Target vehicle identification method, field-end server, and readable storage medium

Abstract

The present application relates to the field of automated driving, specifically to a target vehicle identification method, a field-end server, and a readable storage medium, and aims to solve the issue of accurate identification of a target vehicle by a field end. To this end, according to the present application, a field end senses a sensed tracking target to obtain a plurality of target images, on the basis of the target images, acquires target status data of the sensed tracking target, and, on the basis of the target status data, identifies the sensed tracking target, so as to determine whether the sensed tracking target is a target vehicle. By means of the foregoing configuration mode, in the present application, the target status data of the sensed tracking target can be identified by means of the field end, so as to implement accurate identification of the target vehicle, having advantages such as high robustness and a wide scope of use.

Description

Target vehicle identification method, field server and readable storage medium This application claims priority to Chinese patent application CN 202410327248.2, filed on March 21, 2024, entitled “Target vehicle identification method, field server and readable storage medium”. The entire contents of the above Chinese patent application are incorporated into this application by reference. Technical Field

[0001] The present application relates to the field of autonomous driving, and specifically provides a target vehicle identification method, a field server, and a readable storage medium. Background Art

[0002] Unmanned battery swap parking is a very important application scenario in autonomous driving. Accurately identifying the target vehicle is a key part of unmanned battery swap parking.

[0003] Accordingly, the art needs a new target vehicle identification solution to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects, the present application is proposed to provide a solution or at least partially solve the problem of how to achieve accurate identification of the target vehicle at the terminal.

[0005] In a first aspect, the present application provides a method for identifying a target vehicle, characterized in that the method is applied to a field end and comprises:

[0006] Acquire multiple frames of target images of the sensing and tracking target obtained by the field-side sensing;

[0007] Acquiring target state data of the sensed and tracked target based on the multiple frames of target images;

[0008] The sensing and tracking target is identified according to the target state data to determine whether the sensing and tracking target is a target vehicle.

[0009] In one technical solution of the above-mentioned target vehicle identification method, the target state data includes a light detection state;

[0010] The identifying the sensing and tracking target according to the target state data includes:

[0011] Obtaining a light classification result of the perceived tracking target based on the light detection state of the perceived tracking target for a first preset number of consecutive frames;

[0012] The perception tracking target is identified according to the light classification result.

[0013] In one technical solution of the target vehicle identification method, obtaining a light classification result of the perceived and tracked target based on the light detection state of the perceived and tracked target for a first preset number of consecutive frames includes:

[0014] When the light detection state for a second consecutive preset number of frames in the light detection state for the first consecutive preset number of frames is all in the on state, determining that the light classification result of the sensed tracking target is light on;

[0015] Wherein, the first preset frame number is greater than or equal to the second preset frame number.

[0016] In one technical solution of the above-mentioned target vehicle identification method, the field end is provided with a laser radar and multiple cameras, and the multiple cameras have different viewing angles;

[0017] The acquiring of multiple frames of target images of the sensing and tracking target obtained by the field-side sensing includes:

[0018] Obtaining a three-dimensional detection frame of the perceived tracking target based on the point cloud data collected by the laser radar;

[0019] Acquire multiple frames of continuous images at different viewing angles using the multiple cameras;

[0020] According to the three-dimensional detection frame and the continuous multi-frame images of different viewing angles, continuous multi-frame target images captured by multiple cameras are acquired.

[0021] In one technical solution of the above-mentioned target vehicle identification method, obtaining target state data of the sensed and tracked target based on the multiple frames of target images includes:

[0022] According to the continuous multi-frame target images of each camera, the self-attention features of each view are obtained;

[0023] Obtaining cross-attention features based on the target images captured by multiple cameras at the same time;

[0024] According to the self-attention feature and the cross-attention feature, a light detection result of the perceived tracking target is obtained.

[0025] In one technical solution of the above-mentioned target vehicle identification method, the target state data includes a vehicle type detection result;

[0026] The identifying the sensing and tracking target according to the light classification result includes:

[0027] When the light classification result is that the light is on, judging whether the vehicle type of the historical frame of the perception and tracking target is consistent with that of the current frame according to the vehicle type detection result;

[0028] When the vehicle types are consistent, the sensing tracking target is determined to be a target vehicle.

