Detection method and device for vehicle radar re-calibration and vehicle

Abstract

The invention relates to the technical field of radar calibration, and discloses a detection method and device for vehicle radar re-calibration and a vehicle, and the detection method for vehicle radar re-calibration comprises the steps: obtaining common view point cloud data, located in a common view area of a first radar and a second radar, of the vehicle; judging whether the common viewpoint cloud data meets a preset verification condition, wherein the verification condition comprises at least one of the following conditions: a point verification condition, a distance verification condition and an angle verification condition; if yes, whether the first radar or the second radar shifts or not is judged according to the common viewpoint cloud data, and if yes, the external parameters of the first radar or the second radar are re-calibrated. Based on the point cloud data of the radar with the common visual area, whether the current radar shifts or not can be automatically judged without manual intervention, so that the radar is re-calibrated in time, the navigation precision of a vehicle is ensured, and collision with an obstacle is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of radar calibration, and in particular to a detection method and device for vehicle radar recalibration, and a vehicle. Background Art

[0002] At present, in order to improve the perception ability of vehicles and prevent them from colliding with surrounding obstacles, radars are usually installed around the vehicles. Taking the scenario of using automated guided vehicles (AGVs) to replace manual cargo handling in complex working environments as an example, the navigation and positioning module of AGVs plays a very important role in their ability to operate automatically in complex working environments. The most common, most applicable and most convenient navigation method used by AGVs is local contour navigation, which uses the on-board laser radar sensor to perceive the contour of the environment and locate the vehicle by matching it with the map.

[0003] In the actual application of AGV, navigation accuracy can directly affect the docking accuracy during vehicle operations, such as semiconductors, new energy, automobiles, warehousing and other scenarios. Therefore, when the AGV leaves the factory, it is necessary to manually calibrate the navigation radar accurately, that is, to obtain the relative position (position + posture) of the radar relative to the vehicle body. However, in the operation scenarios of AGV, there are often complex situations such as mixed traffic of people and vehicles, and mixed traffic of vehicles. When the position or posture of the laser radar installed on the vehicle changes due to uncontrollable factors (for example: changes in mechanical stress, collisions with other vehicles or objects, problems with the radar itself, etc.), it will directly affect the positioning accuracy, resulting in the vehicle being unable to meet the docking accuracy requirements, resulting in serious situations such as positioning loss, material damage, and vehicle damage.

[0004] Therefore, after the vehicle leaves the factory, it is particularly important to evaluate in real time whether the radar is displaced and whether it can still meet the positioning requirements and normal operation indicators. The existing technology generally requires personnel to conduct regular inspections; technicians to recalibrate and maintain regularly, or analyze vehicle faults (such as the number of docking failures and the number of abnormal positioning losses) and then intervene manually to observe the operation, which is labor-intensive and has delays. Summary of the invention

[0005] In view of this, the present invention provides a vehicle radar recalibration detection method, device and vehicle to solve the problem that radar recalibration relies on manual labor and has lag.

[0006] In a first aspect, the present invention provides a vehicle radar recalibration detection method, the method comprising:

[0007] Acquire common view point cloud data on the vehicle in the common view area of ​​the first radar and the second radar;

[0008] Determine whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition;

[0009] If it meets the requirements, it is determined whether the first radar or the second radar is displaced according to the common view point cloud data. If it is displaced, the external parameters of the first radar or the second radar are recalibrated.

[0010] The vehicle radar recalibration detection method provided by the present invention obtains the common view point cloud data of the first radar and the second radar in the common view area on the vehicle, and determines whether the common view point cloud data meets the point verification conditions, distance verification conditions and angle verification conditions. If so, it determines whether the first radar or the second radar is displaced according to the common view point cloud data. If so, the external parameters of the first radar or the second radar are recalibrated. Based on the point cloud data of the radar with the common view area, the present invention can automatically determine whether the current radar is displaced without manual intervention, so as to recalibrate the radar in time, ensure the navigation accuracy of the vehicle, and avoid collision with obstacles.

