Method, apparatus, machine, medium, and vehicle for specifying true three-dimensional coordinates of a vehicle

Abstract

The present invention provides a method, apparatus, device, medium, and vehicle for specifying the true three-dimensional coordinates of a vehicle. 【Solution means】In this method, K groups of own vehicle position information collected at K stops by the inertial sensor of the auxiliary collection vehicle and K groups of main sensing data packets collected by the main collection vehicle are obtained. Based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Based on the coordinate transformation matrix and the own vehicle position information collected at a predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle is calculated, and the projected coordinate information is used as the true three-dimensional coordinates of the auxiliary collection vehicle, and the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold. According to the present invention, the true three-dimensional coordinates of a vehicle at an ultra-long distance can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of smart driving technology, and specifically to a method, apparatus, device, medium, and vehicle for identifying the true three-dimensional coordinates of a vehicle.

Background Art

[0002] During the driving of a driverless vehicle, it is necessary to detect the vehicles that appear on the road. Thereby, accidents such as collisions can be avoided, and the safety of driving can be guaranteed.

[0003] Specifically, a vehicle-end sensor, such as a lidar, may be attached to the driverless vehicle, and the surrounding environment during the driving of the driverless vehicle may be sensed by the lidar to obtain the true three-dimensional coordinates of each sensed vehicle.

[0004] The sensing distance of a normal lidar is only 60 meters to 120 meters. When the distance exceeds the upper limit of 120 meters, the lidar cannot obtain any three-dimensional sensing results. Therefore, currently, the true three-dimensional coordinates of a vehicle at an ultra-long distance cannot be obtained.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The present invention provides a method, apparatus, device, medium, and vehicle capable of obtaining the true three-dimensional coordinates of a vehicle at an ultra-long distance. The specific solutions are as follows.

Means for Solving the Problems

[0006] In a first aspect, an embodiment of the present invention provides a method for determining the true three-dimensional coordinates of a vehicle. In the method for determining the true three-dimensional coordinates of a vehicle, in an initial state, both the main collection vehicle and the auxiliary collection vehicle are located at an initial point, and the vehicle-end sensors of both the main collection vehicle and the auxiliary collection vehicle are both facing forward of the vehicle. When the auxiliary collection vehicle is traveling forward and the main collection vehicle is stationary, and every time the auxiliary collection vehicle stops after traveling at a predetermined distance interval, the vehicle-end sensors of both vehicles simultaneously collect sensing data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle continues to travel and stops at a predetermined distance position. At this time, the vehicle-end sensors of both vehicles simultaneously collect the sensing data packets for the predetermined time. The method for determining the true three-dimensional coordinates of a vehicle includes obtaining, by the inertial sensor of the auxiliary collection vehicle, K groups of own-vehicle position information collected at K stops and K groups of main sensing data packets collected by the main collection vehicle, where K is greater than a predetermined number, each group of main sensing data packets is composed of sensing data at the time stamps of a plurality of frames, the sensing data at the time stamp of each frame includes three-dimensional coordinate information of the sensing vehicle, and the sensing vehicle includes at least the auxiliary collection vehicle, calculating, based on the K groups of own-vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle, where the powered-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle at the time of power-on as the origin of the coordinate system, calculating, based on the coordinate transformation matrix and the own-vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle, and using the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, where the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold.

[0007] Optionally, the step of calculating a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the vehicle position information of the K groups and the main sensing data packets of the K groups is as follows: Taking the first group of main sensing data packets among the main sensing data packets of the K groups as the main sensing data packets of the current group, calculating a first relative distance error by calculating the first vehicle position information among the first group of vehicle position information over the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group according to the nearest neighbor method, specifying the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packets of the current group, and specifying the first vehicle position information as the vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main sensing data packets of the current group; For the time stamps of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, calculating a second relative distance error by calculating the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information, specifying the three-dimensional coordinate information of the sensing vehicle corresponding to a second relative distance error smaller than a predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the frame, calculating a third relative distance error by calculating the second three-dimensional coordinate information over the first group of vehicle position information according to the nearest neighbor method, and specifying the vehicle position information corresponding to the minimum third relative distance error as the vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the frame; Calculating a transformation matrix corresponding to the main sensing data packets of the current group based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of each frame, the vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of each frame, and a predetermined general diagram optimization formula, and using the transformation matrix as the current transformation matrix; Use the main perception data packet of the next group of the current group of main perception data packets as the main perception data packet of the current group, and based on the vehicle position information of the next group of the first group of vehicle position information and the current transformation matrix, calculate the first three-dimensional coordinate information of the auxiliary collection vehicle collected at the time stamp of the first frame in the main perception data packet of the current group by the main collection vehicle. And until the transformation matrix corresponding to the main perception data packet of the last group among the K groups of main perception data packets is obtained, for each time stamp of each frame other than the time stamp of the first frame in the main perception data packet of the current group, calculate the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain a second relative distance error, and then return to the execution of the step of obtaining the second relative distance error. The method further includes using the transformation matrix corresponding to the main perception data packet of the last group as the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle.

[0008] Optionally, the step of calculating the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle includes: Multiplying the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle from the left by the coordinate transformation matrix to obtain the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle.

[0009] Optionally, after the step of calculating the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the vehicle three-dimensional coordinate true value identification method further includes: Receive the two-dimensional marking box of the auxiliary collection vehicle manually marked on the target main perception data packet, perform two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information, and when the matching is successful, further execute the step of using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle. The target main perception data packet is the main perception data packet collected by the vehicle-end sensor of the main collection vehicle when the auxiliary collection vehicle continues to drive to a predetermined distance position and stops.

[0010] Optionally, the step of performing two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information includes: Projecting the three-dimensional boundary box corresponding to the projection coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projection boundary box; Calculating the first area of the intersection set and the second area of the union set between the two-dimensional marking box and the two-dimensional projection boundary box, calculating the division value of the first area and the second area, and when the division value is greater than a predetermined value, specifying that the matching is successful.

[0011] Optionally, after the step of using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, the above vehicle true three-dimensional coordinate determination method includes: Further including the step of receiving an adjustment to the vehicle height in the true three-dimensional coordinates of the auxiliary collection vehicle and obtaining the adjusted true three-dimensional coordinates of the auxiliary collection vehicle.

[0012] Optionally, after the step of using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, the above vehicle true three-dimensional coordinate determination method includes: Further including the step of calculating the true three-dimensional coordinates of each sensed vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary perception data packet collected at the predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle.

[0013] In a second aspect, an embodiment of the present invention provides a vehicle three-dimensional coordinate true value identification device. In an initial state, both the main collection vehicle and the auxiliary collection vehicle are located at an initial point, and the vehicle end sensors of both vehicles, namely the main collection vehicle and the auxiliary collection vehicle, are both facing forward of the vehicle. When the auxiliary collection vehicle is traveling forward and the main collection vehicle is stationary, and each time the auxiliary collection vehicle stops after traveling a predetermined distance interval, the vehicle end sensors of both vehicles simultaneously collect sensed data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle continues to travel and stops at a predetermined distance position. At this time, the vehicle end sensors of both vehicles simultaneously collect the sensed data packets for the predetermined time. The vehicle three-dimensional coordinate true value identification device includes an acquisition module, a coordinate transformation matrix identification module, and a first coordinate true value identification module. The acquisition module is configured to acquire K groups of own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at K stops and K groups of main sensed data packets collected by the main collection vehicle. Here, K is greater than a predetermined number. Each group of main sensed data packets is composed of sensed data at time stamps of a plurality of frames. The sensed data at the time stamp of each frame includes three-dimensional coordinate information of the sensed vehicle, and the sensed vehicle includes at least the auxiliary collection vehicle. The coordinate transformation matrix identification module is configured to calculate a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the K groups of own vehicle position information and the K groups of main sensed data packets. The powered-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle when powered on as the origin of the coordinate system. The first coordinate true value identification module is configured to calculate the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position, and set the projected coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle. The distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold.

