A method and device for solving a vehicle trip trajectory matching problem based on multi-source perception data

Abstract

The present application relates to the technical field of intelligent driving, in particular to a method and device for solving vehicle travel trajectory matching problem based on multi-source perception data, wherein the method for solving vehicle travel trajectory matching problem based on multi-source perception data comprises: obtaining first data and second data at a current time, wherein the second data comprises at least one second data; calculating the distance between the first data and each second data according to the first data and at least one second data to form a plurality of feature data groups matched with the first data; obtaining a matching data in each feature data group to form a matching data set; and forming the matching data according to the matching data set.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, specifically to a method and device for solving the problem of vehicle trajectory matching based on multi-source perception data. Background Technology

[0002] In the field of intelligent driving technology, single-vehicle intelligent driving is a necessary condition for the development of intelligent driving. Single-vehicle intelligent driving involves digitizing the surrounding environment through various sensors on the vehicle, facilitating further AI decision-making and processing, and ultimately achieving autonomous or intelligent driving for the vehicle. However, single-vehicle intelligent driving currently has some technical shortcomings. For example, onboard sensors cannot achieve complete environmental perception without blind spots. Traffic itself is a complex systems engineering problem, requiring solutions to comprehensive issues such as vehicle-to-vehicle interaction and vehicle-to-road collaboration. Besides the vehicle's own perception capabilities, a "God's hand" is needed to optimize the allocation of limited road resources to a continuous stream of vehicles; that is, not only do vehicles need to be "intelligent," but roads also need to be "intelligent." Achieving intelligent roads mainly relies on roadside perception equipment. Roadside perception equipment refers to a perception system composed of radar, cameras, and computing units, which can perceive the status of moving targets (vehicles, pedestrians, non-motorized vehicles, etc.) on the road. Radar identification generates radar-side data, mainly including vehicle distance and speed, while camera identification generates camera-side data, with image data mainly including vehicle size, vehicle type, license plate number, and other information. Then, the radar data is synchronized with the camera data in time. Next, spatial location information is transformed from the radar coordinate system to the pixel coordinate system, and then data fusion is performed in the same coordinate system to obtain comprehensive fused perception data, which is then output. However, a challenge exists in the perception data fusion process. After time synchronization and coordinate system transformation, the camera will identify the travel trajectories of multiple vehicles within its field of view, and the radar will also identify the travel trajectories of multiple vehicles within its field of view at the same time. The difficulty lies in how to correctly match the travel trajectories of multiple vehicles on the radar side with those on the camera side.

[0003] Current solutions to this problem can be categorized into local and global approaches. Local approaches typically employ feature recognition and matching, comparing the similarity of features between two points to determine if they belong to the same trajectory. However, this method suffers from a high false positive rate when dealing with situations where trajectory features are indistinct for uniformly moving straight-line vehicles or when the trajectories of multiple vehicles following each other are highly similar. Global approaches generally compare the entire trajectory data set point by point, using a subtraction and averaging method to determine which points represent the same journey trajectory. Differences in the subtraction methods used for data comparison can lead to varying success rates. Summary of the Invention

[0004] To address the aforementioned shortcomings, this application further provides a method and apparatus for solving the vehicle trajectory matching problem based on multi-source sensing data, specifically:

[0005] On the one hand, this application provides a method for solving the vehicle trajectory matching problem based on multi-source sensing data, which includes:

[0006] At the current moment, the first type of data and the second type of data are obtained, and the second type of data contains at least one piece of second data;

[0007] Based on the first data and at least one second data, calculate the distance between the first data and each of the second data to form multiple feature data groups that match the first data;

[0008] In each of the feature data groups, a matching data is obtained to form a matching dataset;

[0009] The matching data is generated based on the matching dataset.

[0010] Preferably, in the above-mentioned method for solving the vehicle trajectory matching problem based on multi-source perception data, a first type of data and a second type of data are acquired at the current moment, wherein the second type of data includes at least one second type of data; wherein the first type of data is data collected by a camera, and the second type of data is data collected by radar.