[0011] In an optional embodiment, after obtaining the common view point cloud data of the vehicle in the common view area of ​​the first radar and the second radar, it also includes: obtaining the vehicle's rotational angular velocity, vehicle speed and angular velocity, and judging whether the vehicle is in a stationary state based on the rotational angular velocity, vehicle speed and angular velocity, and if so, executing the step of judging whether the common view point cloud data meets the preset verification conditions.

[0012] The present invention detects whether the radar needs to be recalibrated when the vehicle is stationary, thereby ensuring that the frequencies of the radars in the common view area are the same, the collected point cloud data are time synchronized, and there is no motion distortion, thereby avoiding interference.

[0013] In an optional embodiment, obtaining common view point cloud data of a first radar and a second radar on a vehicle in a common view area includes: obtaining first original data of the first radar and second original data of the second radar, the perception range of the first radar and the perception range of the second radar having at least one common view area; filtering out trailing points of the first original data and the second original data, and converting them into a vehicle body coordinate system to obtain first point cloud data and second point cloud data; filtering the first point cloud data and the second point cloud data according to the range of the common view area to obtain first common view point cloud data of the first radar within the common view area, and second common view point cloud data of the second radar within the common view area.

[0014] The present invention can ensure the accuracy of point cloud data by filtering out trailing points of point cloud data and avoid interference caused by the divergence characteristics of radar laser pulses. By simultaneously acquiring point cloud data of the common view area of ​​two radars, mutual calibration can be performed based on the two radars, and the relative position of the radars can be determined by recalibration to improve the navigation accuracy of the vehicle.

[0015] In an optional embodiment, determining whether the common view point cloud data meets preset verification conditions includes: clustering the first common view point cloud data, selecting the first cluster of point cloud data and the second cluster of point cloud data with the top two numbers of point cloud points, and clustering the second common view point cloud data, selecting the third cluster of point cloud data and the fourth cluster of point cloud data with the top two numbers of point cloud points; and determining whether the common view point cloud data meets the verification conditions based on the first cluster of point cloud data and the second cluster of point cloud data, the third cluster of point cloud data and the fourth cluster of point cloud data.

[0016] The present invention can determine whether the current common view area environment is suitable for detecting whether the radar is shifted by judging whether the obtained point cloud data meets the verification conditions, thereby ensuring the reliability and accuracy of the detection.

[0017] In an optional implementation, judging whether the common view point cloud data meets the point count verification condition based on the first cluster point cloud data and the second cluster point cloud data, the third cluster point cloud data and the fourth cluster point cloud data, includes: obtaining the first point count of the first cluster point cloud data, the second point count of the second cluster point cloud data, the third point count of the third cluster point cloud data and the fourth point count of the fourth cluster point cloud data; judging whether the first point count, the second point count, the third point count and the fourth point count are all greater than a preset quantity threshold, if they are all greater, the point count verification condition is met; judging whether the common view point cloud data meets the distance verification condition based on the first cluster point cloud data and the second cluster point cloud data, the third cluster point cloud data and the fourth cluster point cloud data, includes: calculating the first centroid of the first cluster point cloud data and the first centroid of the second cluster point cloud data; and determining whether the common view point cloud data meets the distance verification condition based on the first cluster point cloud data and the second cluster point cloud data. A first distance interval between the two centroids, and a second distance interval between the third centroid of the third cluster point cloud data and the fourth centroid of the fourth cluster point cloud data; judging whether the first distance interval and the second distance interval are both less than a preset distance threshold, if both are less than, the distance verification condition is met; judging whether the common view point cloud data meets the angle verification condition according to the first cluster point cloud data and the second cluster point cloud data, the third cluster point cloud data and the fourth cluster point cloud data, including: performing straight line fitting on the first cluster point cloud data and the third cluster point cloud data to obtain a first straight line, and performing straight line fitting on the second cluster point cloud data and the fourth cluster point cloud data to obtain a second straight line; judging whether the angle between the first straight line and the second straight line is less than a preset angle threshold, if so, the angle verification condition is met.

[0018] By judging the number of points in the point cloud data, the present invention can ensure that the current data is sufficient to support radar shift detection and ensure the reliability of the detection result; by judging the distance between the point cloud data of two radars, it can be determined that the two radars are currently detecting the same object, so the positions of the two radars can be calibrated with each other; by judging the angles of point clouds of different clusters, it can detect whether the radar is shifted based on point clouds in different directions, avoiding the situation where subsequent point cloud matching is degraded due to point clouds in one direction.