[0014] Optionally, the coordinate transformation matrix identification module includes a first calculation sub-module, a second calculation sub-module, a current transformation matrix identification sub-module, and a third calculation sub-module. The first calculation sub-module uses the first group of main sensing data packets among the K groups of main sensing data packets as the main sensing data packets of the current group, and calculates the first relative distance error by calculating the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group with respect to the first vehicle position information among the first group of ego-vehicle position information according to the nearest neighbor method, identifies the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packets of the current group, and is configured to identify the first vehicle position information as the ego-vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main sensing data packets of the current group. The second calculation sub-module calculates the second relative distance error for the time stamps of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, with respect to the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information, identifies the three-dimensional coordinate information of the sensing vehicle corresponding to the second relative distance error smaller than the predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the frame, calculates the third relative distance error by calculating the second three-dimensional coordinate information with respect to the first group of ego-vehicle position information according to the nearest neighbor method, and is configured to identify the ego-vehicle position information corresponding to the minimum third relative distance error as the ego-vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the frame. The current transformation matrix identification sub-module calculates a transformation matrix corresponding to the main sensing data packet of the current group based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of each frame, the vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of each frame, and a predetermined general diagram optimization formula, and is configured to use the transformation matrix as the current transformation matrix. The third calculation sub-module uses the main sensing data packet of the next group of the main sensing data packet of the current group as the main sensing data packet of the current group, and based on the vehicle position information of the next group of the vehicle position information of the first group and the current transformation matrix, calculates the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packet of the current group. And until the transformation matrix corresponding to the main sensing data packet of the final group among the K groups of main sensing data packets is obtained, for the time stamps of each frame other than the time stamp of the first frame in the main sensing data packet of the current group, the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information are calculated to obtain a second relative distance error, and then returns to the execution of this step. The transformation matrix corresponding to the main sensing data packet of the final group is configured to be the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle.

[0015] Optionally, specifically, the first coordinate true value identification module multiplies the vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position from the left by the coordinate transformation matrix to obtain the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle.

[0016] Optionally, the above vehicle three-dimensional coordinate true value identification device further includes a receiving module. The receiving module calculates the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position. After that, it receives the two-dimensional marking box of the auxiliary collection vehicle manually marked in the target main sensing data packet, performs two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projected coordinate information, and is configured to trigger the projected coordinate information to be used as the true three-dimensional coordinates of the auxiliary collection vehicle when the matching is successful. The target main sensing data packet is the main sensing data packet collected by the vehicle end sensor of the main collection vehicle when the auxiliary collection vehicle continues to travel and stops at the predetermined distance position.

[0017] Optionally, the receiving module a projection sub-module configured to project the three-dimensional boundary box corresponding to the projected coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projected boundary box, a matching sub-module configured to calculate the first area of the intersection set and the second area of the union set between the two-dimensional marking box and the two-dimensional projected boundary box, calculate the division value of the first area and the second area, and identify that the matching is successful when the division value is greater than a predetermined value.

[0018] Optionally, the above vehicle true three-dimensional coordinate identification device further includes an adjustment module, The adjustment module is configured to receive an adjustment to the vehicle height in the true three-dimensional coordinates of the auxiliary collection vehicle after using the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, and obtain the adjusted true three-dimensional coordinates of the auxiliary collection vehicle.

[0019] Optionally, the above vehicle true three-dimensional coordinate identification device further includes a second true coordinate identification module, After using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, the second true coordinate specifying module is configured to calculate the true three-dimensional coordinates of each sensing vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensing vehicle in the auxiliary sensing data packet collected by the vehicle-end sensor of the auxiliary collection vehicle at the predetermined distance position.

[0020] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium, and when the program is executed by a processor, the vehicle three-dimensional true coordinate specifying method according to any embodiment of the present invention is implemented.

[0021] In a fourth aspect, an embodiment of the present invention provides an electronic device. The electronic device includes one or more processors, the processor is coupled to a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device is caused to implement the vehicle three-dimensional true coordinate specifying method according to any embodiment of the present invention.

[0022] In a fifth aspect, an embodiment of the present invention provides a vehicle. The vehicle includes the vehicle three-dimensional true coordinate specifying device according to any embodiment of the present invention, or the electronic device according to any embodiment of the present invention.

[0023] As can be seen from the above content, in the vehicle three-dimensional coordinate true value determination method, apparatus, device, medium and vehicle according to the embodiments of the present invention, it is possible to obtain K groups of own vehicle position information collected at K times of parking by the inertial sensor of the auxiliary collection vehicle and K groups of main sensing data packets collected by the main collection vehicle. Here, K is greater than a predetermined number of times, each group of main sensing data packets is composed of sensing data at the time stamps of multiple frames, the sensing data at the time stamp of each frame includes the three-dimensional coordinate information of the sensing vehicle, the sensing vehicle includes at least the auxiliary collection vehicle, and based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Here, the power-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle when powered on as the origin of the coordinate system. And based on the coordinate transformation matrix and the own vehicle position information collected at a predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle is calculated, and the projected coordinate information is used as the three-dimensional coordinate true value of the auxiliary collection vehicle. Here, the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold. In the embodiments of the present invention, through the cooperation between the main collection vehicle and the auxiliary collection vehicle, based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Then, based on the coordinate transformation matrix, the own vehicle position information of the auxiliary collection vehicle is projected onto the vehicle coordinate system of the main collection vehicle to obtain the three-dimensional coordinate true value of the auxiliary collection vehicle. Since the distance between the predetermined distance position and the main collection vehicle is greater than the predetermined distance threshold, the obtained three-dimensional coordinate true value of the auxiliary collection vehicle is a three-dimensional coordinate true value at an ultra-long distance, realizing the sensing of the three-dimensional coordinate true value of the vehicle at an ultra-long distance of the main collection vehicle. Of course, to implement any product or method of the present invention, it is not necessary to simultaneously have all the above-mentioned advantages.

Effects of the Invention

[0024] The innovative points of the embodiments of the present invention include the following aspects. 1. Through the cooperation between the main collection vehicle and the auxiliary collection vehicle, based on the self-vehicle position information of group K and the main sensing data packet of group K, calculate the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle. Then, project the self-vehicle position information of the auxiliary collection vehicle onto the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix to obtain the true three-dimensional coordinates of the auxiliary collection vehicle. Since the distance between the predetermined distance position and the main collection vehicle is greater than the predetermined distance threshold, the obtained true three-dimensional coordinates of the auxiliary collection vehicle are the true three-dimensional coordinates of a vehicle at an ultra-long distance, realizing the perception of the true three-dimensional coordinates of the vehicle at an ultra-long distance by the main collection vehicle. 2. In the embodiment of the present invention, only through the cooperation between the main collection vehicle and the auxiliary collection vehicle, without adding sensors, the true three-dimensional coordinates of a vehicle at an ultra-long distance can be accurately obtained. 3. Based on the self-vehicle position information of group K, the main sensing data packet of group K, and the nearest neighbor method, it is possible to calculate the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle. 4. By adjusting the vehicle height in the true three-dimensional coordinates of the auxiliary collection vehicle, the accuracy of the adjusted true three-dimensional coordinates of the auxiliary collection vehicle in the height direction is improved. 5. Through the coordinate transformation method, convert the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected by the auxiliary collection vehicle at a predetermined distance position into that of the vehicle coordinate system of the main collection vehicle, and obtain the true three-dimensional coordinates of the three-dimensional coordinate information of each sensed vehicle sensed by the auxiliary collection vehicle at an ultra-long distance in the vehicle coordinate system of the main collection vehicle. To more clearly explain the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can obtain other drawings from these drawings without spending creative labor.