[0011] Preferably, the above-described method for solving the vehicle trajectory matching problem based on multi-source sensing data, wherein calculating the distance between the first data and each of the second data based on the first data and at least one second data to form multiple feature data groups matching the first data specifically includes:

[0012] Using a predetermined number of first data points as reference points, calculate the distance between each first data point and each second data point to form the distance between the current first data point and each second data point;

[0013] A distance set for each second data point is formed based on the distance between the current first data point and each second data point;

[0014] The feature data group is formed based on the distance set of each second data point.

[0015] Preferably, in the above-described method for solving the vehicle trajectory matching problem based on multi-source sensing data, obtaining matching data from each of the feature data groups to form a matching dataset specifically includes:

[0016] The distance between each first data point and the feature data group is obtained to form each first dataset;

[0017] In each first dataset, the minimum distance value is obtained, and the second data that matches the minimum distance value is used as the initial matching data.

[0018] The matching dataset is formed based on each of the initial matching data.

[0019] Preferably, in the above-described method for solving the vehicle trajectory matching problem based on multi-source sensing data, forming the matching data according to the matching dataset specifically involves: obtaining the proportion of each second data in the matching dataset, and using the second data with the largest proportion as the matching data to match the first data.

[0020] On the other hand, this application further provides a device for solving the vehicle trajectory matching problem based on multi-source sensing data, wherein:

[0021] The acquisition unit acquires first data and second type of data at the current moment, wherein the second type of data contains at least one piece of second data.

[0022] The feature data forming unit calculates the distance between the first data and each of the second data based on the first data and at least one second data to form a plurality of feature data groups that match the first data;

[0023] A matching dataset forming unit acquires matching data from each of the feature data groups to form a matching dataset;

[0024] A matching data forming unit forms the matching data based on the matching dataset.

[0025] Preferably, in the above-mentioned device for solving the vehicle trajectory matching problem based on multi-source sensing data, the feature data forming unit includes:

[0026] The calculation module uses a predetermined number of first data points as reference points to calculate the distance between each first data point and each second data point, thus forming the distance between the current first data point and each second data point.

[0027] The module combines the distance set of each second data point to form a distance set for each second data point based on the distance between the current first data point and each second data point; and forms the feature data group based on the distance set of each second data point.

[0028] Preferably, in the above-mentioned device for solving the vehicle trajectory matching problem based on multi-source sensing data, the matching dataset forming unit includes:

[0029] The first dataset module obtains the distance between each first data point and the feature data group to form each first dataset;

[0030] The initial matching data module obtains the minimum distance value from each first dataset and uses the second data that matches the minimum distance value as the initial matching data.

[0031] The matching dataset module forms the matching dataset based on each of the initial matching data.

[0032] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described above for solving the vehicle trajectory matching problem based on multi-source perception data.

[0033] Fourthly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described above for solving the vehicle trajectory matching problem based on multi-source sensing data.

[0034] Compared with the prior art, the beneficial effects of this application are:

[0035] A method for solving the vehicle trajectory matching problem based on multi-source sensing data adopts a holistic approach. By using the concept of Hausdorff distance calculation, radar data and camera data are compared as a whole, but local differences between radar data and camera data are also compared. Data matching is performed based on the comparison results to determine the correlation between data, which improves the success rate of trajectory matching to a certain extent. Attached Figure Description

[0036] Figure 1 This application provides a flowchart of a method for solving the vehicle trajectory matching problem based on multi-source sensing data;

[0037] Figure 2 This application provides a flowchart of a method for solving the vehicle trajectory matching problem based on multi-source sensing data;

[0038] Figure 3 This application provides a flowchart of a method for solving the vehicle trajectory matching problem based on multi-source sensing data;

[0039] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0040] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0041] Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. The process can be terminated when its operation is complete, but may also have additional steps not included in the figures. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0042] Example 1

[0043] like Figure 1 As shown, this application provides a method for solving the vehicle trajectory matching problem based on multi-source sensing data, which includes:

[0044] Step S110: At the current moment, acquire first data and second type data, where the second type data contains at least one second data point; wherein, the first data is data acquired by the camera, and the second type data is data acquired by the radar. For example, first determine the trajectory data of a vehicle locked by the camera, and then compare it with the trajectory data of multiple targets identified by the radar within the same time period. Assume that the radar identifies a dataset of three targets within this time period, and the first data of a vehicle trajectory acquired by the camera is X1 = {x1, ..., x...} p The radar acquires second type of data Y1 = {A, B, C}, which is the same type of data collected at the same time as the first data. Here, A is the trajectory data identified by the radar, defined as A = {a1, ..., a...}. q}, B represents the trajectory data identified by the radar, which is B = {b1, ..., b}. m}, where C is the trajectory data identified by radar, and C = {c1, ..., c n}

[0045] Step S120: Calculate the distance between the first data and each of the second data based on the first data and at least one of the second data to form multiple feature data groups matching the first data; specifically including:

[0046] like Figure 2As shown, step S1201 involves calculating the distance between each first data point and each second data point using a predetermined number of first data points as reference points, thus forming the distance between the current first data point and each second data point. Indicatively, the predetermined number is 100, but can also be 50 or other arbitrary data set according to actual needs. Indicatively, the distances are calculated based on the bidirectional distances between point X and point A, point X and point B, and point X and point C for every 100 data points.

[0047]

[0048] Where h(X, A) is the one-way Hausdorff distance from set X to set A;

[0049] h(A, X) is the one-way Hausdorff distance from set A to set X;

[0050] Bidirectional Hausdorff distance between X and A: H i (X,A)=max(h i (X,A),h i (A,X))

[0051] Similarly, the bidirectional Hausdorff distance between X and B, and the bidirectional Hausdorff distance between X and C can be obtained using the above method.

[0052] When calculating h(X, A), it is necessary to consider each point x in the point set X. i to a distance x from this point i The nearest midpoint a in set A j Distance between ||x i -a j The distances are sorted, and the maximum value among these distances is taken as the value of h(X,A). The bidirectional Hausdorff distance H1(X,A) is the larger of the one-way distances h1(X,A) and h1(A,X), and it measures the maximum mismatch between the two point sets.

[0053] Step S1202: Form a distance set for each second data point based on the distance between the current first data point and each second data point; schematically:

[0054] For example, the distance set of point A is: H1(X,A), H2(X,A), H3(X,A), H4(X,A)..., the distance set of point B is: H1(X,B), H2(X,B), H3(X,B), H4(X,B)..., and the distance set of point C is: H1(X,C), H2(X,C), H3(X,C), H4(X,C)...

[0055] Step S1203: Form the feature data group based on the distance set of each second data point. Indicatively, the feature data group is as follows:

[0056] H1(X,A), H2(X,A), H3(X,A), H4(X,A)……

[0057] H1(X,B), H2(X,B), H3(X,B), H4(X,B)……

[0058] H1(X,C), H2(X,C), H3(X,C), H4(X,C)……

[0059] Step S130: Obtain matching data from each of the feature data groups to form a matching dataset; specifically including:

[0060] like Figure 3 As shown, step S1301 involves obtaining the distance between the feature data group and each first data point to form each first dataset; illustratively, for example, the first first dataset is H1(X,A), H1(X,B), H1(X,C); the second first dataset is H2(X,A), H2(X,B), H2(X,C); the third first dataset is H3(X,A), H3(X,B), H3(X,C)... the fifth first dataset is H... 100 (X,A), H 100 (X,B),H 100 (X,C).

[0061] Step S1302: Obtain the minimum distance value in each first dataset, and use the second data that matches the minimum distance value as the initial matching data; illustratively:

[0062] Schematic, U1 = min{H1(X,A), H1(X,B), H1(X,C)}; if H1(X,B) is the minimum value in U1, then U1 = B;

[0063] Step S1303: Form the matching dataset based on each of the initial matching data. The matching dataset U = {U1, U2, U3, ... U...} 100}

[0064] Step S140: Form the matching data based on the matching dataset. Calculate the percentage of A, B, and C in set U, and the one with the highest percentage can be inferred to be the actual matching data sought.