[0029] In one technical solution of the above-mentioned target vehicle identification method, the target state data includes a target motion state;

[0030] The identifying the sensing and tracking target according to the light classification result includes:

[0031] When the light classification result is that the light is on, judging whether the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame according to the target motion state;

[0032] When the motion states are consistent, the sensed tracking target is determined to be a target vehicle.

[0033] In one technical solution of the above-mentioned target vehicle identification method, the target state data includes a vehicle type detection result and a target motion state;

[0034] The identifying the sensing and tracking target according to the light classification result includes:

[0035] When the light classification result is that the light is on, judging whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame based on the vehicle type detection result, and judging whether the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame based on the motion state of the target;

[0036] When the vehicle type and the motion states are consistent, the sensing tracking target is determined to be a target vehicle.

[0037] In one technical solution of the target vehicle identification method, judging whether the vehicle type of the historical frame of the perceived and tracked target is consistent with that of the current frame based on the vehicle type detection result includes:

[0038] When the proportion of the vehicle type detection results of all historical frames of the perception and tracking target that is the same as the vehicle type detection result of the current frame is greater than or equal to a first preset proportion, it is determined that the vehicle type of the historical frames of the perception and tracking target is consistent with that of the current frame.

[0039] In one technical solution of the target vehicle identification method, judging whether the motion state of the historical frame of the perceived and tracked target is consistent with the motion state of the current frame according to the motion state of the target includes:

[0040] Acquire historical trajectory data sent by the sensing and tracking target to the field end;

[0041] According to the target motion state, determining whether the historical frame and the current frame are both in a stationary state or a moving state, and determining whether the field-side historical trajectory of the sensed and tracked target collected by the field end is consistent with the historical trajectory data;

[0042] When both the historical frame and the current frame are in a stationary state or a moving state, and the field-side historical trajectory is consistent with the historical trajectory data, it is determined that the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame.

[0043] In one technical solution of the above-mentioned target vehicle identification method, determining that the sensed tracking target is a target vehicle includes:

[0044] Determining whether the sensing tracking target is unique;

[0045] When the sensing tracking target is unique, the sensing tracking target is determined to be a target vehicle.

[0046] In one technical solution of the above-mentioned target vehicle identification method, before identifying the perception tracking target based on the target state data, the method further includes:

[0047] Acquiring relative positioning data sent by the sensing and tracking target to the field end;

[0048] When the relative positioning data is acquired, the sensing and tracking target is identified according to the distance between the sensing and tracking target and the field end;

[0049] When the relative positioning data is not obtained, the step of identifying the perception tracking target according to the target state data is performed.

[0050] In one technical solution of the above-mentioned target vehicle identification method, identifying the perception and tracking target based on the distance between the perception and tracking target and the field end includes:

[0051] When the distance between the sensing and tracking target and the field end is within a preset distance threshold range, the sensing and tracking target with the smallest distance is identified as the target vehicle.

[0052] In a second aspect, the present application provides a field-side server, the field-side server comprising:

[0053] at least one processor;

[0054] and, a memory communicatively coupled to the at least one processor;

[0055] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the target vehicle identification method described in any one of the technical solutions of the above-mentioned target vehicle identification method is implemented.

[0056] In a third aspect, a computer-readable storage medium is provided, which stores a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the target vehicle identification method described in any one of the technical solutions of the above-mentioned target vehicle identification method.

[0057] The above one or more technical solutions of this application have at least one or more of the following beneficial effects:

[0058] In implementing the technical solution of this application, the present application senses the sensing and tracking target through the field end, obtains multiple frames of target images, and obtains target state data of the sensing and tracking target based on the target images. Based on the target state data, the sensing and tracking target is identified, thereby determining whether the sensing and tracking target is a target vehicle. Through the above configuration, the present application can achieve accurate identification of the target vehicle by identifying the target state data of the sensing and tracking target through the field end, which has the advantages of high robustness and a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The disclosure of this application will become more easily understood with reference to the accompanying drawings. Those skilled in the art will readily appreciate that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Among them:

[0060] FIG1 is a schematic flow chart of the main steps of a method for identifying a target vehicle according to an embodiment of the present application;

[0061] FIG2 is a flow chart showing the main steps of a process for identifying and parking a battery swapping vehicle in a battery swapping station according to an embodiment of the present application;

[0062] FIG3 is a schematic diagram of the main implementation framework of a battery-swap vehicle identification method according to an embodiment of the present application;

[0063] FIG4 is a schematic diagram of a visual interface of a battery swap vehicle identification process according to an embodiment of the present application;

[0064] FIG5 is a flow chart showing the main steps of a method for identifying a target vehicle according to an embodiment of the present invention;

[0065] FIG6 is a flowchart illustrating the main steps of obtaining a light classification result according to an implementation of an embodiment of the present application;

[0066] FIG7 is a schematic diagram of the connection relationship between the controller and the processor of the field-side server according to an embodiment of the present application. DETAILED DESCRIPTION

[0067] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the scope of protection of the present application.

[0068] In the description of this application, "module" and "processor" may include hardware, software, or a combination of both. A module may include hardware circuitry, various suitable sensors, communication ports, and memory. It may also include software components, such as program code, or a combination of software and hardware. A processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. A processor has data and / or signal processing capabilities. A processor may be implemented in software, hardware, or a combination of both. Non-transitory computer-readable storage media include any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" refers to all possible combinations of A and B, such as only A, only B, or both A and B. The terms "at least one of A or B" or "at least one of A and B" have similar meanings to "A and / or B" and may include only A, only B, or both A and B. The singular forms "a" and "the" may also include the plural forms.

[0069] Here we first explain some terms involved in this application.

[0070] An automated driving system (ADS) is a system that continuously performs all dynamic driving tasks (DDT) within its operational domain design (ODD). Specifically, the system is only allowed to fully assume the task of autonomous vehicle control under specified appropriate driving scenarios. When the vehicle meets the ODD conditions, the system is activated, replacing the human driver as the vehicle's primary driver. The DDT refers to the continuous lateral (left and right steering) and longitudinal motion control (acceleration, deceleration, and constant speed) of the vehicle, as well as the detection and response to objects and events in the vehicle's driving environment. The ODD refers to the conditions under which the automated driving system can operate safely. These conditions can include geographic location, road type, speed range, weather, time of day, and national and local traffic laws and regulations.

[0071] Referring to FIG. 1 , FIG. 1 is a flow chart illustrating the main steps of a target vehicle identification method according to an embodiment of the present application. As shown in FIG. 1 , the target vehicle identification method in the embodiment of the present application is applied to a field terminal and mainly includes the following steps S101 to S103 .

[0072] Step S101: Acquire multiple frames of target images of the perceived tracking target obtained by field-side perception.

[0073] In this embodiment, the field end may sense the field end environment, thereby obtaining a multi-frame target image of the sensed and tracked target in the field end environment.

[0074] In one embodiment, the field end can be a battery swap station, a smart parking lot, etc., which can realize the association between the vehicle end and the field end, thereby realizing the scenario of automatic parking or automatic battery swap parking function.

[0075] In one embodiment, the field side may be provided with multiple sensors, such as lidar and cameras, for sensing the field side environment.

[0076] Step S102: Acquire target state data of the perceived tracking target based on multiple frames of target images.

[0077] In this embodiment, target state data of the perceived and tracked target can be obtained based on multiple frames of target images, wherein the target state data is data representing the current state of the perceived and tracked target.

[0078] In one embodiment, the target state data may include light detection state, vehicle type detection result, target motion state, etc. The target motion state may include being in motion or stationary state, and motion trajectory, etc.

[0079] Step S103: Identify the sensing and tracking target according to the target state data to determine whether the sensing and tracking target is a target vehicle.

[0080] In this embodiment, the perception tracking target can be identified based on the target state data, thereby determining whether the perception tracking target is a target vehicle.

[0081] Based on steps S101-S103 described above, the embodiment of the present application senses the target being tracked through the field end, obtains multiple frames of target images, and acquires target state data of the target being tracked based on the target images. Based on the target state data, the target being tracked is identified, thereby determining whether the target being tracked is the target vehicle. Through the above configuration, the embodiment of the present application can accurately identify the target vehicle by identifying the target state data of the target being tracked through the field end, thus having the advantages of high robustness and a wide range of applications.

