Auxiliary parking method, device, computer device and computer-readable storage medium

Abstract

The present application relates to an auxiliary parking method, device, computer device, and computer-readable storage medium. The method includes: obtaining point cloud data scanned by a roadside fusion perception system for a preset parking area, and obtaining position and attitude information of a target vehicle in the preset parking area according to the point cloud data; obtaining collision information corresponding to obstacles within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information, where the collision information is used to characterize whether a collision will occur between the target vehicle and the obstacles during the parking process of the target vehicle; generating an auxiliary parking image based on the position and attitude information and the collision information, and displaying the auxiliary parking image through a preset display component corresponding to the target vehicle. Using this method can improve parking safety and the flexibility of auxiliary parking.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technologies, and particularly to an auxiliary parking method, device, computer device, and computer-readable storage medium. Background Art

[0002] With the development of the economy and the improvement of people's living standards, there are more and more vehicles on the road. With the sharp increase in the number of vehicles, various vehicle parking problems have become increasingly prominent.

[0003] For example, usually, the vehicle owner generally parks the vehicle in a parking space based on personal experience. However, due to problems such as the vehicle owner's blind spot of vision or lack of experience, unsafe phenomena such as vehicle scratches often occur. To avoid this phenomenon, currently, some high-end vehicles are equipped with a reverse image and a reverse radar to assist the vehicle owner in parking.

[0004] However, installing a reverse image and a reverse radar will increase the vehicle cost. Therefore, the above method of auxiliary parking by installing a reverse image and a reverse radar has the problem of poor flexibility. Summary of the Invention

[0005] Based on this, it is necessary to provide an auxiliary parking method, device, computer device, and computer-readable storage medium that can improve the safety of parking and the flexibility of auxiliary parking for the above technical problems.

[0006] In a first aspect, an embodiment of this application provides an auxiliary parking method, and the method includes:

[0007] Obtain point cloud data scanned by a roadside fusion perception system for a preset parking area, and based on the point cloud data, obtain the position and attitude information of a target vehicle in the preset parking area;

[0008] Based on the point cloud data and the position and attitude information, obtain collision information corresponding to obstacles within a preset distance range of the target vehicle, where the collision information is used to represent whether the target vehicle and the obstacles will collide during the parking process of the target vehicle;

[0009] Generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0010] In one of the embodiments, the obtaining the position and attitude information of the target vehicle in the preset parking area based on the point cloud data includes:

[0011] Determine target point cloud data corresponding to the target vehicle from the point cloud data;

[0012] Perform object detection on the target point cloud data to obtain the position and attitude information, which is used to represent the current parking position and parking attitude of the target vehicle in the preset parking area.

[0013] In one embodiment, determining the target point cloud data corresponding to the target vehicle from the point cloud data includes:

[0014] Obtain the vehicle image of the target vehicle collected by the roadside fusion perception system, and perform position calibration on the vehicle image to obtain the calibrated position corresponding to the target vehicle;

[0015] Filter the point cloud data according to the calibrated position to obtain the target point cloud data, extract the identity information of the target vehicle from the vehicle image, and store the identity information and the target point cloud data in an associated manner.

[0016] In one embodiment, the position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle; obtaining the collision information of the obstacles within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information includes:

[0017] Determine the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and the preset distance range threshold;

[0018] Perform object detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle bounding box corresponding to the obstacle;

[0019] Obtain the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0020] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes:

[0021] If it is detected that the distance between the target corner point and the first target side of the obstacle bounding box is less than the first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, generate a first collision warning mark, and use the first collision warning mark as the collision information;

[0022] Wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

[0023] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes:

[0024] If it is detected that the distance between the second target side of the target bounding box and the obstacle corner point is less than the second distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, a second collision warning mark is generated, and the second collision warning mark is used as the collision information;

[0025] Wherein, the second collision warning mark is used to add the second collision warning mark to the foot coordinate of the obstacle key point on the second target side and the second target side.

[0026] In one embodiment, before generating the auxiliary parking image based on the position and pose information and the collision information, it further includes:

[0027] Obtaining the target point cloud data corresponding to the target vehicle and the obstacle point cloud data corresponding to the obstacle;

[0028] Correspondingly, generating the auxiliary parking image based on the position and pose information and the collision information includes:

[0029] Adding the collision information to the position and pose information to obtain the added position and pose information;

[0030] Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information.

[0031] In one embodiment, generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information includes:

[0032] Performing coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and pose information respectively to obtain the conversion processing result;

[0033] Generating the auxiliary parking image according to the conversion processing result.

[0034] In one embodiment, before generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information, it further includes:

[0035] Obtaining multiple target images collected by the roadside fusion perception system for the preset parking area;

[0036] Obtain a panoramic image corresponding to the target vehicle according to the multiple target images;

[0037] Correspondingly, the generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information includes:

[0038] Generate the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image.

[0039] In one embodiment, the auxiliary parking image includes a plurality of sub-images, and the image perspectives corresponding to the sub-images are different. The displaying the auxiliary parking image through a preset display component corresponding to the target vehicle includes:

[0040] Synchronize the sub-images at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

[0041] In a second aspect, an embodiment of the present application provides an auxiliary parking device, and the device includes:

[0042] A first acquisition module, configured to acquire point cloud data obtained by a roadside fusion perception system scanning a preset parking area, and acquire position and attitude information of a target vehicle in the preset parking area according to the point cloud data;

[0043] A second acquisition module, configured to acquire collision information corresponding to an obstacle within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information, where the collision information is used to characterize whether a collision will occur between the target vehicle and the obstacle during the parking process of the target vehicle;

[0044] A display module, configured to generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0045] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the method in the first aspect are implemented.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0047] The beneficial effects brought by the technical solution provided by the embodiment of the present application at least include:

[0048] The above-mentioned auxiliary parking method, device, computer device, and computer-readable storage medium obtain point cloud data scanned by a roadside fusion perception system for a preset parking area, and based on the point cloud data, obtain the position and attitude information of a target vehicle in the preset parking area. Then, based on the point cloud data and the position and attitude information, obtain collision information corresponding to obstacles within a preset distance of the target vehicle. This collision information is used to represent whether a collision will occur between the target vehicle and the obstacles during the parking process of the target vehicle. Then, generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle. In this way, assist the vehicle owner to park through the auxiliary parking image, which avoids unsafe phenomena such as vehicle scratches caused by the vehicle owner parking only based on personal experience, thereby improving parking safety. In addition, the embodiments of the present application generate an auxiliary parking image based on the point cloud data scanned by the roadside fusion perception system. Therefore, regardless of whether a reverse image and a reverse radar are installed in the vehicle, auxiliary parking can be achieved through the auxiliary parking image, thus avoiding the problem of poor flexibility of auxiliary parking caused by only installing reverse images and reverse radars in some high-end vehicles in the traditional technology. The embodiments of the present application also improve the flexibility of auxiliary parking. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the implementation environment of the auxiliary parking method in an embodiment;