Brief Description of the Drawings

[0025]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0026] Hereinafter, with reference to the drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and completely. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art without creative efforts based on the embodiments of the present invention are included in the protection scope of the present invention.

[0027] It should be noted that the terms "including", "comprising" or any modification thereof in the embodiments of the present invention and the drawings are intended to cover non-exclusive inclusion. For example, a series of steps or means in a process, method, system, product or device included are not limited to the listed steps or means, and preferably include steps or means not listed, or preferably include other steps or means inherent to these processes, methods, products or devices.

[0028] The embodiments of the present invention provide a vehicle three-dimensional coordinate true value determination method, device, equipment, medium and vehicle, and can obtain the three-dimensional coordinate true value of a vehicle at ultra-long distance. Hereinafter, the embodiments of the present invention will be described in detail.

[0029] FIG. 1 is a schematic flowchart of a vehicle three-dimensional coordinate true value determination method according to an embodiment of the present invention. The method is applied to an electronic device.

[0030] Figure 2 is a schematic diagram showing that the vehicle-end sensors of the main collection vehicle 1 and the auxiliary collection vehicle 2 according to the embodiment of the present invention collect sensed data packets respectively. Referring to Figure 2, in the initial state, both the main collection vehicle 1 and the auxiliary collection vehicle 2 are located at the initial point, and the vehicle-end sensors of both vehicles, namely the main collection vehicle and the auxiliary collection vehicle, are all directed forward of the vehicle, that is, the direction where the arrow is located in Figure 2. When the auxiliary collection vehicle 2 is traveling forward, that is, when traveling along the arrow direction, the main collection vehicle 1 remains stationary.

[0031] Every time the auxiliary collection vehicle 2 travels a predetermined distance interval h and stops, the vehicle-end sensors of both vehicles simultaneously collect sensed data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle 2 continues to travel and stops at a predetermined distance position, point A. At this time, the vehicle-end sensors of both vehicles simultaneously collect sensed data packets for a predetermined time. And when the auxiliary collection vehicle 2 stops at point A, there is another vehicle 3 passing through point A, and the main collection vehicle 1 does not apply the P range when stationary.

[0032] Based on this, the above method specifically includes the following steps S110 to S130.

[0033] In S110, K groups of own vehicle position information collected at K stops by the inertial sensor of the auxiliary collection vehicle and K groups of main sensed data packets collected by the main collection vehicle are obtained. K is greater than a predetermined number, and each group of main sensed data packets is composed of sensed data with multiple frames of timestamps. The sensed data with each frame of timestamp includes the three-dimensional coordinate information of the sensed vehicle, and the sensed vehicle includes at least the auxiliary collection vehicle.

[0034] While the auxiliary collection vehicle 2 is in motion, its inertial sensor collects the ego pose of the auxiliary collection vehicle 2, i.e., its own vehicle position information, in real time. Each time the auxiliary collection vehicle 2 travels a predetermined distance interval h and stops, and the vehicle-end sensor of the auxiliary collection vehicle 2 collects the sensed data packets for a predetermined time, the inertial sensor of the auxiliary collection vehicle 2 can collect a group of its own vehicle position information corresponding to a predetermined time each time it stops. In this way, K groups of its own vehicle position information are obtained after K stops.

[0035] When the auxiliary collection vehicle 2 stops each time it travels a predetermined distance interval h, the vehicle-end sensors of both vehicles collect the sensed data packets for a predetermined time simultaneously. Although the main collection vehicle 1 remains stationary without moving, the main collection vehicle 1 obtains K groups of main sensed data packets during K stops. Here, each group of main sensed data packets is composed of sensed data with time stamps of multiple frames. The sensed data with the time stamp of each frame includes the three-dimensional coordinate information of the sensed vehicle. K is greater than a predetermined number. By way of example, the predetermined number is 3.

[0036] While the auxiliary collection vehicle 2 travels forward and the main collection vehicle 1 remains stationary, the vehicle-end sensor of the main collection vehicle 1 can always sense the auxiliary collection vehicle 2. Therefore, the sensed vehicles sensed by the main collection vehicle 1 include at least the auxiliary collection vehicle.

[0037] In order to obtain the true three-dimensional coordinates of a vehicle at an ultra-long distance, the electronic device obtains the K groups of its own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle during K stops and the K groups of main sensed data packets collected by the main collection vehicle.

[0038] Here, the main collection vehicle 1 is the vehicle for which the true three-dimensional coordinates of a vehicle at an ultra-long distance should be obtained, and the auxiliary collection vehicle 2 is the vehicle that supports the main collection vehicle 1 to obtain the true three-dimensional coordinates of a vehicle at an ultra-long distance.

[0039] In S120, based on the own vehicle position information of group K and the main sensing data packet of group K, a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle is calculated. The powered-on vehicle coordinate system is a vehicle coordinate system with the position at the time when the in-vehicle computer of the vehicle is powered on as the origin of the coordinate system.

[0040] After the own vehicle position information of group K and the main sensing data packet of group K are acquired, the electronic device needs to calculate a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle 1 and the powered-on vehicle coordinate system of the auxiliary collection vehicle 2 based on the own vehicle position information of group K and the main sensing data packet of group K. The powered-on vehicle coordinate system is a vehicle coordinate system with the position at the time when the in-vehicle computer of the vehicle is powered on as the origin of the coordinate system.

[0041] Specifically, step S120 is as follows: Regarding the first group of main sensing data packets among the main sensing data packets of group K as the main sensing data packets of the current group, perform calculations with the first own vehicle position information among the own vehicle position information of the first group over the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group according to the nearest neighbor method to obtain the first relative distance error, specify the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packets of the current group, and specify the first own vehicle position information as the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main sensing data packets of the current group; For each frame timestamp in the current group's main perception data packet other than the timestamp of the first frame, calculate the second relative distance error for the three-dimensional coordinate information of the perception vehicle corresponding to the timestamp of the frame and the first three-dimensional coordinate information, and identify the three-dimensional coordinate information of the perception vehicle corresponding to the second relative distance error smaller than the predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the timestamp of the frame. Then, calculate the third relative distance error for the second three-dimensional coordinate information over the ego vehicle position information of the first group according to the nearest neighbor method, and identify the ego vehicle position information corresponding to the minimum third relative distance error as the ego vehicle position information of the auxiliary collection vehicle corresponding to the timestamp of the frame. Based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the timestamp of each frame, the ego vehicle position information of the auxiliary collection vehicle corresponding to the timestamp of each frame, and a predetermined general diagram optimization formula, calculate the transformation matrix corresponding to the current group's main perception data packet, and set the transformation matrix as the current transformation matrix. Set the next group's main perception data packet of the current group's main perception data packet as the current group's main perception data packet. Based on the next group's ego vehicle position information of the first group and the current transformation matrix, calculate the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the timestamp of the first frame in the current group's main perception data packet. Then, until the transformation matrix corresponding to the last group's main perception data packet among the K groups of main perception data packets is obtained, return to execute the step of calculating the second relative distance error for the three-dimensional coordinate information of the perception vehicle corresponding to the timestamp of each frame other than the timestamp of the first frame and the first three-dimensional coordinate information in the current group's main perception data packet. Set the transformation matrix corresponding to the last group's main perception data packet as the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle. This may include the above steps.