[0019] In an optional embodiment, it is determined whether the first radar or the second radar is displaced according to the common view point cloud data, and if so, the external parameters of the first radar or the second radar are recalibrated, including: using a preset point cloud registration algorithm to iteratively align a first total point cloud composed of a first cluster of point cloud data and a second cluster of point cloud data, and a second total point cloud composed of a third cluster of point cloud data and a fourth cluster of point cloud data until preset registration parameters are met; calculating the error between the current relative position and the calibrated relative position between the first total point cloud and the second total point cloud according to the registration result; determining whether the relative position between the first radar and the second radar is displaced according to the error and a preset error threshold, and if so, adjusting the external parameters of the first radar or the second radar according to the error.

[0020] By matching the point cloud data of two radars, the present invention can determine whether the relative positions of the two radars have changed, thereby detecting whether the radars have shifted. If so, the external parameters of the radars are adjusted so that the relative positions of the radars correspond to the external parameters, thereby ensuring navigation accuracy.

[0021] In an optional embodiment, it is determined whether the relative position between the first radar and the second radar is displaced according to the error and the preset error threshold, and if displaced, the external parameters of the first radar or the second radar are adjusted according to the error, including: if the error is less than the preset error threshold, it is determined that the first radar and / or the second radar are not displaced, the error includes: lateral distance error, longitudinal distance error and angle error, and the preset error threshold includes a first preset distance error threshold, a second preset distance error threshold and a preset angle error threshold; if the error is greater than the preset error threshold and less than a preset multiple of the preset error threshold, it is determined that the first radar and / or the second radar is displaced, a warning signal is generated, and after the number of consecutive times of being met exceeds a preset number, the external parameters of the first radar or the second radar are adjusted according to the lateral distance error, the longitudinal distance error and / or the angle error; if the error is greater than the preset multiple of the preset error threshold, it is determined that a fault occurs, and a fault signal is generated.

[0022] The present invention can automatically adjust the displacement of the radar in a timely manner to a certain extent through three-level processing, and send a warning signal to the staff. In order to avoid the situation where the excessive displacement cannot be adjusted, the present invention can send a fault signal to remind the staff to intervene manually, thereby reducing the vehicle failure rate and recovering unnecessary economic losses.

[0023] In a second aspect, the present invention provides a detection device for vehicle radar recalibration, the device comprising:

[0024] A data acquisition module, used to acquire common view point cloud data on the vehicle in a common view area of ​​the first radar and the second radar;

[0025] A condition verification module, used to determine whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition;

[0026] The displacement determination module is used to determine whether the first radar or the second radar is displaced according to the common view point cloud data if it meets the requirements, and if so, recalibrate the external parameters of the first radar or the second radar.

[0027] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle radar recalibration detection method of the above-mentioned first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0028] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the vehicle radar recalibration detection method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0029] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the vehicle radar recalibration detection method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0030] In a sixth aspect, the present invention provides a vehicle, comprising the above-mentioned computer device. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 is a flow chart of a detection method for vehicle radar recalibration according to an embodiment of the present invention;

[0033] Figure 2 is a radar schematic diagram of a vehicle radar recalibration detection method according to an embodiment of the present invention;

[0034] Figure 3 is a schematic diagram of a common view area of ​​a vehicle radar recalibration detection method according to an embodiment of the present invention;

[0035] Figure 4 is a scanning schematic diagram of a vehicle radar recalibration detection method according to an embodiment of the present invention;

[0036] Figure 5 is a flow chart of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0037] Figure 6 is a schematic diagram of a trailing point of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0038] Figure 7 is a schematic diagram of point cloud screening of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0039] Figure 8 is a schematic diagram of a specific flow chart of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0040] Fig. 9 is a flow chart of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0041] Fig.10 is a schematic diagram in a parallel direction of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0042] Fig.11 is a schematic diagram of two clusters of point clouds according to another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0043] Fig.12 is a straight line fitting schematic diagram of another vehicle radar recalibration detection method according to an embodiment of the present invention;