[0065] The aforementioned method for solving the vehicle trajectory matching problem based on multi-source sensing data adopts a holistic approach. By using the concept of Hausdorff distance calculation, it compares radar data and camera data as a whole, but also compares the local differences between radar data and camera data. Based on the comparison results, it performs data matching to determine the correlation between data, which enhances the success rate of trajectory matching to a certain extent.

[0066] Example 2

[0067] This application further provides a device for solving the vehicle trajectory matching problem based on multi-source sensing data, comprising:

[0068] The acquisition unit acquires first data and second type of data at the current moment, wherein the second type of data contains at least one piece of second data.

[0069] The feature data forming unit calculates the distance between the first data and each of the second data based on the first data and at least one second data to form a plurality of feature data groups that match the first data;

[0070] A matching dataset forming unit acquires matching data from each of the feature data groups to form a matching dataset;

[0071] A matching data forming unit forms the matching data based on the matching dataset.

[0072] As a further preferred embodiment, in the above-mentioned device for solving the vehicle trajectory matching problem based on multi-source sensing data, the feature data forming unit includes:

[0073] The calculation module uses a predetermined number of first data points as reference points to calculate the distance between each first data point and each second data point, thus forming the distance between the current first data point and each second data point.

[0074] The module combines the distance set of each second data point to form a distance set for each second data point based on the distance between the current first data point and each second data point; and forms the feature data group based on the distance set of each second data point.

[0075] As a further preferred embodiment, the above-mentioned device for solving the vehicle trajectory matching problem based on multi-source sensing data includes a matching dataset forming unit comprising:

[0076] The first dataset module obtains the distance between each first data point and the feature data group to form each first dataset;

[0077] The initial matching data module obtains the minimum distance value from each first dataset and uses the second data that matches the minimum distance value as the initial matching data.

[0078] The matching dataset module forms the matching dataset based on each of the initial matching data.

[0079] The working principle of the device for solving the vehicle trajectory matching problem based on multi-source sensing data and the method for solving the vehicle trajectory matching problem based on multi-source sensing data are the same, and will not be elaborated here.

[0080] Example 3

[0081] This application also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the following:

[0082] At the current moment, the first type of data and the second type of data are obtained, and the second type of data contains at least one piece of second data;

[0083] Based on the first data and at least one second data, calculate the distance between the first data and each of the second data to form multiple feature data groups that match the first data;

[0084] In each of the feature data groups, a matching data is obtained to form a matching dataset;

[0085] The matching data is generated based on the matching dataset.

[0086] Storage medium – any type of memory device or storage device. The term “storage medium” is intended to include: mounting media, such as CD-ROM, floppy disk, or magnetic tape devices; computer system memory or random access memory, such as DRAM, DDRRAM, SRAM, EDORAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. Storage medium may also include other types of memory or combinations thereof. Additionally, storage medium may reside in a computer system in which a program is executed, or it may reside in a different second computer system connected to the computer system via a network (such as the Internet). The second computer system can provide program instructions to the computer for execution. The term “storage medium” can include two or more storage media that may reside in different locations (e.g., in different computer systems connected via a network). Storage medium may store program instructions (e.g., specifically implemented as a computer program) that can be executed by one or more processors.

[0087] Of course, the computer-executable instructions provided in the embodiments of this application are not limited to the synchronous operation of scene recognition as described above, but can also execute related operations in the method for solving the vehicle travel trajectory matching problem based on multi-source perception data provided in any embodiment of this application.

[0088] Example 4

[0089] This application provides an electronic device that can integrate the scene recognition device provided in this application. Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of this application. Figure 4 As shown, this embodiment provides an electronic device 400, which includes: one or more processors 420; and a storage device 410 for storing one or more programs, which, when executed by the one or more processors 420, cause the one or more processors 420 to perform:

[0090] At the current moment, the first type of data and the second type of data are obtained, and the second type of data contains at least one piece of second data;

[0091] Based on the first data and at least one second data, calculate the distance between the first data and each of the second data to form multiple feature data groups that match the first data;

[0092] In each of the feature data groups, a matching data is obtained to form a matching dataset;

[0093] The matching data is generated based on the matching dataset.