[0050] Figure 2 Schematic flowchart of the auxiliary parking method in an embodiment;

[0051] Figure 3 Schematic flowchart of obtaining the position and attitude information of the target vehicle in another embodiment;

[0052] Figure 4 Schematic flowchart of step 301 in another embodiment;

[0053] Figure 5 Schematic flowchart of step 202 in another embodiment;

[0054] Figure 6 Schematic diagram of an exemplary target bounding box and an obstacle bounding box;

[0055] Figure 7 Schematic diagram of the position between the target corner point of an exemplary target bounding box and the first target edge of an obstacle bounding box;

[0056] Figure 8 Schematic diagram of the position between the second target edge of an exemplary target bounding box and the obstacle corner point of an obstacle bounding box;

[0057] Figure 9It is a schematic diagram of the positions of an exemplary target bounding box and an obstacle bounding box;

[0058] Figure 10 It is a schematic flowchart of an auxiliary parking method in another embodiment;

[0059] Figure 11 It is a schematic diagram of the effect of coordinate system conversion processing;

[0060] Figure 12 It is a schematic diagram of an exemplary auxiliary parking image;

[0061] Figure 13 It is a schematic diagram of another exemplary auxiliary parking image;

[0062] Figure 14 It is a schematic diagram of another exemplary auxiliary parking image;

[0063] Figure 15 It is a schematic diagram of another exemplary auxiliary parking image;

[0064] Figure 16 It is a schematic flowchart of an auxiliary parking method in another embodiment;

[0065] Figure 17 It is a schematic diagram of a display interface in an exemplary target vehicle;

[0066] Figure 18 It is a structural block diagram of an auxiliary parking device in an embodiment;

[0067] Figure 19 It is an internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0068] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0069] The auxiliary parking method, device, computer device and computer-readable storage medium provided by the embodiments of the present application can improve the parking safety of the vehicle and the flexibility of auxiliary parking. The technical solutions of the present application will be described in detail below through embodiments and with reference to the accompanying drawings. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0070] The auxiliary parking method provided by the embodiments of the present application can be applied to an implementation environment as shown in Figure 1 As shown in Figure 1As shown, the implementation environment includes a computer device 101 and a roadside integrated perception system 102, and the computer device 101 and the roadside integrated perception system 102 can communicate through a wired network or a wireless network.

[0071] Among them, the computer device 101 can be a roadside computing unit / terminal / edge server. Optionally, the computer device 101 can also be a cloud server, an in-vehicle computing unit / terminal on the vehicle side, etc. The type of the computer device 101 is not specifically limited here.

[0072] The roadside integrated perception system 102 is also known as a roadside base station or a smart base station, and it can include one or more of sensors such as a millimeter-wave radar sensor, a lidar sensor, a camera, etc. The roadside integrated perception system 102 can also be a perception system composed of multiple sensors. This perception system can run relevant algorithms to process the data collected by multiple sensors to achieve environmental perception. The algorithms include but are not limited to object detection algorithms for the corresponding sensor data, multi-sensor fusion perception algorithms, etc.

[0073] In one embodiment, as Figure 2 shown, an auxiliary parking method is provided. Taking the computer device in Figure 1 as an example for illustration. It should be noted that the method of the embodiment of the present application can also be executed by a combined system of computing devices on the vehicle side, the cloud side, and the roadside side. The specific task allocation in this method can be flexibly set based on requirements, and the embodiment of the present application does not make a limitation.

[0074] This method includes step 201, step 202, and step 203:

[0075] Step 201, the computer device obtains the point cloud data scanned by the roadside integrated perception system for a preset parking area, and based on the point cloud data, obtains the position and attitude information of the target vehicle in the preset parking area.

[0076] The preset parking area can be the perception area of the roadside integrated perception system. During the working process, the roadside integrated perception system can scan the perception area and then output point cloud data. The point cloud data contains a lot of information, such as coordinate information, color information, reflection intensity information, etc.

[0077] After the computer device obtains the point cloud data, it obtains the position and attitude information of the target vehicle in the preset parking area according to the point cloud data. Among them, the target vehicle can be any vehicle in the preset parking area, and the position and attitude information of the target vehicle is used to represent the current parking position and parking attitude of the target vehicle in the preset parking area. The position and attitude information may include the coordinates of the target key points of the target bounding box corresponding to the target vehicle, and the target vehicle can be located in the target bounding box. The target key points can be, for example, the corner points of the target bounding box, and of course, they can also be other key points of the target bounding box.

[0078] The following describes the process by which the computer device obtains the position and attitude information of the target vehicle according to the point cloud data.

[0079] In a possible implementation, the computer device can perform target detection on the point cloud data to obtain the position and attitude information of each vehicle in the preset parking area, and the computer device filters out the position and attitude information of the target vehicle from the position and attitude information of each vehicle.

[0080] In another possible implementation, refer to Figure 3 The computer device can execute Figure 3 Steps 301 and 302 shown to implement the process of obtaining the position and attitude information of the target vehicle in the preset parking area according to the point cloud data:

[0081] Step 301, the computer device determines the target point cloud data corresponding to the target vehicle from the point cloud data.

[0082] The roadside fusion perception system scans the preset parking area and outputs point cloud data, which may include the point cloud data of multiple vehicles. In this way, the computer device determines the target point cloud data corresponding to the target vehicle from the point cloud data output by the roadside fusion perception system.

[0083] In a possible implementation of step 301, refer to Figure 4 The computer device can execute Figure 4 Steps 3011 and 3012 shown to implement the process of step 301:

[0084] Step 3011, the computer device obtains the vehicle image of the target vehicle collected by the roadside fusion perception system, and performs position calibration on the vehicle image to obtain the calibrated position corresponding to the target vehicle.

[0085] In the embodiments of the present application, the roadside fusion perception system may include a camera. Exemplarily, the camera can be installed at the entrance of the preset parking area. In this way, the vehicle image of the target vehicle can be collected through the camera.

[0086] Since the installation position of the camera is fixed, the computer device can calibrate the position of the target vehicle according to the position of the target vehicle (or of course, the license plate of the target vehicle) in the vehicle image, and obtain the calibrated position corresponding to the target vehicle.

[0087] Step 3012: The computer device filters the point cloud data according to the calibrated position to obtain target point cloud data, extracts the identity information of the target vehicle from the vehicle image, and stores the identity information and the target point cloud data in an associated manner.

[0088] In the embodiment of the present application, the roadside fusion perception system may further include a lidar sensor. The point cloud data may be obtained by the lidar sensor scanning a preset parking area. In this way, the computer device can project the calibrated position corresponding to the target vehicle onto the lidar coordinate system according to the relative pose between the camera and the lidar sensor to obtain a projection position.