[0042] When the main collection vehicle 1 collects the first group of main sensing data packets, the distance from the auxiliary collection vehicle 2 is relatively close. At this time, in order to obtain the three-dimensional coordinate information of the auxiliary collection vehicle 2 in the main collection vehicle 1 at that time, the first group of main sensing data packets among the K groups of main sensing data packets are used as the main sensing data packets of the current group, and according to the nearest neighbor method, for the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group, calculations are performed with the first vehicle position information among the first group of own vehicle position information to obtain the first relative distance error, and the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error is specified by the main collection vehicle 1 as the first three-dimensional coordinate information of the auxiliary collection vehicle 2 collected at the time stamp of the first frame in the main sensing data packets of the current group. Here, the first group of own vehicle position information is the first group of own vehicle position information ranked first among the K groups of own vehicle position information.

[0043] That is, the three-dimensional coordinate information corresponding to the minimum first relative distance error is the three-dimensional coordinate information collected by the auxiliary collection vehicle 2 at the time stamp of the first frame in the main collection vehicle 1. Then, the first vehicle position information is specified as the vehicle position information of the auxiliary collection vehicle 2 corresponding to the time stamp of the first frame in the main sensing data packets of the current group.

[0044] Currently, only the three-dimensional coordinate information of the auxiliary collection vehicle 2 at the time stamp of the first frame in the first group of main sensing data packets has been obtained. In order to obtain the three-dimensional coordinate information of the auxiliary collection vehicle 2 at other time stamps, for each time stamp of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, calculations are performed on the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain the second relative distance error, and it is necessary to specify the three-dimensional coordinate information of the sensing vehicle corresponding to the second relative distance error smaller than the predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected at the time stamp of the frame by the main collection vehicle. Thereby, the three-dimensional coordinate information of the auxiliary collection vehicle 2 at other time stamps in the first group of main sensing data packets is obtained. As an example, the predetermined distance error threshold may be 1 m.

[0045] Thereafter, according to the nearest neighbor method, calculations are performed with respect to the vehicle position information of the first group and the second three-dimensional coordinate information to obtain a third relative distance error, and the vehicle position information corresponding to the minimum third relative distance error is specified as the vehicle position information of the auxiliary collection vehicle 2 corresponding to the time stamp of the frame. Finally, the three-dimensional coordinate information of the auxiliary collection vehicle 2 collected by the main collection vehicle 1 at the time stamp of each frame and the vehicle position information of the auxiliary collection vehicle 2 corresponding to the time stamp of each frame are obtained.

[0046] After the three-dimensional coordinate information of the auxiliary collection vehicle 2 collected by the main collection vehicle 1 at the time stamp of each frame and the vehicle position information of the auxiliary collection vehicle 2 corresponding to the time stamp of each frame are obtained, based on the three-dimensional coordinate information of the auxiliary collection vehicle 2 collected by the main collection vehicle 1 at the time stamp of each frame, the vehicle position information of the auxiliary collection vehicle 2 corresponding to the time stamp of each frame, and a predetermined general diagram optimization formula, a transformation matrix corresponding to the main sensing data packet of the current group is calculated, and the transformation matrix is used as the current transformation matrix. As an example, the current transformation matrix is a 4x4 transformation matrix.

[0047] In this way, the processing of the main sensing data packets of the first group is completed, and a current transformation matrix representing the transformation relationship between the three-dimensional coordinate information of the auxiliary collection vehicle 2 collected by the main collection vehicle 1 and the vehicle position information of the auxiliary collection vehicle 2 is obtained.

[0048] The three-dimensional coordinate information of the auxiliary collection vehicle 2 and the vehicle position information of the auxiliary collection vehicle 2 in the vehicle coordinate system of the main collection vehicle 1 satisfy the following mathematical formula.

[0049]

Equation

[0050] However, B is the vehicle coordinate system of the auxiliary collection vehicle, A is the vehicle coordinate system of the main collection vehicle, argmin is the numerical value of the minimum value that satisfies the formula, i is the count point, b is vehicle B, exp is the exponential mapping, and || ||2 is the 2-norm.

[0051] The following is to process the main sensing data packets of other groups except the first group of main sensing data packets in the main sensing data packets of the K groups, set the main sensing data packets of the next group of the current group of main sensing data packets as the main sensing data packets of the current group, and when the main collection vehicle 1 is collecting the main sensing data packets of the second group, since the distance from the auxiliary collection vehicle 2 is relatively far, there is an error between the vehicle position information of the second group and the three-dimensional coordinate information of the auxiliary collection vehicle 2 in the main collection vehicle 1. Therefore, it is necessary to substitute the vehicle position information of the second group using the calculated current transformation matrix to obtain the three-dimensional coordinate information of the auxiliary collection vehicle 2 in the main collection vehicle 1 at that time.

[0052] Specifically, based on the vehicle position information of the next group of the first group of vehicle position information and the current transformation matrix, the first three-dimensional coordinate information of the auxiliary collection vehicle 2 collected at the time stamp of the first frame in the main sensing data packets of the current group by the main collection vehicle 1 is calculated, and until the transformation matrix corresponding to the main sensing data packets of the final group among the main sensing data packets of the K groups is obtained, for each time stamp of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, calculations are performed on the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain the second relative distance error, and then return to the execution of the step of obtaining the second relative distance error, and set the transformation matrix corresponding to the main sensing data packets of the final group as the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle 1 and the powered-on vehicle coordinate system of the auxiliary collection vehicle 2.

[0053] Thereby, based on the vehicle position information of the K groups, the main sensing data packets of the K groups and the nearest neighbor method, it is possible to calculate the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle.

[0054] In S130, based on the coordinate transformation matrix and the vehicle position information of the host collection vehicle collected at a predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the host collection vehicle is calculated, and the projected coordinate information is used as the true three-dimensional coordinates of the auxiliary collection vehicle. The distance between the predetermined distance position and the host collection vehicle is greater than a predetermined distance threshold.

[0055] When the auxiliary collection vehicle 2 is traveling forward, since the host collection vehicle 1 remains stationary all the time, the vehicle coordinate system of the host collection vehicle 1 becomes the powered-on vehicle coordinate system. Therefore, the coordinate transformation matrix may be considered as the transformation matrix of the powered-on vehicle coordinate system of the auxiliary collection vehicle 2 with respect to the powered-on vehicle coordinate system of the host collection vehicle 1. Thus, at that time, at the predetermined distance position, the three-dimensional coordinate information of the auxiliary collection vehicle 2 in the host collection vehicle 1 becomes the coordinate information of the auxiliary collection vehicle 2 in the powered-on vehicle coordinate system of the host collection vehicle 1.

[0056] Therefore, after the coordinate transformation matrix is obtained, based on the coordinate transformation matrix and the vehicle position information of the auxiliary collection vehicle 2 collected at the predetermined distance position A by the inertial sensor, the projected coordinate information of the auxiliary collection vehicle 2 in the vehicle coordinate system of the host collection vehicle 1 can be obtained, and the projected coordinate information is used as the true three-dimensional coordinates of the auxiliary collection vehicle 2. Here, the distance between the predetermined distance position and the host collection vehicle 1 is greater than the predetermined distance threshold. As an example, the predetermined distance threshold is 300 m.

[0057] The fact that the distance between the predetermined distance position and the host collection vehicle 1 is greater than the predetermined distance threshold means that the predetermined distance position is far away from the host collection vehicle 1. Therefore, the obtained true three-dimensional coordinates of the auxiliary collection vehicle 2 are true three-dimensional coordinates at an ultra-long distance.