[0044] Fig.13 is a structural block diagram of a detection device for vehicle radar recalibration according to an embodiment of the present invention;

[0045] Fig.14 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0047] The embodiments of the present invention are applicable to the scenario where a vehicle equipped with a radar detects the displacement of the radar during use, with the automatic guided vehicle AGV as an example. AGV operations are generally carried out in indoor environments, and there are certain requirements for the flatness of the ground. Therefore, the navigation and positioning radar used by AGV often uses a single-line laser radar to output the position of objects on the radar sensing plane. The operating environment of AGV is a complex industrial production scene. The larger the radar viewing angle, the more accurate and stable the positioning. The requirements for the navigation and positioning accuracy of the AGV body are: position <10mm, angle <1°. For many backpack AGVs, in order to cover the 360° full field of view, one vehicle generally needs to be equipped with two radars. The embodiments of the present invention provide a detection method for vehicle radar recalibration, which detects whether the radar is displaced based on radar common view point cloud data to achieve the effect of automatically recalibrating the radar.

[0048] According to an embodiment of the present invention, an embodiment of a detection method for vehicle radar recalibration is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0049] In this embodiment, a vehicle radar recalibration detection method is provided, which can be used for the above-mentioned AGV built-in positioning software module. Figure 1 is a flow chart of a vehicle radar recalibration detection method according to an embodiment of the present invention, such as Figure 1 As shown, the process includes the following steps:

[0050] Step S101, obtaining common view point cloud data of a vehicle in a common view area between a first radar and a second radar.

[0051] Specifically, in the embodiment of the present invention, Figure 2 As shown in the figure, taking the example of a front radar deployed on the right front of the AGV and a rear radar deployed on the left rear, when the AGV leaves the factory, the external parameters of the two radars have been calibrated, that is, the position and angle of the radar relative to the center of the vehicle body are accurately calculated; during use, the point cloud of environmental objects scanned by the radar will be converted to the coordinate system of the center of the vehicle body, and then the navigation and positioning calculation will be performed. Therefore, for the same object in the radar common view area, whether it is scanned by the front radar or the rear radar, its position in the vehicle body coordinate system should be exactly the same. The radar has a certain perception range. If the perception ranges of different radars overlap, the radars have a common view area, such as Figure 3 As shown in FIG. 1 , there is a front common viewing area and a rear common viewing area between the front radar and the rear radar of the AGV, which is only an example and is not limited to this. If the front radar and the rear radar of the AGV are scanning in a rectangular room, such as Figure 4In the scanned image shown, the dotted line is the point cloud scanned by the front radar, and the solid line is the point cloud scanned by the rear radar. The front left and rear right of the vehicle body are the front common viewing area (common viewing area 1) and the rear common viewing area (common viewing area 2) of the radar, that is, both the front and rear radars can scan. If the radar is calibrated very well, the point clouds of the corners seen by the front and rear radars will completely overlap. Therefore, in order to perform mutual calibration based on two radars, the embodiment of the present invention obtains the common viewing point cloud data collected by the common viewing areas of the two radars, that is, the point cloud data of the front radar within the common viewing area and the point cloud data of the rear radar within the common viewing area.

[0052] Step S102, determining whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition.

[0053] Specifically, in an embodiment of the present invention, in order to determine whether the radar is displaced based on the common view point cloud data of two radars, it is necessary to predetermine whether the common view area environment represented by the common view point cloud data obtained at this time is suitable for displacement detection, so it is necessary to determine whether the common view point cloud data meets the verification conditions. The verification conditions set in the embodiment of the present invention include: point number verification conditions, distance verification conditions, and angle verification conditions. Among them, by judging the number of points in the point cloud data, it can be ensured that the current data is sufficient to support radar displacement detection and ensure the reliability of the detection results; by judging the distance between the point cloud data of the two radars, it can be determined that the two radars are currently detecting the same object, so the positions can be calibrated based on the two radars; by judging the angles of different cluster point clouds, it is possible to detect whether the radar is displaced based on point clouds in different directions, avoiding the situation where the subsequent point cloud matching is degraded due to point clouds in one direction.