[0089] The point cloud data obtained by the roadside fusion perception system scanning the preset parking area may include the point cloud data of multiple vehicles in the preset parking area. Based on the point cloud data of each vehicle, the perception position of the vehicle can be obtained. In this way, the computer device can compare the projection position with each perception position, and use the point cloud data corresponding to the perception position with a coincidence degree greater than the preset coincidence degree threshold as the target point cloud data. In this way, the target point cloud data of the target vehicle is filtered out.

[0090] Furthermore, the computer device may also extract the identity information of the target vehicle from the vehicle image. This identity information may be, for example, license plate information. The computer device stores this identity information and the target point cloud data in an associated manner in the database.

[0091] Step 302: The computer device performs target detection on the target point cloud data to obtain position and attitude information.

[0092] After the computer device filters out the target point cloud data corresponding to the target vehicle, it uses a target detection algorithm to perform target detection on the target point cloud data to obtain position and attitude information.

[0093] It should be noted that the process of target detection may also be executed by the roadside fusion perception system, and the computer device obtains the position and attitude information of the target vehicle detected by the roadside fusion perception system.

[0094] Step 202: The computer device obtains collision information corresponding to obstacles within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information.

[0095] Through the above implementation, after the computer device obtains the position and attitude information of the target vehicle, the computer device obtains the collision information corresponding to the obstacles within the preset distance range of the target vehicle according to the point cloud data and the position and attitude information, and the collision information is used to characterize whether the target vehicle will collide with the obstacles during the parking process of the target vehicle.

[0096] As an implementation, the computer device can determine the obstacle point cloud data corresponding to the obstacles from the point cloud data according to the position and attitude information, and then detect whether the target vehicle will collide with the obstacles during the parking process of the target vehicle according to the obstacle point cloud data and the position and attitude information of the target vehicle, so as to obtain the collision information.

[0097] Step 203, the computer device generates an auxiliary parking image based on the position and attitude information and the collision information, and displays the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0098] In a possible implementation, the computer device can add the collision information to the position and attitude information to obtain the added position and attitude information, and the computer device performs rendering processing on the added position and attitude information to obtain an auxiliary parking image, and displays the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0099] Exemplarily, the computer device can send the auxiliary parking image to the target vehicle for the target vehicle to display the auxiliary parking image through the preset display component configured thereon. In this way, the vehicle owner can intuitively understand the current parking position and parking attitude of the target vehicle in the preset parking area, and the reminder of whether the target vehicle will collide with the obstacles during the parking process through the auxiliary parking image, so as to assist the vehicle owner to park safely.

[0100] In the above embodiment, the point cloud data obtained by the roadside fusion perception system scanning the preset parking area is acquired, and according to the point cloud data, the position and attitude information of the target vehicle in the preset parking area is obtained. Then, according to the point cloud data and the position and attitude information, the collision information corresponding to the obstacles within the preset distance range of the target vehicle is obtained, and the collision information is used to represent whether a collision will occur between the target vehicle and the obstacles during the parking process of the target vehicle. Furthermore, an auxiliary parking image is generated based on the position and attitude information and the collision information, and the auxiliary parking image is displayed through a preset display component corresponding to the target vehicle. In this way, the owner is assisted in parking through the auxiliary parking image, which avoids unsafe phenomena such as vehicle scratches caused by the owner parking only based on personal experience, thereby improving the parking safety. In addition, the embodiment of the present application generates the auxiliary parking image based on the point cloud data scanned by the roadside fusion perception system. Therefore, regardless of whether a reverse image and a reverse radar are installed in the vehicle, the auxiliary parking can be realized through the auxiliary parking image, thus avoiding the problem of poor flexibility of auxiliary parking caused by only installing reverse images and reverse radars in some high-end vehicles in the traditional technology. The embodiment of the present application also improves the flexibility of auxiliary parking, which is conducive to the popularization of auxiliary parking.

[0101] The auxiliary parking method in this embodiment is particularly applicable when there is no positioning information such as GPS or Beidou positioning signal, such as in an underground parking lot. Since the vehicle cannot obtain its own position when there is no GPS or Beidou positioning information, and the general auxiliary parking scheme is based on the position (it is necessary to know the relative position relationship between the vehicle and the parking space to park), this embodiment can realize auxiliary parking based on the positioning of the lidar when there is no signal.

[0102] Based on the same inventive concept, as a variant, another embodiment of the present application provides another auxiliary parking method. The method includes: acquiring in real time the real-time point cloud data obtained by the roadside fusion perception system scanning the preset parking area, and according to the real-time point cloud data, acquiring the position and attitude information of the target vehicle in the preset parking area; according to the real-time point cloud data and the position and attitude information, acquiring the real-time collision information corresponding to the obstacles within the preset distance range of the target vehicle, and the real-time collision information is used to represent the risk level of a collision occurring between the target vehicle and the obstacles during the real-time parking process of the target vehicle; generating a real-time auxiliary parking image based on the position and attitude information and the real-time collision information, and displaying the real-time auxiliary parking image through a preset display component corresponding to the target vehicle, and / or inputting the real-time auxiliary parking image into an automatic parking decision model to generate a parking strategy for the target vehicle. The difference between this embodiment and the embodiment above mainly lies in the practical expansion of the obtained auxiliary information. The basic principle is basically similar and will not be elaborated here.

[0103] In another embodiment, when the assisted parking scenario is roadside parking, after the roadside fusion perception system obtains the detection result, it can calibrate the target detection result. The specific method is as follows: If the number of targets in the detection result of the camera in the roadside fusion perception system is greater than the detection number of the lidar, it is highly likely that the lidar detects multiple targets as a single target at this time. Therefore, the detection target of the lidar can be adjusted and optimized based on the image to obtain an accurate detection result. This method is applicable to the detection of smaller targets such as non-motor vehicles. At this time, due to the small size of the target and the relatively lower safety of the personnel involved compared to the vehicle inside the vehicle, more careful and precise assisted parking is required.

[0104] In another embodiment of the present application, the assisted parking method may include at least two modes: one is the normal mode and the other is the cautious parking mode. Specifically: When the obstacles within the preset distance range of the target vehicle meet the cautious parking mode, it is adjusted to the cautious mode. This condition can be detecting pedestrians or non-motor vehicles. The cautious parking mode can be outputting a prompt message, suggesting that the driver wait a little while, controlling the vehicle to be temporarily stationary, etc. In one embodiment, based on Figure 2 the embodiment shown, refer to Figure 5 , this embodiment relates to the process of how a computer device obtains the collision information corresponding to the obstacles within the preset distance range of the target vehicle according to the point cloud data and the position and attitude information. As Figure 5 shown, the position and attitude information includes the coordinates of the target key points of the target inclusion box corresponding to the target vehicle. Step 202 includes Step 501, Step 502, and Step 503:

[0105] Step 501, the computer device determines the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and the preset distance range threshold.

[0106] As an implementation manner, the target key points can be the corner points of the target inclusion box. In this way, the computer device can determine the center point coordinates of the target inclusion box according to the coordinates of the target key points.