[0058] Specifically, calculating the projected coordinate information of the auxiliary collection vehicle 2 in the vehicle coordinate system of the host collection vehicle 1 based on the coordinate transformation matrix and the vehicle position information of the auxiliary collection vehicle 2 collected at the predetermined distance position by the inertial sensor may include multiplying the coordinate transformation matrix on the left by the vehicle position information of the auxiliary collection vehicle 2 collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle 2 to obtain the projected coordinate information of the auxiliary collection vehicle 2 in the vehicle coordinate system of the host collection vehicle 1.

[0059] As can be seen from the above content, in the embodiment of the present invention, it is possible to obtain K groups of own vehicle position information collected at K stops by the inertial sensor of the auxiliary collection vehicle and K groups of main sensing data packets collected by the main collection vehicle. Here, K is greater than a predetermined number of times, and each group of main sensing data packets is composed of sensing data with time stamps of multiple frames. The sensing data with the time stamp of each frame includes the three-dimensional coordinate information of the sensing vehicle. The sensing vehicle includes at least the auxiliary collection vehicle. And based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Here, the power-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle when powered on as the origin of the coordinate system. And based on the coordinate transformation matrix and the own vehicle position information collected at a predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle is calculated, and the projected coordinate information is used as the true three-dimensional coordinate of the auxiliary collection vehicle. Here, the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold. In the embodiment of the present invention, through the cooperation between the main collection vehicle and the auxiliary collection vehicle, based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Then, based on the coordinate transformation matrix, the own vehicle position information of the auxiliary collection vehicle is projected onto the vehicle coordinate system of the main collection vehicle to obtain the true three-dimensional coordinate of the auxiliary collection vehicle. Since the distance between the predetermined distance position and the main collection vehicle 1 is greater than the predetermined distance threshold, the obtained true three-dimensional coordinate of the auxiliary collection vehicle 2 is the true three-dimensional coordinate at an ultra-long distance, realizing the sensing of the true three-dimensional coordinate of the vehicle at an ultra-long distance by the main collection vehicle.

[0060] Also, in the embodiment of the present invention, only through the cooperation between the main collection vehicle and the auxiliary collection vehicle, without adding sensors, it is possible to accurately obtain the true three-dimensional coordinate of the vehicle at an ultra-long distance.

[0061] In one implementation manner, after step S130, the vehicle three-dimensional coordinate true value determination method may further include a step of receiving an adjustment to the vehicle height in the three-dimensional coordinate true value of the auxiliary collection vehicle and obtaining the three-dimensional coordinate true value of the adjusted auxiliary collection vehicle.

[0062] Since the road surface does not have no height at all, the obtained three-dimensional coordinate true value of the auxiliary collection vehicle 2 has a certain offset in the height direction, and it is necessary to manually adjust the height. The electronic device receives an adjustment to the vehicle height in the three-dimensional coordinate true value of the auxiliary collection vehicle 2 and obtains the three-dimensional coordinate true value of the adjusted auxiliary collection vehicle 2.

[0063] Thereby, by receiving an adjustment to the vehicle height in the three-dimensional coordinate true value of the auxiliary collection vehicle, the accuracy of the adjusted three-dimensional coordinate true value of the auxiliary collection vehicle in the height direction is improved.

[0064] In another implementation manner, after step S130, the vehicle three-dimensional coordinate true value determination method may further include a step of calculating the three-dimensional coordinate true value of each sensed vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle.

[0065] As described above, when the projection coordinate information is used as the three-dimensional coordinate true value of the auxiliary collection vehicle 2, only the three-dimensional coordinate true value of the auxiliary collection vehicle 2 sensed at a very long distance by the main collection vehicle 1 is obtained. Furthermore, the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the auxiliary collection vehicle 2 may all be converted to that in the vehicle coordinate system of the main collection vehicle 1. That is, based on the projection coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle, the three-dimensional coordinate true value of each sensed vehicle in the vehicle coordinate system of the main collection vehicle is calculated.

[0066] Specifically, calculating the true three-dimensional coordinates of each sensed vehicle in the vehicle coordinate system of the main collection vehicle based on the projected coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle is It may include multiplying the projected coordinate information from the left by the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle to obtain the true three-dimensional coordinates of each sensed vehicle in the vehicle coordinate system of the main collection vehicle.

[0067] In this way, by means of the coordinate transformation method, all the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at a predetermined distance position by the auxiliary collection vehicle 2 is converted into that in the vehicle coordinate system of the main collection vehicle 1, and the true three-dimensional coordinates of the three-dimensional coordinate information of each sensed vehicle sensed by the auxiliary collection vehicle 2 at an ultra-long distance in the vehicle coordinate system of the main collection vehicle 1 are obtained.

[0068] In another implementation method, after obtaining the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at a predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the above vehicle true three-dimensional coordinate determination method may further include the step of receiving the two-dimensional marking box of the auxiliary collection vehicle manually marked in the target main sensing data packet, performing two-dimensional-three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projected coordinate information, and when the matching is successful, executing the step of using the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, where the target main sensing data packet is the main sensing data packet collected by the vehicle-end sensor of the main collection vehicle when the auxiliary collection vehicle continues to travel and stops at a predetermined distance position.

[0069] Since the obtained projection coordinate information is not necessarily accurate, it is necessary to determine whether the projection coordinate information is accurate. Specifically, receive the two-dimensional marking box of the auxiliary collection vehicle 2 manually marked in the target main sensing data packet, perform two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information, and when the matching is successful, it means that the obtained projection coordinate information is accurate. At that time, execute the step of using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle. Here, the target main sensing data packet is the main sensing data packet collected by the vehicle-end sensor of the main collection vehicle 1 when the auxiliary collection vehicle 2 continues to travel to the parking position A at a predetermined distance.

[0070] Specifically, performing two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information means projecting the three-dimensional boundary box corresponding to the projection coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projected boundary box, calculating the first area of the intersection set and the second area of the union set between the two-dimensional marking box and the two-dimensional projected boundary box, calculating the division value of the first area and the second area, and when the division value is greater than a predetermined value, specifying that the matching is successful.

[0071] The method of performing two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information is to lower the three-dimensional to two-dimensional and then perform matching with the two-dimensional marking box, that is, project the three-dimensional boundary box corresponding to the projection coordinate information onto the image plane where the two-dimensional marking box is located, obtain the two-dimensional projected boundary box, calculate the first area of the intersection set and the second area of the union set between the two-dimensional marking box and the two-dimensional projected boundary box, calculate the division value of the first area and the second area, and when the division value is greater than a predetermined value, it means that the two are similar and specify that the matching is successful.

[0072] In this way, it is determined whether or not the two-dimensional marking box matches the three-dimensional boundary box corresponding to the projection coordinate information by means of two-dimensional-three-dimensional matching.

[0073] FIG. 3 is a structural schematic diagram of a vehicle three-dimensional coordinate true value specifying device according to an embodiment of the present invention. Referring to FIG. 3, in the initial state, both the main collection vehicle and the auxiliary collection vehicle are located at the initial point, and the vehicle end sensors of both vehicles, namely the main collection vehicle and the auxiliary collection vehicle, both face forward of the vehicle. When the auxiliary collection vehicle is traveling forward and the main collection vehicle is stationary, and when the auxiliary collection vehicle stops every time it travels a predetermined distance interval, the vehicle end sensors of both vehicles simultaneously collect the sensed data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle continues to travel and stops at a predetermined distance position. At this time, the vehicle end sensors of both vehicles simultaneously collect the sensed data packets for the predetermined time. The vehicle three-dimensional coordinate true value specifying device may include an acquisition module 310, a coordinate transformation matrix specifying module 320, and a first coordinate true value specifying module 330.