[0054] In some optional implementations, if any one of the point verification condition, the distance verification condition, and the angle verification condition is not met, it is proved that the common view area environment corresponding to the common view point cloud data at this time is not suitable for displacement detection, that is, the displacement detected may be inaccurate. However, if the point verification condition, the distance verification condition, and the angle verification condition are all met, it is proved that the common view area environment corresponding to the common view point cloud data at this time is suitable for displacement detection.

[0055] Step S103: If the conditions are met, determine whether the first radar or the second radar is displaced based on the common view point cloud data; if so, recalibrate the external parameters of the first radar or the second radar.

[0056] Specifically, in an embodiment of the present invention, if it is proved that the common view area environment corresponding to the common view point cloud data of the front radar and the rear radar is suitable for displacement detection, the point cloud data of the front radar in the common view area and the point cloud data of the rear radar in the common view area are matched. Because the external parameters of the two radars have been calibrated when the AGV leaves the factory, the current external parameters correspond to the fixed relative position of the front radar and the rear radar. If the front radar and the rear radar are not displaced, then after scanning an object, the point cloud data at the same position of the object is basically consistent, that is, the point cloud data of the front radar in the common view area and the point cloud data of the rear radar in the common view area are matched to a certain extent, and the matching error is within an allowable range.

[0057] In some optional implementations, if the embodiment of the present invention determines that the matching error of the point cloud data corresponding to the two radars exceeds the allowable error range, it means that the front radar or the rear radar is displaced, and the relative positions of the two radars are changed. If the displacement is not serious, that is, the error is within a certain range, the AGV can directly recalibrate the external parameters of the front radar or the rear radar according to the error. After recalibration, the relative positions of the two radars are consistent with when the AGV leaves the factory, thereby ensuring the accuracy of positioning. When recalibrating the radar, the embodiment of the present invention selects the front radar as the reference and adjusts the external parameters of the rear radar according to the error, but is not limited to this.

[0058] The vehicle radar recalibration detection method provided by the present invention obtains the common view point cloud data of the first radar and the second radar in the common view area on the vehicle, and determines whether the common view point cloud data meets the point verification conditions, distance verification conditions and angle verification conditions. If so, it determines whether the first radar or the second radar is displaced according to the common view point cloud data. If so, the external parameters of the first radar or the second radar are recalibrated. Based on the point cloud data of the radar with the common view area, the present invention can automatically determine whether the current radar is displaced without manual intervention, so as to recalibrate the radar in time, ensure the navigation accuracy of the vehicle, and avoid collision with obstacles.

[0059] In this embodiment, a vehicle radar recalibration detection method is provided, which can be used in the above-mentioned AGV built-in positioning software module. Figure 5 is a flow chart of a vehicle radar recalibration detection method according to an embodiment of the present invention, such as Figure 5 As shown, the process includes the following steps:

[0060] Step S501: Acquire common view point cloud data of a vehicle in a common view area between a first radar and a second radar.

[0061] Specifically, the above step S501 includes:

[0062] Step S5011: Acquire first original data of a first radar and second original data of a second radar, wherein the sensing range of the first radar and the sensing range of the second radar have at least one common view area.

[0063] Specifically, in the embodiment of the present invention, Figure 3 As shown in the figure, during the operation of the AGV, the front radar and the rear radar perform laser scanning of the surrounding environment in real time to obtain the original point cloud data within the sensing range. The sensing ranges of the front radar and the rear radar overlap, so there are two common viewing areas: the front common viewing area (common viewing area 1) and the rear common viewing area (common viewing area 2).

[0064] Step S5012: filtering out the trailing points of the first original data and the second original data, and transforming them into a vehicle body coordinate system to obtain first point cloud data and second point cloud data.