[0082] Steps S101 to S103 are further described below.

[0083] In one embodiment of the present application, a laser radar and multiple cameras are provided at the field end, and the multiple cameras have different viewing angles; step S101 may further include the following steps S1011 to S1013:

[0084] Step S1011: Obtain a three-dimensional detection frame of the perceived tracking target based on the point cloud data collected by the lidar.

[0085] Step S1012: Acquire multiple frames of continuous images at different viewing angles using multiple cameras.

[0086] Step S1013: acquiring a continuous multi-frame target image captured by multiple cameras according to the three-dimensional detection frame and the continuous multi-frame images of different viewing angles.

[0087] In this embodiment, the field end can be equipped with a lidar and multiple cameras with different viewing angles. This multi-angle camera setup ensures comprehensive coverage of the field end environment, preventing issues such as occlusion that can cause the inability to identify and track target status data. By simultaneously acquiring target images based on the perception of the lidar and cameras, multi-view, multi-sensor integration is achieved, improving the accuracy of target status data.

[0088] Specifically, the system acquires a 3D detection frame of the perceived tracking target based on the point cloud data collected by the LiDAR. Specifically, the LiDAR is used to detect the perceived tracking target in the field environment, thereby obtaining a 3D detection frame of the perceived tracking target. Targets outside the area of ​​interest are filtered out based on the coordinates of the 3D detection frame. Multiple images from different perspectives are then acquired using cameras with different viewpoints, ensuring a more complete picture of the perceived tracking target's appearance and lighting conditions. Based on the 3D detection frame and multiple frames of images from different perspectives, multiple frames of target images of the perceived tracking target are acquired.

[0089] In one embodiment, step S102 may further include the following steps S1021 to S1023:

[0090] Step S1021: Obtain the self-attention features of each perspective based on the continuous multi-frame target images of each camera.

[0091] Step S1022: Obtain cross-attention features based on target images captured by multiple cameras at the same time.

[0092] Step S1023: Obtain the light detection result of the perceived tracking target based on the self-attention feature and the cross-attention feature.

[0093] In this embodiment, a neural network model can be applied to extract features from multiple consecutive target images from each camera, obtaining self-attention features for a continuous time series at each viewpoint. Cross-attention features from multiple cameras at the same moment are also obtained. Through training, the weight distribution of self-attention features and cross-attention features is determined. Based on the self-attention features, cross-attention features, and their corresponding weights, the light detection results of the perceived tracking target are predicted.

[0094] In one embodiment, step S103 may further include the following steps S1031 and S1032:

[0095] Step S1031: obtaining a light classification result of the perceived tracking target according to the light detection state of the perceived tracking target for a first preset number of consecutive frames.

[0096] In this embodiment, step S1031 may be further configured as follows:

[0097] When the light detection state for a second preset number of consecutive frames is on in the light detection state for a first preset number of consecutive frames, the light classification result of the perceived tracking target is determined to be light on; wherein the first preset number of frames is greater than or equal to the second preset number of frames.

[0098] In this embodiment, if the light detection state is on for a second preset number of consecutive frames in the light detection state for a first preset number of consecutive frames, it can be considered that the light of the light classification result of the perceived tracking target is on.

[0099] In an example, if the first preset number of frames is 8 and the second preset number of frames is 5, if in 8 consecutive frames of light detection status, the light detection status in the frame number interval [2, 6] is all on, then the light classification result can be considered as light on.

[0100] Step S1032: Identify the perceived tracking target based on the light classification result.

[0101] In this embodiment, the sensing and tracking target can be identified based on the light classification result. That is, if the light classification result is that the light is on, the sensing and tracking target can be considered to be the target vehicle.

[0102] In another embodiment, the target state data may include a vehicle type detection result, and step S1032 may further include steps S10321 to S10322:

[0103] Step S10321: When the light classification result is that the light is on, based on the vehicle type detection result, determine whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame.

[0104] In this embodiment, the target image can be identified to obtain a vehicle type detection result in the target image. When the light classification result is that the light is on, it can be determined based on the vehicle type detection result whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame.