[0074] The acquisition module 310 is configured to acquire K groups of own vehicle position information collected at the time of K stops by the inertial sensor of the auxiliary collection vehicle and K groups of main sensed data packets collected by the main collection vehicle. K is greater than a predetermined number. Each group of main sensed data packets is composed of sensed data at the time stamps of a plurality of frames. The sensed data at the time stamp of each frame includes three-dimensional coordinate information of the sensed vehicle, and the sensed vehicle includes at least the auxiliary collection vehicle.

[0075] The coordinate transformation matrix specifying module 320 is configured to calculate a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the K groups of own vehicle position information and the K groups of main sensed data packets. The powered-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle at the time of power-on as the origin of the coordinate system.

[0076] The first coordinate true value specifying module 330 calculates the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information of the own vehicle collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle, and is configured to use the projected coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle. The distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold value.

[0077] The device according to an embodiment of the present invention can acquire K groups of own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at K times of parking and K groups of main sensing data packets collected by the main collection vehicle. Here, K is greater than a predetermined number of times. Each group of main sensing data packets is composed of sensing data at the time stamps of a plurality of frames. The sensing data at the time stamp of each frame includes the three-dimensional coordinate information of the sensing vehicle. The sensing vehicle includes at least the auxiliary collection vehicle. Based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Here, the power-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle at the time of power-on as the origin of the coordinate system. Based on the coordinate transformation matrix and the own vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle is calculated, and the projected coordinate information is used as the three-dimensional coordinate true value of the auxiliary collection vehicle. Here, the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold value. In the embodiment of the present invention, through the cooperation between the main collection vehicle and the auxiliary collection vehicle, based on the K groups of own vehicle position information and the K groups of main sensing data packets, a coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle is calculated. Then, based on the coordinate transformation matrix, the own vehicle position information of the auxiliary collection vehicle is projected into the vehicle coordinate system of the main collection vehicle to obtain the three-dimensional coordinate true value of the auxiliary collection vehicle. Since the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold value, the obtained three-dimensional coordinate true value of the auxiliary collection vehicle is a three-dimensional coordinate true value at an ultra-long distance, realizing the sensing of the three-dimensional coordinate true value of the vehicle at an ultra-long distance by the main collection vehicle.

[0078] In one implementation manner, the coordinate transformation matrix determination module 320 may include a first calculation sub-module, a second calculation sub-module, a current transformation matrix determination sub-module, and a third calculation sub-module.

[0079] The first calculation sub-module uses the first group of main sensing data packets among the K groups of main sensing data packets as the main sensing data packets of the current group, and calculates the first relative distance error by calculating the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group with respect to the first vehicle position information among the first group of ego vehicle position information according to the nearest neighbor method, and determines the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packets of the current group, and is configured to determine the first vehicle position information as the ego vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main sensing data packets of the current group.

[0080] The second calculation sub-module calculates the second relative distance error for the time stamps of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, by calculating the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information, and determines the three-dimensional coordinate information of the sensing vehicle corresponding to the second relative distance error smaller than the predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the frame, and calculates the third relative distance error by calculating the second three-dimensional coordinate information with respect to the first group of ego vehicle position information according to the nearest neighbor method, and is configured to determine the ego vehicle position information corresponding to the minimum third relative distance error as the ego vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the frame.

[0081] The current transformation matrix determination sub-module is configured to calculate a transformation matrix corresponding to the main sensing data packet of the current group based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of each frame, the vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of each frame, and a predetermined general diagram optimization formula, and set the transformation matrix as the current transformation matrix.

[0082] The third calculation sub-module sets the main sensing data packet of the next group of the main sensing data packet of the current group as the main sensing data packet of the current group, and based on the vehicle position information of the next group of the vehicle position information of the first group and the current transformation matrix, calculates the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packet of the current group, and until the transformation matrix corresponding to the main sensing data packet of the final group among the K groups of main sensing data packets is obtained, for the time stamps of each frame other than the time stamp of the first frame in the main sensing data packet of the current group, returns to execute the step of obtaining the second relative distance error by performing calculations on the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information, and configures the transformation matrix corresponding to the main sensing data packet of the final group as the coordinate transformation matrix between the power-on vehicle coordinate system of the main collection vehicle and the power-on vehicle coordinate system of the auxiliary collection vehicle.

[0083] Optionally, the first coordinate true value determination module 330 specifically may multiply the vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position from the left by the coordinate transformation matrix to obtain the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle.

[0084] Optionally, the above vehicle three-dimensional coordinate true value determination device further includes a receiving module. The receiving module calculates the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position. After that, it receives the two-dimensional marking box of the auxiliary collection vehicle manually marked in the target main sensing data packet, performs two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projected coordinate information, and when the matching is successful, it is configured to trigger the projected coordinate information to be used as the true three-dimensional coordinates of the auxiliary collection vehicle. The target main sensing data packet is the main sensing data packet collected by the vehicle end sensor of the main collection vehicle when the auxiliary collection vehicle continues to travel and stops at the predetermined distance position.

[0085] Optionally, the receiving module may include a projection sub-module and a matching sub-module. The projection sub-module is configured to project the three-dimensional boundary box corresponding to the projected coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projected boundary box. The matching sub-module is configured to calculate the first area of the intersection set and the second area of the union set between the two-dimensional marking box and the two-dimensional projected boundary box, calculate the division value of the first area and the second area, and when the division value is greater than a predetermined value, identify that the matching is successful.

[0086] Optionally, the above vehicle true three-dimensional coordinate identification device may further include an adjustment module. The adjustment module is configured to receive an adjustment to the vehicle height in the true three-dimensional coordinates of the auxiliary collection vehicle after using the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, and obtain the adjusted true three-dimensional coordinates of the auxiliary collection vehicle.

[0087] Optionally, the above vehicle true three-dimensional coordinate identification device may further include a second true coordinate identification module. After using the projection coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, the second coordinate true value determination module is configured to calculate the true three-dimensional coordinates of each sensing vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensing vehicle in the auxiliary sensing data packet collected by the vehicle-end sensor of the auxiliary collection vehicle at the predetermined distance position.

[0088] The above device embodiment corresponds to the method embodiment and has the same technical effects as the method embodiment. For details, please refer to the method embodiment. The device embodiment is obtained based on the method embodiment. For specific descriptions, please refer to the part of the method embodiment and will not be repeated here.

[0089] FIG. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. As shown in FIG. 4, the electronic device includes one or more processors 410. The processor 410 is coupled to a storage device 420, and the storage device 420 stores one or more programs. When the one or more programs are executed by the one or more processors 410, the electronic device is caused to implement the vehicle three-dimensional coordinate true value determination method according to any embodiment of the present invention.

[0090] Based on the above embodiments, another embodiment of the present invention provides a vehicle. The vehicle includes the vehicle three-dimensional coordinate true value determination device according to any embodiment of the present invention, or the electronic device according to any embodiment of the present invention.

[0091] FIG. 5 is a schematic diagram of a vehicle according to an embodiment of the present invention. As shown in FIG. 5, the vehicle includes a speed sensor 51, an ECU (Electronic Control Unit) 52, a GPS (Global Positioning System) positioning device 53, and a T-Box (Telematics Box) 54. The speed sensor 51 measures the vehicle speed and uses the vehicle speed as the experienced speed for model training. The GPS positioning device 53 acquires the current geographical location of the vehicle. The T-Box 54 may communicate with the server as a gateway. The ECU 52 may execute the vehicle three-dimensional coordinate true value identification method described above.