[0065] Specifically, in an embodiment of the present invention, during the laser radar scanning process, due to the divergence angle of the laser pulse, when the laser pulse hits the surface of an object, it will not only form a light spot on the target object, but may also form an extension of the light spot on the front and rear objects. This phenomenon is called the tailing phenomenon. The tailing phenomenon can cause errors in the laser radar when processing point cloud data, and it is impossible to accurately judge the distance and position of the object. The tailing point is equivalent to a noise point, and it will also affect the judgment of the radar displacement of the present invention. Therefore, the embodiment of the present invention filters out the tailing points of the original point cloud data of the front radar and the original point cloud data of the rear radar. Among them, there are many filtering methods in the conventional technical means in this field. The embodiment of the present invention chooses to judge the angles of the continuous scanning points p1 and p2 in the radar original point cloud data to filter out the tailing points, such as Figure 6 As shown, the judgment conditions are as follows. As long as one of the following judgment conditions is met, p2 is determined to be a trailing point and deleted:

[0066] ∠Op1p2>max_angle

[0067] ∠Op1p2<min_angle

[0068] In some optional embodiments, after the embodiments of the present invention delete the trailing points in the original point cloud data, the original point cloud data needs to be converted into the vehicle body coordinate system, which is equivalent to unifying the coordinates of the front radar and the rear radar, thereby obtaining the first point cloud data of the front radar in the vehicle body coordinate system and the second point cloud data of the rear radar in the vehicle body coordinate system. The point cloud data at this time is all the point cloud data after deleting the trailing points.

[0069] Step S5013, filtering the first point cloud data and the second point cloud data according to the common view area range to obtain the first common view point cloud data of the first radar within the common view area range, and the second common view point cloud data of the second radar within the common view area range.

[0070] Specifically, in the embodiment of the present invention, Figure 7 As shown in the figure, the perception ranges of the front radar and the rear radar overlap to form common viewing areas 1 and 2. According to the actual mechanical size of the AGV vehicle and the installation position of the radar, combined with the scanning range of each radar, the common viewing area range of the radar can be calculated, so as to filter out the first common viewing point cloud data of the front radar in the common viewing area and the second common viewing point cloud data of the rear radar in the common viewing area according to the coordinate information of the point cloud and the coordinate range of the common viewing area.

[0071] Step S502, obtaining the rotational angular velocity, vehicle speed and angular velocity of the vehicle, and determining whether the vehicle is stationary based on the rotational angular velocity, vehicle speed and angular velocity.

[0072] Specifically, in the embodiment of the present invention, considering the moving vehicle, since the frequencies of the front radar and the rear radar are different, when the vehicle is moving or rotating at high speed, since the point clouds collected by the radar are not obtained at the same time, there will be motion distortion, and the distorted point cloud data will have a great interference in determining whether the radar is shifted. Figure 8 As shown, the AGV of the embodiment of the present invention will perform vehicle radar recalibration detection only when it is stationary, such as during task intervals, parking and charging periods, interactive vehicle return periods, task avoidance stages, etc.

[0073] In some optional implementations, an IMU (Inertial Measurement Unity) and a wheel speed sensor are pre-installed on the AGV. The IMU can output the rotational angular velocity angular_gyro and the acceleration of the vehicle body. The wheel speed sensor can calculate and output the vehicle body velocity vel_vehicle and the angular velocity angular_vehicle according to the vehicle body model. According to the parameter values ​​collected by the two sensors, it can be determined whether the vehicle body is in a stationary state. The judgment conditions for the stationary state are as follows. When the following judgment conditions are met at the same time, the AGV is determined to be in a stationary state:

[0074] angular_gyro<2° / s

[0075] vel_vehicle<0.005m / s

[0076] angular_vehicle<0.5° / s

[0077] Step S503: determine whether the common view point cloud data meets the preset verification conditions, where the verification conditions include at least one of the following: point number verification condition, distance verification condition and angle verification condition. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0078] Step S504: If the conditions are met, determine whether the first radar or the second radar is displaced based on the common view point cloud data. If so, recalibrate the external parameters of the first radar or the second radar. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0079] The vehicle radar recalibration detection method provided by the present invention obtains the common view point cloud data of the first radar and the second radar in the common view area on the vehicle, and determines whether the common view point cloud data meets the point verification conditions, distance verification conditions and angle verification conditions. If so, it determines whether the first radar or the second radar is displaced according to the common view point cloud data. If so, the external parameters of the first radar or the second radar are recalibrated. Based on the point cloud data of the radar with the common view area, the present invention can automatically determine whether the current radar is displaced without manual intervention, so as to recalibrate the radar in time, ensure the navigation accuracy of the vehicle, and avoid collision with obstacles.