[0105] In one embodiment, a first preset ratio can be set. When the proportion of vehicle type detection results of all historical frames of the perception and tracking target that is the same as the vehicle type detection result of the current frame is greater than or equal to the first preset ratio, it can be determined that the vehicle type of the historical frames of the perception and tracking target is consistent with that of the current frame.

[0106] In one example, the first preset ratio may be 50%. That is, if the ratio of the vehicle type detection results of the perceived target in all historical frames is the same as that in the current frame is greater than or equal to 50%, it can be determined that the vehicle type of the perceived target in the historical frames is the same as that in the current frame.

[0107] Step S10322: When the vehicle types are consistent, the sensing tracking target is determined to be the target vehicle.

[0108] In this embodiment, if the vehicle type matches, the perceived tracking target is determined to be the target vehicle. Combining the light classification results and the vehicle type, a comprehensive judgment on whether the perceived tracking target is the target vehicle can further improve the robustness of the recognition process.

[0109] In another embodiment, the target state data may include a vehicle type detection result, and step S1032 may further include steps S10323 to S10324:

[0110] Step S10323: When the light classification result is that the light is on, determine whether the motion state of the historical frame of the perceived tracking target is consistent with the motion state of the current frame based on the target motion state.

[0111] In this embodiment, step S10323 may further include the following steps S103231 to S103233:

[0112] Step S103231: Obtain historical trajectory data sent by the sensing and tracking target to the field end.

[0113] Step S103232: According to the target motion state, determine whether the historical frame and the current frame are both in a stationary state or a moving state, and determine whether the field-side historical trajectory of the perception and tracking target collected by the field side is consistent with the historical trajectory data.

[0114] Step S103233: When both the historical frame and the current frame are in a stationary state or a moving state, and the field-side historical trajectory is consistent with the historical trajectory data, it is determined that the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame.

[0115] In this embodiment, the field-side historical trajectory of the field-side perception and tracking target can be obtained based on multiple frames of target images, and the field-side perception and tracking target can be identified as being in a stationary state or in motion. If both the historical frame and the current frame are in a stationary state or in motion, and the field-side historical trajectory is consistent with the historical trajectory data sent by the perception and tracking target to the field-side, it can be considered that the motion state of the perception and tracking target in the historical frame and the current frame is consistent.

[0116] In one embodiment, a trajectory ratio threshold may be set. When the ratio of the historical trajectory of the field end to the historical trajectory data sent by the sensing and tracking target to the field end is greater than the trajectory ratio threshold, the trajectories may be considered consistent.

[0117] In one embodiment, the historical trajectory data of the target being tracked may be obtained by using a wheel speed meter of the target being tracked.

[0118] Step S10324: When the motion states are consistent, the sensed tracking target is determined to be the target vehicle.

[0119] In this embodiment, if the motion state is consistent, the perceived tracking target can be considered to be the target vehicle. Comprehensively considering the light classification results and motion state of the perceived tracking target can further improve the robustness of the recognition process.

[0120] In another embodiment, step S1032 may further include steps S10325 to S10326:

[0121] Step S10325: When the light classification result is that the light is on, based on the vehicle type detection result, determine whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame, and based on the target motion state, determine whether the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame.

[0122] Step S10326: When the vehicle type and motion state are consistent, the sensing tracking target is determined to be the target vehicle.

[0123] In this embodiment, the light classification results, vehicle type detection results and target motion state can be combined to comprehensively judge whether the perceived target is the target vehicle, which can further improve the robustness of the recognition process.

[0124] In one embodiment, after determining that the perceived tracking target is the target vehicle based on light classification results, or light classification results combined with vehicle type detection results, or light classification results combined with target motion status, or light classification results combined with vehicle type detection results and target motion status, the system can further determine whether the perceived tracking target is unique. If the perceived tracking target is unique, the system can ultimately determine that the perceived tracking target is the target vehicle. This can avoid the problem of the terminal being unable to implement automatic parking, battery swap parking, and other functions when there are multiple target vehicles, thereby improving the robustness of terminal control.