[0107] The computer device determines the candidate point cloud data within the preset distance range threshold of the center point coordinates in the point cloud data scanned by the roadside fusion perception system. The candidate point cloud data includes the target point cloud data corresponding to the position and attitude information of the target vehicle and the obstacle point cloud data corresponding to the obstacle. In this way, the computer device determines the obstacle point cloud data.

[0108] Step 502, the computer device performs target detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle inclusion box corresponding to the obstacle.

[0109] The computer device uses a target detection algorithm to perform target detection on the obstacle point cloud data, and obtains the coordinates of the obstacle key points of the obstacle bounding box corresponding to the obstacle. The obstacle can be located within the obstacle bounding box. The obstacle key points can be the corner points of the obstacle bounding box, and can also be other key points such as the midpoints of the sides of the obstacle bounding box.

[0110] Exemplarily, taking the obstacle as a vehicle, refer to Figure 6 , Figure 6 which is a schematic diagram of an exemplary target bounding box and an obstacle bounding box. As Figure 6 shown, the arrow direction is the vehicle head direction. The computer device performs target detection on the obstacle point cloud data within a preset distance range threshold (as shown by the circle in Figure 6 ), and obtains the coordinates of the obstacle key points (the two corner points of the obstacle bounding box 1 and the midpoints of the two sides of the obstacle bounding box 1) of the obstacle bounding box 1 corresponding to the obstacle 1, and the coordinates of the obstacle key points (the midpoints of the two sides of the obstacle bounding box 2) of the obstacle bounding box 2 corresponding to the obstacle 2.

[0111] Step 503, the computer device obtains collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0112] The computer device compares the distances between the coordinates of the target key points and the coordinates of the obstacle key points. If the distance between the target key point and the obstacle key point is small, the computer device determines that the target vehicle will collide with the obstacle, otherwise, it determines that the target vehicle will not collide with the obstacle, thereby obtaining collision information.

[0113] In a possible implementation manner of step 503, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points. The computer device obtains collision information according to the coordinates of the target key points and the coordinates of the obstacle key points. Specifically, it can detect whether the distance between the target corner point and the first target side of the obstacle bounding box is less than the first distance threshold according to the coordinates of the target corner points and the coordinates of the obstacle corner points, and obtain collision information according to the detection result.

[0114] Refer to Figure 7 , Figure 7It is a schematic diagram of the position between the target corner point of an exemplary target bounding box and the first target edge of an obstacle bounding box. The target corner point can be the corner point of the target bounding box closest to the obstacle bounding box, and the first target edge can be the edge of the obstacle bounding box closest to the target bounding box. Based on the coordinates of the obstacle corner points in the obstacle bounding box, the computer device can determine the position of the first target edge containing the obstacle corner point. In this way, the computer device can calculate the distance between the target corner point and the first target edge of the obstacle bounding box, and detect whether the distance between the target corner point and the first target edge is less than the first distance threshold.

[0115] If the computer device detects, based on the coordinates of the target corner point and the coordinates of the obstacle corner point, that the distance between the target corner point and the first target edge of the obstacle bounding box is less than the first distance threshold, the computer device generates a first collision warning mark and uses the first collision warning mark as collision information. This first collision warning mark is used to add a first collision warning mark to the coordinates of the target corner point.

[0116] In another possible implementation of step 503, the coordinates of the target key point include the coordinates of the target corner point, and the coordinates of the obstacle key point include the coordinates of the obstacle corner point. The computer device obtains collision information based on the coordinates of the target key point and the coordinates of the obstacle key point. Specifically, it can detect whether the distance between the second target edge of the target bounding box and the obstacle corner point is less than the second distance threshold based on the coordinates of the target corner point and the coordinates of the obstacle corner point, and obtain collision information based on the detection result.

[0117] See Figure 8 , Figure 8 It is a schematic diagram of the position between the second target edge of an exemplary target bounding box and the obstacle corner point of an obstacle bounding box. The obstacle corner point can be the corner point of the obstacle bounding box closest to the target bounding box, and the second target edge can be the edge of the target bounding box closest to the obstacle bounding box. Based on the coordinates of the target corner point, the computer device can determine the position of the second target edge containing the target corner point. In this way, the computer device can calculate the distance between the second target edge and the obstacle corner point, and detect whether the distance between the second target edge and the obstacle corner point is less than the second distance threshold.

[0118] If the computer device detects, based on the coordinates of the target corner point and the coordinates of the obstacle corner point, that the distance between the second target edge of the target bounding box and the obstacle corner point is less than the second distance threshold, the computer device generates a second collision warning mark and uses the second collision warning mark as collision information. This second collision warning mark is used to add a second collision warning mark to the foot coordinate of the obstacle key point on the second target edge and the second target edge.

[0119] It should be noted that the two implementation manners of step 503 described above can also be implemented simultaneously. Refer to Figure 9 , which is a schematic diagram of the positions of an exemplary target bounding box and an obstacle bounding box. As Figure 9 shown, the computer device can detect whether the distance between the target corner point and the first target side of the obstacle bounding box is less than the first distance threshold based on the coordinates of the target corner point and the coordinates of the obstacle corner point, and detect whether the distance between the second target side of the target bounding box and the obstacle corner point is less than the second distance threshold.

[0120] If the distance between the target corner point and the first target side of the obstacle bounding box is less than the first distance threshold, the computer device generates a first collision warning mark. If the distance between the second target side of the target bounding box and the obstacle corner point is less than the second distance threshold, the computer device generates a second collision warning mark, thereby obtaining collision information.

[0121] In this way, the computer device adds collision information to the position and attitude information. Specifically, it adds a first collision warning mark to the coordinates of the target corner point, and adds a second collision warning mark to the foot coordinate of the obstacle corner point on the second target side of the target bounding box. Of course, it can also add a second collision warning mark to the second target side, and so on.

[0122] After the computer device adds collision information to the position and attitude information, it can perform rendering processing on the added position and attitude information to obtain an auxiliary parking image, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0123] In this way, the vehicle owner can more intuitively see the possible collision positions between the target vehicle and obstacles during the parking process through the collision information in the auxiliary parking image, thereby assisting the vehicle owner to park safely and improving parking safety.

[0124] In one embodiment, based on Figure 2 the embodiment shown, refer to Figure 10 , this embodiment relates to the process of how the computer device generates an auxiliary parking image. As Figure 10 shown, in the auxiliary parking method of this embodiment, before step 203, there is also step 204:

[0125] Step 204, the computer device acquires target point cloud data corresponding to the target vehicle and obstacle point cloud data corresponding to the obstacle.

[0126] The process by which the computer device acquires target point cloud data corresponding to the target vehicle and obstacle point cloud data corresponding to the obstacle according to the point cloud data scanned by the roadside fusion perception system for a preset parking area can refer to the implementation process of the above embodiment and will not be elaborated here.