[0094] like Figure 4 As shown, the electronic device 400 includes a processor 420, a storage device 410, an input device 430, and an output device 440; the number of processors 420 in the electronic device can be one or more. Figure 4 Taking a processor 420 as an example; the processor 420, storage device 410, input device 430, and output device 440 in the electronic device can be connected via a bus or other means. Figure 4 For example, China and Israel are connected via bus 450.

[0095] Storage device 410, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and module units, such as the program instructions corresponding to the method for solving the vehicle travel trajectory matching problem based on multi-source sensing data in the embodiments of this application.

[0096] Storage device 410 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, storage device 410 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, storage device 410 may further include memory remotely located relative to processor 420, which can be connected via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0097] Input device 430 can be used to receive input digital, character, or voice information, and to generate key signal inputs related to user settings and function control of the electronic device. Output device 440 may include devices such as a display screen and a speaker.

[0098] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for solving the vehicle trajectory matching problem based on multi-source sensing data, characterized in that, include: At the current moment, first data and second type of data are acquired, and the second type of data contains at least one second data; wherein, the first data is data collected by the camera and the second type of data is data collected by the radar, the trajectory data of a vehicle locked by the camera is determined, and the difference is compared with the trajectory data of multiple targets identified by the radar in the same time period; Based on the first data and at least one second data, the distance between the first data and each of the second data is calculated to form multiple feature data groups that match the first data, specifically including: Using a predetermined number of first data points as reference points, calculate the distance between each first data point and each second data point to form the distance between the current first data point and each second data point; A distance set for each second data point is formed based on the distance between the current first data point and each second data point; The feature data group is formed based on the distance set of each second data point; In each of the feature data groups, a matching data is obtained to form a matching dataset; The matching data is generated based on the matching dataset.

2. The method for solving the vehicle trajectory matching problem based on multi-source sensing data according to claim 1, characterized in that, Obtaining matching data from each of the aforementioned feature data groups to form a matching dataset specifically includes: The distance between each first data point and the feature data group is obtained to form each first dataset; In each first dataset, the minimum distance value is obtained, and the second data that matches the minimum distance value is used as the initial matching data. The matching dataset is formed based on each of the initial matching data.

3. The method for solving the vehicle trajectory matching problem based on multi-source sensing data according to claim 2, characterized in that, The matching data is formed based on the matching dataset by: obtaining the proportion of each second data in the matching dataset, and using the second data with the largest proportion as the matching data to match the first data.

4. A device for solving the vehicle trajectory matching problem based on multi-source sensing data, characterized in that, include: The acquisition unit acquires first data and second type of data at the current moment, wherein the second type of data contains at least one second data; wherein the first data is data acquired by the camera and the second type of data is data acquired by the radar, and determines the trajectory data of a vehicle locked by the camera, and performs a difference comparison with the trajectory data of multiple targets identified by the radar within the same time period; The feature data forming unit calculates the distance between the first data and each of the second data based on the first data and at least one second data to form a plurality of feature data groups that match the first data; The feature data forming unit includes: The calculation module uses a predetermined number of first data points as reference points to calculate the distance between each first data point and each second data point, thus forming the distance between the current first data point and each second data point. The module combines the distance set of each second data point to the distance between the current first data point and each second data point; and forms the feature data group based on the distance set of each second data point. A matching dataset forming unit acquires matching data from each of the feature data groups to form a matching dataset; A matching data forming unit forms the matching data based on the matching dataset.

5. The device for solving the vehicle trajectory matching problem based on multi-source sensing data according to claim 4, characterized in that, The matching dataset forming unit includes: The first dataset module obtains the distance between each first data point and the feature data group to form each first dataset; The initial matching data module obtains the minimum distance value from each first dataset and uses the second data that matches the minimum distance value as the initial matching data. The matching dataset module forms the matching dataset based on each of the initial matching data.

6. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the method for solving the vehicle trajectory matching problem based on multi-source perception data as described in any one of claims 1-3 above.

7. An electronic 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 method for solving the vehicle trajectory matching problem based on multi-source perception data as described in any one of claims 1-3.

Images