[0125] In one implementation of the embodiment of the present application, before step S103, the present application may further include the following steps S104 to S106:

[0126] Step S104: Acquire the relative positioning data sent by the sensing and tracking target to the field end.

[0127] In this embodiment, after the sensing and tracking target establishes a communication connection with the field end, it will send relative positioning data to the field end.

[0128] Step S105: When the relative positioning data is obtained, the sensing and tracking target is identified according to the distance between the sensing and tracking target and the field end.

[0129] Step S106: When the relative positioning data is not obtained, execute step S103.

[0130] In this embodiment, if the terminal acquires relative positioning data sent by a sensing and tracking target, it can determine whether the distance between the sensing and tracking target and the terminal is within a preset distance threshold based on the relative positioning data. If the distance between the sensing and tracking target and the terminal is within the preset distance threshold, the sensing and tracking target with the shortest distance can be identified as the target vehicle. If relative positioning data is not acquired, the sensing and tracking target can be identified based on the target status data.

[0131] The following is a description of the target vehicle identification method of the present application in conjunction with Figures 2 to 5, taking the field end as a battery swap station and the target vehicle as a vehicle to be battery swapped as an example. Figure 2 is a flow chart of the main steps of the identification and parking control process of a battery swap vehicle in a battery swap station according to an embodiment of the present application; Figure 3 is a schematic diagram of the main implementation framework of the battery swap vehicle identification method according to an embodiment of the present application; Figure 4 is a schematic diagram of the visual interface of the battery swap vehicle identification process according to an embodiment of the present application; Figure 5 is a flow chart of the main steps of the target vehicle identification method according to an embodiment of the book application.

[0132] As shown in Figure 2, the battery swap station can perform target detection through sensor data, obtain perception and tracking targets, and identify battery swap vehicles based on the perception and tracking targets. After obtaining the battery swap vehicles, the battery swap vehicles can be optimized and the targets can be tracked, thereby realizing the integration and planning control of the vehicles to be battery swapped, so as to realize battery swap parking.

[0133] As shown in Figure 3, the identification process of the vehicle to be replaced by the battery-swapping vehicle identification module is implemented. The station-side input data of the battery-swapping vehicle identification module are the brightness / darkness (on / off) of the lights and the vehicle model information (vehicle model detection results) collected by the station. The vehicle-side input data of the battery-swapping vehicle identification module are the brightness / darkness (on / off) of the lights, the vehicle model information (vehicle model detection results) and the positioning of the vehicle in the station system (relative positioning data). Among them, the positioning of the vehicle in the station system is optional input data. The output of the battery-swapping vehicle identification module is the target to be parked (the vehicle to be replaced by the battery) or other dynamic targets.

[0134] As shown in Figure 4, cameras with different viewing angles installed at the battery swap station can capture target images of the battery swap vehicle from different viewing angles. The laser radar installed at the battery swap station can obtain a three-dimensional detection frame of the battery swap vehicle, where the three-dimensional detection frame is shown as mark 1 in Figure 4.

[0135] As shown in FIG5 , the target vehicle identification process may include the following steps S201 to S201 to S212:

[0136] Step S201: Update the information of the vehicle to be replaced.

[0137] Step S202: Determine whether the relative positioning data of the vehicle to be replaced is valid; if so, jump to step S203; if not, jump to step S204.

[0138] Step S203: Determine whether there is a station-side sensing tracking target within the distance threshold range; if so, jump to step S205; if not, jump to step S206.

[0139] Step S204: Check whether the lighting status of the vehicle to be replaced is valid; if so, jump to step S207; if not, jump to step S206.

[0140] Step S205: Output the “nearest” perception tracking target as the vehicle to be replaced with a battery.

[0141] Step S206: The vehicle to be replaced does not exist.

[0142] Step S207: Determine whether the classification result of the light of the perceived tracking target is on; if so, jump to step S208; if not, jump to step S206.

[0143] Step S208: Determine whether the target vehicle model being sensed and tracked is consistent; if so, jump to step S209; if not, jump to step S206.

[0144] In this embodiment, step S208 is similar to the method described in the aforementioned step S10321, and for the sake of simplicity, it will not be repeated here.