[0127] Correspondingly, step 203 includes step 2031, step 2032, and step 2033:

[0128] In step 2031, the computer device adds collision information to the position and attitude information to obtain the added position and attitude information.

[0129] As described above, if the distance between the target corner point of the target bounding box and the first target edge of the obstacle bounding box is less than the first distance threshold, the computer device adds a first collision warning mark to the coordinates of the target corner point; if the distance between the second target edge of the target bounding box and the obstacle corner point of the obstacle bounding box is less than the second distance threshold, the computer device adds a second collision warning mark to the foot coordinate in the second target edge of the target bounding box and adds a second collision warning mark to the second target edge to obtain the added position and attitude information.

[0130] In step 2032, the computer device generates an auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information.

[0131] In a possible implementation manner of step 2032, the computer device can simultaneously display the target point cloud data, the obstacle point cloud data, and the added position and attitude information in the auxiliary parking image. In this way, the car owner can not only intuitively see the possible collision positions with obstacles during the parking process of the target vehicle, but also see the obstacle point cloud data, which is beneficial for the car owner to master the general position of the obstacle and further improves the parking safety.

[0132] In another possible implementation manner of step 2032, the computer device can respectively perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and attitude information to obtain the conversion processing result. Then, the computer device generates an auxiliary parking image according to the conversion processing result.

[0133] Since the point cloud data and the position and attitude information of the target vehicle obtained from the point cloud data are both from the perception angle of the roadside fusion perception system, this perception angle may not conform to the user's habit. To make it conform to the user's habit, the computer device can perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and attitude information. For example, if the angle between the target bounding box in the added position and attitude information and the horizontal plane is greater than 0, the computer device can adjust the angle between the target bounding box and the horizontal plane to 0 through coordinate system conversion processing.

[0134] Exemplarily, refer to Figure 11, which is a schematic diagram of the exemplary coordinate system conversion processing effect. Assume that the angle between the target bounding box and the horizontal plane is θ (θ>0), and the center point coordinates of the target bounding box are (x0, y0, z0). For any point (x1, y1, z1) in the added position and attitude information, the computer device can perform coordinate system conversion processing using Formula 1 to obtain the coordinates (x1′, y1′, z1′) after the coordinate system conversion processing.

[0135]

[0136] In this way, the computer device performs coordinate system conversion processing on each point in the target point cloud data, obstacle point cloud data, and the added position and attitude information to obtain the conversion processing result. The conversion processing result is the target point cloud data after the coordinate system conversion processing, the obstacle point cloud data after the coordinate system conversion processing, and the position and attitude information after the coordinate system conversion processing. Then, the computer device generates an auxiliary parking image according to the conversion processing result.

[0137] Step 2033, the computer device displays the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0138] Exemplarily, refer to Figures 12 - 15 , Figures 12 - 15 which are respectively an exemplary auxiliary parking image. Among them, Figure 12 is a three-dimensional schematic diagram of the target point cloud data, obstacle point cloud data, and the added position and attitude information after the coordinate system conversion processing, Figure 13 as shown is a top view schematic diagram of the target point cloud data, obstacle point cloud data, and the added position and attitude information after the coordinate system conversion processing, Figure 14 as shown is a side view schematic diagram of the target point cloud data, obstacle point cloud data, and the added position and attitude information after the coordinate system conversion processing, Figure 15 as shown is a front view schematic diagram of the target point cloud data, obstacle point cloud data, and the added position and attitude information after the coordinate system conversion processing.

[0139] As Figures 12 - 15 shown, the point cloud data within the target bounding box is the target point cloud data after the coordinate system conversion processing, and the point cloud data outside the target bounding box is the obstacle point cloud data after the coordinate system conversion processing. It should be noted that the obstacle point cloud data after the coordinate system conversion processing may also include scene point cloud data, which is not specifically limited here.

[0140] In a possible implementation manner, since Figures 12 - 15 are auxiliary parking images from different image perspectives, the computer device can also display Figures 12 - 15 at different display positions on the same display interface through a preset display component corresponding to the target vehicle.each of the images, so that the vehicle owner can intuitively see the auxiliary parking images from different image perspectives, improving the reliability of auxiliary parking.

[0141] In the above embodiment, collision information is added to the position and attitude information, and then, coordinate system conversion processing is respectively performed on the target point cloud data, the obstacle point cloud data, and the added position and attitude information to obtain the conversion processing result, and then an auxiliary parking image is generated according to the conversion processing result. In this way, through the coordinate system conversion processing, the generated auxiliary parking image can be made more in line with the user's habits, improving the flexibility of auxiliary parking.

[0142] Based on Figure 10 the embodiment shown, see Figure 16 , in this embodiment, before step 2032, steps 2034 and 2035 are further included:

[0143] Step 2034, the computer device acquires multiple target images collected by the roadside fusion perception system for a preset parking area.

[0144] The roadside fusion perception system may include multiple cameras installed at different positions, and multiple target images are acquired by the multiple cameras for the preset parking area.

[0145] Step 2035, the computer device acquires a panoramic image corresponding to the target vehicle according to the multiple target images.

[0146] The computer device obtains the perceived position of the target vehicle according to the target point cloud data of the target vehicle. The computer device intercepts the image areas within a preset distance range around the target vehicle from each target image according to the perceived position, and performs image stitching on the intercepted image areas to obtain a panoramic image, where the preset distance range can be set by itself during implementation, such as set to 5 meters, 3 meters, etc.

[0147] During the image stitching process, the computer device can use the SIFT (Scale Invariant Feature Transform) operator to extract image features, use the Euclidean distance to evaluate the matching results, use KNN (K-Nearest Neighbor) for search, use RANSAC (RANdom SAmple Consensus) for iterative optimization to find a set of transformation parameters with the most inliers, and finally align and stitch the pixels to obtain a panoramic image.

[0148] Correspondingly, step 2032 may include:

[0149] Step 2032a, the computer device generates an auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image.

[0150] After the computer device obtains the panoramic image, the computer device can extract the size, position coordinates, and heading angle of the target bounding box from the added position and attitude information and the target point cloud data, extract the position coordinates of the obstacles from the obstacle point cloud information, and extract information such as the vehicle type and color from the panoramic image. Then, according to the extracted vehicle type, a preset 3D model is selected, and the other extracted information is used to render the 3D model to obtain the auxiliary parking image.

[0151] As an implementation, the computer device can also send all the extracted information to the target vehicle / driver terminal for the target vehicle / driver terminal to generate an auxiliary parking image according to the extracted information.

[0152] It should be noted that, similar to the above embodiments, before step 2032a, the computer device can also perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image, and then generate the auxiliary parking image after the coordinate system conversion processing, which is not specifically limited here.

[0153] In the embodiments of the present application, the auxiliary parking image can include multiple sub-images, and the image perspectives corresponding to the sub-images are different. The process of the computer device displaying the auxiliary parking image through the preset display component corresponding to the target vehicle can specifically be that the computer device synchronizes the sub-images at different display positions on the same display interface through the preset display component corresponding to the target vehicle.

