. The objects that match each feature data item are identified as paired objects.

6. An object matching device, characterized in that, include: The data acquisition unit is used to acquire feature vectors generated by the upstream radar and the downstream radar respectively. The feature vectors are used to characterize the relative relationships between objects located within the target area. A data matching unit is used to match each feature vector generated by the downstream radar with each feature vector generated by the upstream radar. The object matching unit is used to determine the starting objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that match each other as paired objects, and to determine the ending objects corresponding to the feature vectors generated by the downstream radar and the feature vectors generated by the upstream radar that match each other as paired objects. The object matching device is further configured to: sequentially take each object detected by the downstream radar as the current object; and each time the current object is updated, calculate the difference between each feature data of the current object and the corresponding feature data associated with each object detected by the upstream radar, wherein the feature data includes the relative lane position of each object within the target area, and the relative lane position is calculated based on the relative position between the object and the reference lane line, the lane identification number of the reference lane line, and the width of each lane in the road; Each time the current object is updated, objects whose differences between each associated corresponding feature data and each feature data associated with the current object are less than a difference threshold are identified as the paired objects of the current object.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the matching method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the matching method as described in any one of claims 1 to 5.

Images