[0145] Step S209: Determine whether the motion state of the perceived tracking target is consistent with that of the vehicle; if so, jump to step S210; if not, jump to step S206.

[0146] In this embodiment, step S209 is similar to the method described in the aforementioned step S10323, and for the sake of simplicity, it is not repeated here.

[0147] Step S210: Determine whether the perception tracking target is unique; if so, jump to step S211; if not, jump to step S206.

[0148] Step S211: Output the perception tracking target as the vehicle to be replaced with a battery.

[0149] In one embodiment, referring to FIG6 , FIG6 is a flowchart of the main steps of obtaining light classification results according to an embodiment of the present application. As shown in FIG6 , the light classification results can be obtained according to the following steps S301 to S309:

[0150] Step S301: Laser radar detection obtains a candidate frame (three-dimensional detection frame).

[0151] Step S302: Filter candidate objects outside the frame according to coordinates.

[0152] Step S303: Projecting the candidate box to the multi-view camera coordinate system.

[0153] Step S304: capturing candidate target pictures (target images) from the data collected by the multi-view camera.

[0154] Step S305: The neural network extracts target image features.

[0155] Step S306: Self-attention feature extraction.

[0156] Step S307: Cascade with historical features.

[0157] Step S308: Cross-attention features of the current frame.

[0158] Step S309: Obtain light classification results.

[0159] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of the present application, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present application.

[0160] It will be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment of the present application can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium can include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program code. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electric carrier signals and telecommunication signals.

[0161] Another aspect of the present application provides a field-side server, which may include at least one processor and a memory communicatively connected to the at least one processor. The memory stores a computer program that, when executed by the at least one processor, implements the method described in any of the above embodiments. Referring to Figure 7, Figure 7 is a schematic diagram illustrating the connection relationship between a controller and a processor of a field-side server according to one embodiment of the present application. As shown in Figure 7, the controller and processor of the field-side server are communicatively connected via a bus.

[0162] Furthermore, the present application also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present application, the computer-readable storage medium can be configured to store a program for executing the target vehicle identification method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned target vehicle identification method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium can be a memory device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.

[0163] Furthermore, it should be understood that since the configuration of each module is merely for the purpose of illustrating the functional units of the apparatus of the present application, the physical devices corresponding to these modules may be the processor itself, or a portion of the software in the processor, a portion of the hardware, or a combination of software and hardware. Therefore, the number of modules in the figure is merely illustrative.

[0164] Those skilled in the art will appreciate that the various modules in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principles of this application. Therefore, the technical solutions after splitting or merging will fall within the scope of protection of this application.

[0165] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, and based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0166] The user personal information processed by this application will vary depending on the specific product / service scenario and will be based on the specific scenario in which the user uses the product / service. This may involve the user's account information, device information, driving information, vehicle information, or other related information. The applicant will treat the user's personal information and its processing with a high degree of diligence.

[0167] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0168] Thus far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A method for identifying a target vehicle, characterized in that: The method is applied to a field end, and includes: Acquire multiple frames of target images of the sensing and tracking target obtained by the field-side sensing; Acquiring target state data of the sensed and tracked target based on the multiple frames of target images; The sensing and tracking target is identified according to the target state data to determine whether the sensing and tracking target is a target vehicle.

2. The target vehicle identification method according to claim 1, characterized in that: The target state data includes light detection state; The identifying the sensing and tracking target according to the target state data includes: Obtaining a light classification result of the perceived tracking target based on the light detection state of the perceived tracking target for a first preset number of consecutive frames; The perception tracking target is identified according to the light classification result.

3. The target vehicle identification method according to claim 2, characterized in that: The obtaining of a light classification result of the perceived tracking target according to the light detection state of the perceived tracking target for a first preset number of consecutive frames includes: When the light detection state for a second consecutive preset number of frames in the light detection state for the first consecutive preset number of frames is all in the on state, determining that the light classification result of the sensed tracking target is light on; Wherein, the first preset frame number is greater than or equal to the second preset frame number.