[0154] See Figure 17 , Figure 17 is a schematic diagram of the display interface in an exemplary target vehicle. That is, for each image perspective sensor in the roadside fusion perception system, the computer device obtains the sub-image corresponding to the image perspective through the above embodiments, and then displays the sub-images of different image perspectives at different display positions on the same display interface, so that the vehicle owner can intuitively see the auxiliary parking images of different image perspectives, improving the reliability of auxiliary parking.

[0155] In one embodiment, an auxiliary parking method is provided, including:

[0156] Step a, the computer device obtains the point cloud data scanned by the roadside fusion perception system for a preset parking area, and obtains the vehicle image of the target vehicle collected by the roadside fusion perception system, and performs position calibration on the vehicle image to obtain the calibration position corresponding to the target vehicle.

[0157] Step b: The computer device filters the point cloud data according to the calibrated position to obtain the target point cloud data of the target vehicle, extracts the identity information of the target vehicle from the vehicle image, and stores the identity information in association with the target point cloud data.

[0158] Step c: The computer device performs target detection on the target point cloud data to obtain the position and attitude information.

[0159] The position and attitude information is used to represent the current parking position and parking attitude of the target vehicle in the preset parking area. The position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle.

[0160] Step d: The computer device determines the obstacle corresponding obstacle point cloud data in the point cloud data according to the coordinates of the target key points and the preset distance range threshold.

[0161] Step e: The computer device performs target detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle corresponding obstacle bounding box.

[0162] Step f: The computer device obtains the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0163] The collision information is used to represent whether a collision will occur between the target vehicle and the obstacle during the parking process of the target vehicle.

[0164] Optionally, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes:

[0165] If it is detected that the distance between the target corner point and the first target edge of the obstacle bounding box is less than the first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, a first collision warning mark is generated, and the first collision warning mark is used as the collision information; wherein, the first collision warning mark is used to add a first collision warning mark to the coordinates of the target corner point.

[0166] Optionally, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes:

[0167] If it is detected that the distance between the second target edge of the target bounding box and the obstacle corner point is less than the second distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, a second collision warning mark is generated, and the second collision warning mark is used as the collision information; wherein, the second collision warning mark is used to add a second collision warning mark to the foot coordinate of the obstacle key point on the second target edge and the second target edge.

[0168] Step g, the computer device adds collision information to the position and attitude information to obtain the added position and attitude information.

[0169] Step h, the computer device respectively performs coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and attitude information to obtain the conversion processing result.

[0170] Step i, the computer device acquires multiple target images collected by the roadside fusion perception system for a preset parking area, and based on the multiple target images, acquires a panoramic image corresponding to the target vehicle.

[0171] Step j, the computer device generates an auxiliary parking image based on the conversion processing result and the panoramic image, and displays the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0172] The auxiliary parking image includes multiple sub-images, and the image perspectives corresponding to the sub-images are different. The computer device can synchronize the sub-images at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

[0173] It should be understood that although the steps in the above flowchart are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above flowchart may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps or stages in other steps.

[0174] In one embodiment, as Figure 18 shown, an auxiliary parking device is provided, including:

[0175] A first acquisition module 100, configured to acquire point cloud data scanned by the roadside fusion perception system for a preset parking area, and based on the point cloud data, acquire the position and attitude information of a target vehicle in the preset parking area;

[0176] A second acquisition module 200, configured to acquire collision information corresponding to an obstacle within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information, where the collision information is used to characterize whether a collision will occur between the target vehicle and the obstacle during the parking process of the target vehicle;

[0177] A display module 300 is configured to generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0178] In one embodiment, the first acquisition module includes:

[0179] A first determination unit configured to determine target point cloud data corresponding to the target vehicle from the point cloud data;

[0180] A target detection unit configured to perform target detection on the target point cloud data to obtain the position and attitude information, where the position and attitude information is used to characterize the current parking position and parking attitude of the target vehicle in the preset parking area.

[0181] In one embodiment, the first determination unit is specifically configured to acquire a vehicle image of the target vehicle collected by the roadside fusion perception system, perform position calibration on the vehicle image to obtain a calibrated position corresponding to the target vehicle; screen the point cloud data according to the calibrated position to obtain the target point cloud data, and extract identity information of the target vehicle from the vehicle image, and store the identity information in association with the target point cloud data.

[0182] In one embodiment, the position and attitude information includes coordinates of target key points of a target bounding box corresponding to the target vehicle; the second acquisition module includes:

[0183] A second determination unit configured to determine obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold;

[0184] A detection unit configured to perform target detection on the obstacle point cloud data to obtain coordinates of obstacle key points of an obstacle bounding box corresponding to the obstacle;

[0185] An acquisition unit configured to acquire the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0186] The coordinates of the target key points include coordinates of target corner points, and the coordinates of the obstacle key points include coordinates of obstacle corner points. Optionally, the acquisition unit is specifically configured to generate a first collision warning mark if it is detected that the distance between the target corner point and a first target edge of the obstacle bounding box is less than a first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, and use the first collision warning mark as the collision information; wherein the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

[0187] Optionally, the obtaining unit is specifically configured to, if it is detected that the distance between the second target side of the target bounding box and the obstacle corner point is less than a second distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, generate a second collision warning mark and use the second collision warning mark as the collision information; wherein, the second collision warning mark is used to add the second collision warning mark to the foot coordinate of the obstacle key point on the second target side and the second target side.

[0188] In one embodiment, the apparatus further includes:

[0189] A third obtaining module, configured to obtain target point cloud data corresponding to the target vehicle and obstacle point cloud data corresponding to the obstacle;

[0190] The display module includes:

[0191] An adding unit, configured to add the collision information to the position and attitude information to obtain the added position and attitude information;

[0192] A generating unit, configured to generate the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information.

[0193] In one embodiment, the generating unit is specifically configured to perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and attitude information respectively to obtain a conversion processing result; and generate the auxiliary parking image according to the conversion processing result.

[0194] In one embodiment, the generating unit is specifically configured to obtain multiple target images collected by the roadside fusion perception system for the preset parking area; obtain a panoramic image corresponding to the target vehicle according to the multiple target images; and generate the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image.

[0195] In one embodiment, the auxiliary parking image includes multiple sub-images, and the image perspectives corresponding to the sub-images are different. The display module further includes:

[0196] A display unit, configured to synchronize the sub-images at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

[0197] Specific limitations on the auxiliary parking device can be referred to the limitations on the auxiliary parking method in the above text, which will not be elaborated here. Each module in the above auxiliary parking device can be implemented in whole or in part by software, hardware, and their combinations. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0198] In one embodiment, a computer device is provided, and its internal structure diagram can be as Figure 19 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data of the auxiliary parking method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements an auxiliary parking method.