[0080] In this embodiment, a vehicle radar recalibration detection method is provided, which can be used in the above-mentioned AGV built-in positioning software module. Fig. 9 is a flow chart of a vehicle radar recalibration detection method according to an embodiment of the present invention, such as Fig. 9 As shown, the process includes the following steps:

[0081] Step S901, obtain the common view point cloud data of the vehicle in the common view area of ​​the first radar and the second radar. Figure 5 Step S501 of the illustrated embodiment will not be described in detail here.

[0082] Step S902 , determining whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition.

[0083] Specifically, the above step S902 includes:

[0084] Step S9021, clustering the first common view point cloud data, selecting the first cluster of point cloud data and the second cluster of point cloud data with the top two points in terms of point cloud points, and clustering the second common view point cloud data, selecting the third cluster of point cloud data and the fourth cluster of point cloud data with the top two points in terms of point cloud points.

[0085] Specifically, in an embodiment of the present invention, the point cloud data is a series of discrete points, representing the contour of the scanned object. In order to better understand and analyze the data, the first common view point cloud data of the front radar in the common view area and the second common view point cloud data of the rear radar in the common view area are clustered. Point cloud data clustering is a technology that groups points in a point cloud according to spatial position and attribute similarity. Through clustering, the point cloud data can be divided into multiple clusters, and the points in each cluster have high similarity, while the points between different clusters have large differences. This grouping helps to reveal the inherent structure and pattern of the point cloud data, so as to better perform subsequent processing and analysis. In an embodiment of the present invention, two clusters of point clouds with the top two points in point cloud number after clustering are selected, that is, after clustering the first common view point cloud data of the front radar, two clusters of point cloud data with the top two points in point cloud number (the first cluster of point cloud data and the second cluster of point cloud data) are selected, and after clustering the second common view point cloud data of the rear radar, two clusters of point cloud data with the top two points in point cloud number (the third cluster of point cloud data and the fourth cluster of point cloud data) are selected. The reason for taking two clusters is that if the radar point cloud is distributed in only one direction, such as a straight wall, then when the subsequent point cloud is matched, degradation will occur in the direction parallel to the wall, and it will be impossible to identify whether the two radars are consistent in the direction parallel to the wall. Fig.10 As shown in the figure, if there is only a common view point cloud in the X direction, it is impossible to determine whether the radar position has changed in the Y direction.

[0086] In some optional implementations, under normal circumstances, the first cluster of point cloud data corresponds to the third cluster of point cloud data and is located in the same area of ​​the same common viewing zone, and the second cluster of point cloud data corresponds to the fourth cluster of point cloud data and is located in the same area of ​​the same common viewing zone. However, the two clusters of point cloud data of the front radar or the two clusters of point cloud data of the rear radar may be located in different areas of the same common viewing zone, or may be located in different common viewing zones. For example, during the scanning process, no object is scanned in the rear common viewing zone, and at this time, all clusters of point cloud data are in the front common viewing zone. The embodiment of the present invention takes two clusters of point cloud data in two common viewing zones as an example, such as Fig.11 shown.

[0087] Step S9022: judging whether the common view point cloud data meets the verification condition according to the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data and the fourth cluster of point cloud data.

[0088] Specifically, in the implementation of the present invention, in order to verify whether the common view area environment corresponding to the point cloud data obtained at this time can accurately detect whether the radar is displaced, point verification conditions, distance verification conditions and angle verification conditions are set. Only when all verification conditions are met can it be ensured that the current point cloud data can be used to accurately detect whether the radar is displaced.

[0089] In some optional implementations, the above step S9022 includes:

[0090] Step a1, obtain the first point number of the first cluster point cloud data, the second point number of the second cluster point cloud data, the third point number of the third cluster point cloud data and the fourth point number of the fourth cluster point cloud data; determine whether the first point number, the second point number, the third point number and the fourth point number are all greater than a preset quantity threshold, if they are all greater, then the point verification condition is met.