4. The target vehicle identification method according to claim 2, characterized in that: The field end is provided with a laser radar and multiple cameras, and the multiple cameras have different viewing angles; The acquiring of multiple frames of target images of the perceived tracking target obtained by the field-side perception includes: Obtaining a three-dimensional detection frame of the perceived tracking target based on the point cloud data collected by the laser radar; Acquire multiple frames of continuous images at different viewing angles using the multiple cameras; According to the three-dimensional detection frame and the continuous multi-frame images of different viewing angles, continuous multi-frame target images captured by multiple cameras are acquired.

5. The target vehicle identification method according to claim 4, characterized in that: The acquiring target state data of the perceived tracking target according to the multiple frames of target images includes: According to the continuous multi-frame target images of each camera, the self-attention features of each view are obtained; Obtaining cross-attention features based on the target images captured by multiple cameras at the same time; According to the self-attention feature and the cross-attention feature, a light detection result of the perceived tracking target is obtained.

6. The target vehicle identification method according to claim 2, characterized in that: The target state data includes vehicle type detection results; The identifying the sensing and tracking target according to the light classification result includes: When the light classification result is that the light is on, judging whether the vehicle type of the historical frame of the perception and tracking target is consistent with that of the current frame according to the vehicle type detection result; When the vehicle types are consistent, the sensing tracking target is determined to be a target vehicle.

7. The target vehicle identification method according to claim 2, characterized in that: The target state data includes target motion state; The identifying the sensing and tracking target according to the light classification result includes: When the light classification result is that the light is on, judging whether the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame according to the target motion state; When the motion states are consistent, the sensed tracking target is determined to be a target vehicle.

8. The target vehicle identification method according to claim 2, characterized in that: The target state data includes vehicle type detection results and target motion state; The identifying the sensing and tracking target according to the light classification result includes: When the light classification result is that the light is on, judging whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame based on the vehicle type detection result, and judging whether the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame based on the motion state of the target; When the vehicle type and the motion states are consistent, the sensing tracking target is determined to be a target vehicle.

9. The target vehicle identification method according to claim 6 or 8, characterized in that: The determining, based on the vehicle type detection result, whether the vehicle type of the historical frame of the perceived tracking target is consistent with that of the current frame includes: When the proportion of the vehicle type detection results of all historical frames of the perception and tracking target that is the same as the vehicle type detection result of the current frame is greater than or equal to a first preset proportion, it is determined that the vehicle type of the historical frames of the perception and tracking target is consistent with that of the current frame.

10. The target vehicle identification method according to claim 7 or 8, characterized in that: The determining, based on the target motion state, whether the motion state of the historical frame of the perceived tracking target is consistent with the motion state of the current frame includes: Acquire historical trajectory data sent by the sensing and tracking target to the field end; According to the target motion state, determining whether the historical frame and the current frame are both in a stationary state or a moving state, and determining whether the field-side historical trajectory of the sensed and tracked target collected by the field end is consistent with the historical trajectory data; When both the historical frame and the current frame are in a stationary state or a moving state, and the field-side historical trajectory is consistent with the historical trajectory data, it is determined that the motion state of the historical frame of the perceived tracking target is consistent with that of the current frame.

11. The target vehicle identification method according to claim 6, 7 or 8, characterized in that: Determining that the sensed tracking target is a target vehicle includes: Determining whether the sensing tracking target is unique; When the sensing tracking target is unique, the sensing tracking target is determined to be a target vehicle.

12. The target vehicle identification method according to claim 1, characterized in that: Before identifying the sensing and tracking target according to the target state data, the method further includes: Acquiring relative positioning data sent by the sensing and tracking target to the field end; When the relative positioning data is acquired, the sensing and tracking target is identified according to the distance between the sensing and tracking target and the field end; When the relative positioning data is not obtained, the step of identifying the perception tracking target according to the target state data is performed.

13. The target vehicle identification method according to claim 12, characterized in that: The identifying the sensing and tracking target according to the distance between the sensing and tracking target and the field end includes: When the distance between the sensing and tracking target and the field end is within a preset distance threshold range, the sensing and tracking target with the smallest distance is identified as the target vehicle.

14. A field-side server, characterized in that: The field-side server includes: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the target vehicle identification method according to any one of claims 1 to 13 is implemented.

15. A computer-readable storage medium storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by a processor to execute the target vehicle identification method according to any one of claims 1 to 13.

Images