[0199] Those skilled in the art can understand that Figure 19 the structure shown in

[0200] is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0200] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0201] Obtain the point cloud data scanned by the roadside fusion perception system for a preset parking area, and based on the point cloud data, obtain the position and attitude information of the target vehicle in the preset parking area;

[0202] Based on the point cloud data and the position and attitude information, obtain the collision information corresponding to the obstacles within a preset distance range of the target vehicle. The collision information is used to characterize whether a collision will occur between the target vehicle and the obstacles during the parking process of the target vehicle;

[0203] Generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0204] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0205] Determine the target point cloud data corresponding to the target vehicle from the point cloud data;

[0206] Perform target detection on the target point cloud data to obtain the position and attitude information, where the position and attitude information is used to characterize the current parking position and parking attitude of the target vehicle in the preset parking area.

[0207] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0208] Obtain the vehicle image of the target vehicle collected by the roadside fusion perception system, and perform position calibration on the vehicle image to obtain the calibrated position corresponding to the target vehicle;

[0209] Filter the point cloud data according to the calibrated position to obtain the target point cloud data, extract the identity information of the target vehicle from the vehicle image, and store the identity information in association with the target point cloud data.

[0210] In one embodiment, the position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle. When the processor executes the computer program, the following steps are further implemented:

[0211] Determine the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold;

[0212] Perform target detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle bounding box corresponding to the obstacle;

[0213] Obtain the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0214] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points. When the processor executes the computer program, the following steps are further implemented:

[0215] If it is detected that the distance between the target corner point and the first target edge of the obstacle bounding box is less than the first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, generate a first collision warning mark, and use the first collision warning mark as the collision information;

[0216] Wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

[0217] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points. When the processor executes the computer program, the following steps are further implemented:

[0218] If, according to the coordinates of the target corner points and the coordinates of the obstacle corner points, it is detected that the distance between the second target side of the target bounding box and the obstacle corner point is less than the second distance threshold, a second collision warning mark is generated, and the second collision warning mark is used as the collision information;

[0219] Wherein, the second collision warning mark is used to add the second collision warning mark to the foot coordinate of the obstacle key point on the second target side and the second target side.

[0220] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0221] Obtain the target point cloud data corresponding to the target vehicle and the obstacle point cloud data corresponding to the obstacle;

[0222] Correspondingly, generating the auxiliary parking image based on the position and pose information and the collision information includes:

[0223] Adding the collision information to the position and pose information to obtain the added position and pose information;

[0224] Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data and the added position and pose information.

[0225] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0226] Perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data and the added position and pose information respectively to obtain the conversion processing result;

[0227] Generate the auxiliary parking image according to the conversion processing result.

[0228] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0229] Obtain multiple target images collected by the roadside fusion perception system for the preset parking area;

[0230] Obtain the panoramic image corresponding to the target vehicle according to the multiple target images;

[0231] Correspondingly, generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information includes:

[0232] Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image.

[0233] In one embodiment, the auxiliary parking image includes multiple sub-images, and the image perspectives corresponding to each sub-image are different. When the processor executes the computer program, the following steps are further implemented:

[0234] Synchronize each of the sub-images at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

[0235] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0236] Obtain the point cloud data scanned by the roadside fusion perception system for a preset parking area, and obtain the position and attitude information of the target vehicle in the preset parking area according to the point cloud data;

[0237] Obtain the collision information corresponding to the obstacles within a preset distance range of the target vehicle according to the point cloud data and the position and attitude information, where the collision information is used to characterize whether a collision will occur between the target vehicle and the obstacles during the parking process of the target vehicle;

[0238] Generate an auxiliary parking image based on the position and attitude information and the collision information, and display the auxiliary parking image through a preset display component corresponding to the target vehicle.

[0239] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0240] Determine the target point cloud data corresponding to the target vehicle from the point cloud data;

[0241] Perform target detection on the target point cloud data to obtain the position and attitude information, where the position and attitude information is used to characterize the current parking position and parking attitude of the target vehicle in the preset parking area.

[0242] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented:

[0243] Obtain the vehicle image of the target vehicle collected by the roadside fusion perception system, and perform position calibration on the vehicle image to obtain the calibrated position corresponding to the target vehicle;

[0244] Filter the point cloud data according to the calibrated position to obtain the target point cloud data, extract the identity information of the target vehicle from the vehicle image, and store the identity information and the target point cloud data in an associated manner.

[0245] In one embodiment, the position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle. When the computer program is executed by a processor, the following steps are further implemented:

[0246] Determine the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold;

[0247] Perform target detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle bounding box corresponding to the obstacle;

[0248] Obtain the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points.

[0249] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points. When the computer program is executed by a processor, the following steps are further implemented:

[0250] If it is detected according to the coordinates of the target corner points and the coordinates of the obstacle corner points that the distance between the target corner points and the first target side of the obstacle bounding box is less than a first distance threshold, generate a first collision warning mark and use the first collision warning mark as the collision information;

[0251] Wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner points.

[0252] In one embodiment, the coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points. When the computer program is executed by a processor, the following steps are further implemented:

[0253] If it is detected according to the coordinates of the target corner points and the coordinates of the obstacle corner points that the distance between the second target side of the target bounding box and the obstacle corner points is less than a second distance threshold, generate a second collision warning mark and use the second collision warning mark as the collision information;

[0254] Wherein, the second collision warning mark is used to add the second collision warning mark to the foot coordinates of the obstacle key points on the second target side and the second target side.

[0255] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0256] Obtain the target point cloud data corresponding to the target vehicle and the obstacle point cloud data corresponding to the obstacle;

[0257] Correspondingly, generating the auxiliary parking image based on the position and attitude information and the collision information includes:

[0258] Add the collision information to the position and attitude information to obtain the added position and attitude information;

[0259] Generate the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information.

[0260] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0261] Perform coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and attitude information respectively to obtain the conversion processing result;

[0262] Generate the auxiliary parking image according to the conversion processing result.

[0263] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0264] Obtain multiple target images collected by the roadside fusion perception system for the preset parking area;

[0265] Obtain the panoramic image corresponding to the target vehicle according to the multiple target images;

[0266] Correspondingly, generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and attitude information includes:

[0267] Generate the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and attitude information, and the panoramic image.