[0092] Further, the vehicle may further include a V2X (Vehicle-to-Everything) module 55, a radar 56, and a camera 57. The V2X module 55 communicates with other vehicles, roadside devices, etc. The radar 56 or the camera 57 senses road environment information in the front and / or other directions and acquires original point cloud data. The radar 56 and / or the camera 57 may be disposed at the front and / or the rear of the vehicle body.

[0093] Based on the above method embodiment, another embodiment of the present invention provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium, and when the program is executed by a processor, the vehicle three-dimensional coordinate true value identification method according to any embodiment of the present invention is implemented.

[0094] As can be understood by those skilled in the art, the drawings are merely schematic diagrams of one embodiment, and the modules or flows in the drawings are not necessarily essential for implementing the present invention.

[0095] As can be understood by those skilled in the art, the modules in the device in the embodiments may be distributed in the devices of the embodiments in accordance with the description of the embodiments, or may be located in one or more devices that are appropriately changed and different from the present embodiments. The modules of the above embodiments may be integrated into one module, or may be further divided into a plurality of sub-modules.

[0096] Finally, it should be noted that the above embodiments are merely for explaining the technical solutions of the present invention and do not constitute a limitation thereto. Although the present invention has been described in detail with reference to the above embodiments, as can be understood by those skilled in the art, corrections can still be made to the technical solutions described in the above embodiments, or some of the technical features thereof can be replaced by equivalents. Furthermore, the gist of the corresponding technical solutions will not deviate from the spirit and scope of the technical solutions of each embodiment of the present invention due to these corrections or replacements.

Claims

1. A method for determining the true three-dimensional coordinates of a vehicle, comprising: In an initial state, both the main collection vehicle and the auxiliary collection vehicle are located at the initial point, and the vehicle-end sensors of both the main collection vehicle and the auxiliary collection vehicle are both facing forward of the vehicle. When the auxiliary collection vehicle is moving forward and the main collection vehicle is stationary, and each time the auxiliary collection vehicle stops after traveling at a predetermined distance interval, the vehicle-end sensors of both vehicles simultaneously collect the sensed data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle continues to travel and stops at a predetermined distance position. At this time, the vehicle-end sensors of both vehicles simultaneously collect the sensed data packets for the predetermined time. The method for determining the true three-dimensional coordinates of the vehicle further comprises: Obtaining K groups of own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at K stops and K groups of main sensed data packets collected by the main collection vehicle, where K is greater than a predetermined number, each group of main sensed data packets is composed of sensed data at the time stamps of multiple frames, the sensed data at the time stamp of each frame includes the three-dimensional coordinate information of the sensed vehicle, and the sensed vehicle includes at least the auxiliary collection vehicle. Calculating a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the K groups of own vehicle position information and the K groups of main sensed data packets, where the powered-on vehicle coordinate system is a vehicle coordinate system with the position of the in-vehicle computer of the vehicle when powered on as the origin of the coordinate system. Calculating the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position, and using the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle, where the distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold. A method for determining the true three-dimensional coordinates of a vehicle, characterized by comprising the above steps.

2. The step of calculating a coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the K groups of own vehicle position information and the K groups of main sensed data packets comprises: Regarding the first group of main sensing data packets among the main sensing data packets of the K group, use the main sensing data packets of the first group as the main sensing data packets of the current group. Calculate the first relative distance error by calculating the first vehicle position information among the first group of own vehicle position information with respect to the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of each frame in the main sensing data packets of the current group according to the nearest neighbor method. Identify the three-dimensional coordinate information of the sensing vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main sensing data packets of the current group, and identify the first vehicle position information as the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main sensing data packets of the current group. Regarding the time stamps of each frame other than the time stamp of the first frame in the main sensing data packets of the current group, calculate the second relative distance error by calculating the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information, and identify the three-dimensional coordinate information of the sensing vehicle corresponding to the second relative distance error smaller than the predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the frame. Calculate the third relative distance error by calculating the second three-dimensional coordinate information with respect to the first group of own vehicle position information according to the nearest neighbor method, and identify the own vehicle position information corresponding to the minimum third relative distance error as the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the frame. Based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of each frame, the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of each frame, and a predetermined general map optimization formula, calculate a transformation matrix corresponding to the main sensing data packets of the current group, and set the transformation matrix as the current transformation matrix. Use the primary sensing data packet of the next group of the current group of primary sensing data packets as the primary sensing data packet of the current group. Based on the vehicle position information of the next group of the first group of own vehicle position information and the current transformation matrix, calculate the first three-dimensional coordinate information of the auxiliary collection vehicle collected at the time stamp of the first frame in the primary sensing data packet of the current group by the primary collection vehicle. And until the transformation matrix corresponding to the primary sensing data packet of the last group among the K groups of primary sensing data packets is obtained, for the time stamps of each frame other than the time stamp of the first frame in the primary sensing data packet of the current group, perform calculations on the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain a second relative distance error, and return to the execution of the step of obtaining the second relative distance error. Set the transformation matrix corresponding to the primary sensing data packet of the last group as the coordinate transformation matrix between the powered-on vehicle coordinate system of the primary collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle. The method for specifying the true value of the vehicle three-dimensional coordinates according to claim 1, characterized by including the above steps.

3. The step of calculating the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the primary collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle is: The method for specifying the true value of the vehicle three-dimensional coordinates according to claim 1 or 2, characterized by including the step of multiplying the coordinate transformation matrix from the left by the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle to obtain the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the primary collection vehicle.

4. After the step of calculating the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the primary collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle, the method for specifying the true value of the vehicle three-dimensional coordinates is: Further include the step of receiving the two-dimensional marking box of the auxiliary collection vehicle manually marked on the target primary sensing data packet, performing two-dimensional-three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projected coordinate information, and when the matching is successful, setting the projected coordinate information as the true three-dimensional coordinates of the auxiliary collection vehicle. The vehicle three-dimensional coordinate true value determination method according to claim 1, wherein the target main sensing data packet is a main sensing data packet collected by a vehicle-end sensor of the main collection vehicle when the auxiliary collection vehicle continues to travel and stops at a predetermined distance position.

5. The step of performing two-dimensional to three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projection coordinate information includes: projecting the three-dimensional boundary box corresponding to the projection coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projected boundary box; calculating a first area of the intersection set and a second area of the union set between the two-dimensional marking box and the two-dimensional projected boundary box, calculating a division value of the first area and the second area, and when the division value is greater than a predetermined value, specifying that the matching is successful, the vehicle three-dimensional coordinate true value determination method according to claim 4.

6. After the step of using the projection coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle, the vehicle three-dimensional coordinate true value determination method further includes: receiving an adjustment to the vehicle height in the three-dimensional coordinate true value of the auxiliary collection vehicle, and obtaining the three-dimensional coordinate true value of the adjusted auxiliary collection vehicle, the vehicle three-dimensional coordinate true value determination method according to claim 1.

7. After the step of using the projection coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle, the vehicle three-dimensional coordinate true value determination method further includes: calculating the three-dimensional coordinate true value of each sensed vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at the predetermined distance position by the vehicle-end sensor of the auxiliary collection vehicle, the vehicle three-dimensional coordinate true value determination method according to claim 1.