[0268] In one embodiment, the auxiliary parking image includes multiple sub-images, and the image perspectives corresponding to each sub-image are different. When the computer program is executed by a processor, the following steps are further implemented:

[0269] Synchronize each sub-image at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

[0270] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0271] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0272] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. An auxiliary parking method, characterized in that, The method includes: Obtaining in real time the real-time point cloud data scanned by the roadside fusion perception system for a preset parking area, and obtaining the position and attitude information of a target vehicle in the preset parking area according to the real-time point cloud data; Obtaining real-time collision information corresponding to obstacles within a preset distance range of the target vehicle according to the real-time point cloud data and the position and attitude information, where the real-time collision information is used to characterize the risk level of a collision with the obstacles during the real-time parking process of the target vehicle; Generating auxiliary parking information based on the position and attitude information and the real-time collision information, and displaying the auxiliary parking information through a preset display component corresponding to the target vehicle, where the auxiliary parking information includes an image showing the real-time parking process of the target vehicle and / or parking strategy information; The position and attitude information includes the coordinates of target key points of a target bounding box corresponding to the target vehicle; the obtaining of the real-time collision information corresponding to obstacles within a preset distance range of the target vehicle according to the real-time point cloud data and the position and attitude information includes: Determining the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold; Performing target detection on the obstacle point cloud data to obtain the coordinates of obstacle key points of an obstacle bounding box corresponding to the obstacle; Obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points; The coordinates of the target key points include the coordinates of target corner points, and the coordinates of the obstacle key points include the coordinates of obstacle corner points; the obtaining of the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes: If it is detected that the distance between the target corner point and a first target side of the obstacle bounding box is less than a first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, generating a first collision warning mark and using the first collision warning mark as the collision information; Wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

2. The method according to claim 1, wherein The obtaining of the position and attitude information of the target vehicle in the preset parking area according to the point cloud data includes: Determining the target point cloud data corresponding to the target vehicle from the point cloud data; Performing target detection on the target point cloud data to obtain the position and attitude information, where the position and attitude information is used to characterize the current parking position and parking attitude of the target vehicle in the preset parking area.

3. The method according to claim 2, wherein The determining of the target point cloud data corresponding to the target vehicle from the point cloud data includes: Obtaining a vehicle image of the target vehicle collected by the roadside fusion perception system, and performing position calibration on the vehicle image to obtain a calibrated position corresponding to the target vehicle; Filtering the point cloud data according to the calibrated position to obtain the target point cloud data, and extracting the identity information of the target vehicle from the vehicle image and associatively storing the identity information with the target point cloud data.

4. The method according to claim 1, characterized in that, The coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; Obtaining the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes: If, according to the coordinates of the target corner points and the coordinates of the obstacle corner points, it is detected that the distance between the second target side of the target bounding box and the obstacle corner point is less than a second distance threshold, a second collision warning mark is generated, and the second collision warning mark is used as the collision information; Wherein, the second collision warning mark is used to add the second collision warning mark to the foot coordinate of the obstacle key point on the second target side and the second target side.

5. The method according to claim 1, characterized in that Before generating the auxiliary parking image based on the position and pose information and the collision information, it further includes: Obtaining the target point cloud data corresponding to the target vehicle and the obstacle point cloud data corresponding to the obstacle; Correspondingly, generating the auxiliary parking image based on the position and pose information and the collision information includes: Adding the collision information to the position and pose information to obtain the added position and pose information; Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information.

6. The method according to claim 5, characterized in that Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information includes: Performing coordinate system conversion processing on the target point cloud data, the obstacle point cloud data, and the added position and pose information respectively to obtain the conversion processing result; Generating the auxiliary parking image according to the conversion processing result.

7. The method according to claim 5, wherein Before generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information, it further includes: Obtaining multiple target images collected by the roadside fusion perception system for the preset parking area; Obtaining a panoramic image corresponding to the target vehicle according to the multiple target images; Correspondingly, generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, and the added position and pose information includes: Generating the auxiliary parking image according to the target point cloud data, the obstacle point cloud data, the added position and pose information, and the panoramic image.

8. The method according to claim 1, wherein The auxiliary parking image includes multiple sub-images, and the image perspectives corresponding to each sub-image are different. Displaying the auxiliary parking image through a preset display component corresponding to the target vehicle includes: Synchronizing each sub-image at different display positions on the same display interface through a preset display component corresponding to the target vehicle.

9. An auxiliary parking method, characterized in that, The method includes: Real-time obtaining the real-time point cloud data scanned by the roadside fusion perception system for the preset parking area, and obtaining the position and pose information of the target vehicle in the preset parking area according to the real-time point cloud data; Based on the real-time point cloud data and the position and attitude information, obtain the real-time collision information corresponding to the obstacles within the preset distance range of the target vehicle, where the real-time collision information is used to characterize the risk level of collision with the obstacles during the real-time parking process of the target vehicle; Generate a real-time assisted parking image based on the position and attitude information and the real-time collision information, and display the real-time assisted parking image through a preset display component corresponding to the target vehicle, and / or input the real-time assisted parking image into an automatic parking decision model to generate a parking strategy for the target vehicle; The position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle; the obtaining of the real-time collision information corresponding to the obstacles within the preset distance range of the target vehicle according to the real-time point cloud data and the position and attitude information includes: Determine the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold; Perform object detection on the obstacle point cloud data to obtain the coordinates of the obstacle key points of the obstacle bounding box corresponding to the obstacle; Obtain the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points; The coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the obstacle key points include the coordinates of the obstacle corner points; the obtaining of the collision information according to the coordinates of the target key points and the coordinates of the obstacle key points includes: If it is detected that the distance between the target corner point and the first target edge of the obstacle bounding box is less than a first distance threshold according to the coordinates of the target corner point and the coordinates of the obstacle corner point, generate a first collision warning mark and use the first collision warning mark as the collision information; Wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

10. An auxiliary parking device, characterized in that, The device includes: A first acquisition module, configured to acquire the point cloud data obtained by the roadside fusion perception system scanning a preset parking area, and obtain the position and attitude information of the target vehicle in the preset parking area according to the point cloud data; A second acquisition module, configured to obtain collision information corresponding to the obstacles within the preset distance range of the target vehicle according to the point cloud data and the position and attitude information, where the collision information is used to characterize whether the target vehicle will collide with the obstacles during the parking process of the target vehicle; A display module, configured to generate an assisted parking image based on the position and attitude information and the collision information, and display the assisted parking image through a preset display component corresponding to the target vehicle; The position and attitude information includes the coordinates of the target key points of the target bounding box corresponding to the target vehicle; the second acquisition module includes: A second determination unit, configured to determine the obstacle point cloud data corresponding to the obstacle in the point cloud data according to the coordinates of the target key points and a preset distance range threshold; A detection unit, configured to perform target detection on the obstacle point cloud data to obtain the coordinates of the key points of the obstacle bounding box corresponding to the obstacle; An acquisition unit, configured to acquire the collision information according to the coordinates of the target key points and the coordinates of the key points of the obstacle; The coordinates of the target key points include the coordinates of the target corner points, and the coordinates of the key points of the obstacle include the coordinates of the obstacle corner points; specifically, the acquisition unit is configured to generate a first collision warning mark and use the first collision warning mark as the collision information if it is detected according to the coordinates of the target corner points and the coordinates of the obstacle corner points that the distance between the target corner point and the first target edge of the obstacle bounding box is less than a first distance threshold; wherein, the first collision warning mark is used to add the first collision warning mark to the coordinates of the target corner point.

11. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images