8. A vehicle three-dimensional coordinate true value determination device, In the initial state, both the main collection vehicle and the auxiliary collection vehicle are located at the initial point, and the vehicle-end sensors of both the main collection vehicle and the auxiliary collection vehicle face forward. When the auxiliary collection vehicle is moving forward and the main collection vehicle is stationary, and the auxiliary collection vehicle stops every time it travels a predetermined distance interval, the vehicle-end sensors of both vehicles simultaneously collect the sensed data packets for a predetermined time. When the number of stops reaches K times, the auxiliary collection vehicle continues to travel and stops at a predetermined distance position. At that time, the vehicle-end sensors of both vehicles simultaneously collect the sensed data packets for the predetermined time. The vehicle three-dimensional coordinate true value identification device includes an acquisition module, a coordinate transformation matrix identification module, and a first coordinate true value identification module. The acquisition module is configured to acquire K groups of own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle during K stops and K groups of main sensed data packets collected by the main collection vehicle. K is greater than a predetermined number. Each group of main sensed data packets is composed of sensed data with time stamps of a plurality of frames. The sensed data with the time stamp of each frame includes the three-dimensional coordinate information of the sensed vehicle. The sensed vehicle includes at least the auxiliary collection vehicle. The coordinate transformation matrix identification module is configured to calculate the coordinate transformation matrix between the powered-on vehicle coordinate system of the main collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle based on the K groups of own vehicle position information and the K groups of main sensed data packets. The powered-on vehicle coordinate system is a vehicle coordinate system with the position of the vehicle-mounted computer when the vehicle is powered on as the origin of the coordinate system. The first coordinate true value identification module is configured to calculate the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the main collection vehicle based on the coordinate transformation matrix and the own vehicle position information collected by the inertial sensor of the auxiliary collection vehicle at the predetermined distance position, and use the projected coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle. The distance between the predetermined distance position and the main collection vehicle is greater than a predetermined distance threshold. A vehicle three-dimensional coordinate true value identification device characterized by this.

9. The coordinate transformation matrix identification module includes a first calculation sub-module, a second calculation sub-module, a current transformation matrix identification sub-module, and a third calculation sub-module. The first calculation sub-module designates the first group of main perception data packets among the main perception data packets of the K groups as the main perception data packets of the current group, and calculates the first vehicle position information among the first group of own vehicle position information with respect to the three-dimensional coordinate information of the perception vehicle corresponding to the time stamp of each frame in the main perception data packets of the current group according to the nearest neighbor method to obtain a first relative distance error, designates the three-dimensional coordinate information of the perception vehicle corresponding to the minimum first relative distance error as the first three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the first frame in the main perception data packets of the current group, and is configured to designate the first vehicle position information as the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the first frame in the main perception data packets of the current group. The second calculation sub-module calculates, for the time stamps of each frame other than the time stamp of the first frame in the main perception data packets of the current group, the three-dimensional coordinate information of the perception vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain a second relative distance error, designates the three-dimensional coordinate information of the perception vehicle corresponding to a second relative distance error smaller than a predetermined distance error threshold as the second three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of the frame, calculates the second three-dimensional coordinate information with respect to the first group of own vehicle position information according to the nearest neighbor method to obtain a third relative distance error, and is configured to designate the own vehicle position information corresponding to the minimum third relative distance error as the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of the frame. The current transformation matrix identification sub-module calculates a transformation matrix corresponding to the main perception data packets of the current group based on the three-dimensional coordinate information of the auxiliary collection vehicle collected by the main collection vehicle at the time stamp of each frame, the own vehicle position information of the auxiliary collection vehicle corresponding to the time stamp of each frame, and a predetermined general map optimization formula, and is configured to use the transformation matrix as the current transformation matrix. The third calculation sub-module uses the primary sensing data packet of the next group of the current group's primary sensing data packet as the primary sensing data packet of the current group, and based on the vehicle position information of the next group of the first group and the current transformation matrix, calculates the first three-dimensional coordinate information of the auxiliary collection vehicle collected at the time stamp of the first frame in the primary sensing data packet of the current group by the primary collection vehicle. And until the transformation matrix corresponding to the primary sensing data packet of the last group among the K groups of primary sensing data packets is obtained, for the time stamps of each frame other than the time stamp of the first frame in the primary sensing data packet of the current group, calculations are performed on the three-dimensional coordinate information of the sensing vehicle corresponding to the time stamp of the frame and the first three-dimensional coordinate information to obtain a second relative distance error, and then returns to the execution of the step of obtaining the second relative distance error. The vehicle three-dimensional coordinate true value identification device according to claim 8, wherein the transformation matrix corresponding to the primary sensing data packet of the last group is configured to be the coordinate transformation matrix between the powered-on vehicle coordinate system of the primary collection vehicle and the powered-on vehicle coordinate system of the auxiliary collection vehicle.

10. The first coordinate true value identification module multiplies the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle from the left by the coordinate transformation matrix to obtain the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the primary collection vehicle. The vehicle three-dimensional coordinate true value identification device according to claim 8 or 9, characterized in that.

11. The vehicle three-dimensional coordinate true value identification device further comprises a receiving module The receiving module calculates the projected coordinate information of the auxiliary collection vehicle in the vehicle coordinate system of the primary collection vehicle based on the coordinate transformation matrix and the vehicle position information collected at the predetermined distance position by the inertial sensor of the auxiliary collection vehicle, and then receives the two-dimensional marking box of the auxiliary collection vehicle manually marked on the target primary sensing data packet. Perform two-dimensional-three-dimensional matching on the two-dimensional marking box and the three-dimensional boundary box corresponding to the projected coordinate information, and when the matching is successful, trigger the projected coordinate information to be used as the three-dimensional coordinate true value of the auxiliary collection vehicle. The vehicle three-dimensional coordinate true value identification device according to claim 8, wherein the target main sensing data packet is the main sensing data packet collected by the vehicle end sensor of the main collection vehicle when the auxiliary collection vehicle continues to travel and stops at a predetermined distance position.

12. The receiving module A projection sub-module configured to project a three-dimensional bounding box corresponding to the projection coordinate information onto the image plane where the two-dimensional marking box is located to obtain a two-dimensional projected bounding box; The vehicle three-dimensional coordinate true value identification device according to claim 11, further comprising: a matching sub-module configured to calculate a first area of an intersection set and a second area of a union set between the two-dimensional marking box and the two-dimensional projected bounding box, calculate a division value of the first area and the second area, and when the division value is greater than a predetermined value, identify that the matching is successful.

13. The vehicle three-dimensional coordinate true value identification device further comprises an adjustment module The vehicle three-dimensional coordinate true value identification device according to claim 8, wherein the adjustment module is configured to receive adjustment with respect to the vehicle height in the three-dimensional coordinate true value of the auxiliary collection vehicle after using the projection coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle, and obtain the adjusted three-dimensional coordinate true value of the auxiliary collection vehicle.

14. The vehicle three-dimensional coordinate true value identification device further comprises a second coordinate true value identification module The vehicle three-dimensional coordinate true value identification device according to claim 1, wherein the second coordinate true value identification module is configured to calculate the three-dimensional coordinate true value of each sensed vehicle in the vehicle coordinate system of the main collection vehicle based on the projection coordinate information and the three-dimensional coordinate information of each sensed vehicle in the auxiliary sensing data packet collected at the predetermined distance position by the vehicle end sensor of the auxiliary collection vehicle after using the projection coordinate information as the three-dimensional coordinate true value of the auxiliary collection vehicle.

15. A computer-readable storage medium storing a computer program, The computer-readable storage medium, wherein when the computer program is executed by a processor, the vehicle three-dimensional coordinate true value identification method according to any one of claims 1 to 7 is implemented.

16. An electronic device, Comprising one or more processors, The processor is coupled to a storage device for storing one or more programs, An electronic device, characterized in that when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle three-dimensional coordinate true value identification method according to any one of claims 1 to 7.

17. A vehicle, characterized by comprising the vehicle three-dimensional coordinate true value identification device according to any one of claims 8 to 14, or the electronic device according to claim 16.

Images