[0091] Specifically, in the embodiment of the present invention, Figure 8 As shown, in order to ensure that each cluster of point cloud data selected after clustering is representative, it is necessary to ensure that the number of points in each cluster of point cloud data is higher than the preset number threshold, that is, the number of points size_cluster of the first cluster of point cloud data and the second cluster of point cloud data of the front radar, the third cluster of point cloud data and the fourth cluster of point cloud data of the rear radar are higher than the preset number threshold cluster_num_thre, and the common view point cloud data is determined to meet the point verification condition. In this embodiment of the present invention, the preset number threshold cluster_num_thre is set to 50, but it is not limited thereto. The corresponding point verification condition is as follows:

[0092] size_cluster>cluster_num_thre

[0093] Step a2, calculate the first distance interval between the first centroid of the first cluster point cloud data and the second centroid of the second cluster point cloud data, and the second distance interval between the third centroid of the third cluster point cloud data and the fourth centroid of the fourth cluster point cloud data; determine whether the first distance interval and the second distance interval are both less than a preset distance threshold, if both are less than, the distance verification condition is met.

[0094] Specifically, in the embodiment of the present invention, Figure 8 As shown, in order to determine whether the front radar and the rear radar scan the same object, it is necessary to judge the distance between the corresponding cluster point cloud data of the front radar and the rear radar. Only when the distance meets a certain distance threshold, it is determined that the front radar and the rear radar scan the same object. Only under this premise can it be detected whether the radar is displaced based on the point cloud data. In the embodiment of the present invention, the centroid of each cluster of point cloud data represents the position of each cluster of point cloud data. For example, the centroid of the cluster point cloud data of the front radar in the common view area 1 is center_1_front, and the centroid of the cluster point cloud data of the rear radar in the common view area 1 is center_1_rear. If the distance between the two centroids is less than the preset distance interval dis_thre, and the distance between the centroids of the cluster point cloud data of the front radar and the rear radar in the corresponding common view area 2 is also less than the preset distance interval dis_thre, then the distance verification condition is met. In the embodiment of the present invention, the preset distance threshold dis_thre is set to 0.1, but is not limited thereto. The corresponding distance verification condition is as follows:

[0095] dis(center_1_front,center_1_rear)<dis_thre

[0096] dis(center_2_front,center_2_rear)<dis_thre

[0097] Step a3, performing straight line fitting on the first cluster point cloud data and the third cluster point cloud data to obtain a first straight line, and performing straight line fitting on the second cluster point cloud data and the fourth cluster point cloud data to obtain a second straight line; judging whether the angle between the first straight line and the second straight line is less than a preset angle threshold, if so, the angle verification condition is met.

[0098] Specifically, in the embodiment of the present invention, Figure 8 As shown, selecting two clusters of point cloud data from the point cloud data of each radar can only avoid the degradation caused by a single direction to a certain extent, but at this time it is not certain that the two clusters of point cloud data correspond to different directions. Therefore, in order to accurately determine whether the radar is displaced, it is necessary to determine whether the two clusters of point cloud data obtained from a certain radar are distributed in the X and Y directions. In this embodiment of the present invention, a first cluster of point cloud data and a third cluster of point cloud data in the same area of ​​the same common view area are linearly fitted to obtain a corresponding first straight line line1, and at the same time, a second cluster of point cloud data and a third cluster of point cloud data in another area of ​​the same common view area or in another common view area are linearly fitted to obtain a corresponding second straight line line2, as shown in FIG. Fig.12 As shown, if the angle between the two straight lines exceeds the preset angle threshold angle_thre, the angle verification condition is satisfied. In the embodiment of the present invention, the preset angle threshold angle_thre is set to 45°, but is not limited thereto. The corresponding angle verification condition is as follows:

[0099] angle(line1,line2)>angle_thre

[0100] Step S903: If the conditions are met, determine whether the first radar or the second radar is displaced based on the common view point cloud data; if so, recalibrate the external parameters of the first radar or the second radar.

[0101] Specifically, the above step S903 includes:

[0102] Step S9031, using a preset point cloud registration algorithm, iteratively register the first total point cloud composed of the first cluster point cloud data and the second cluster point cloud data, and the second total point cloud composed of the third cluster point cloud data and the fourth cluster point cloud data until the preset registration parameters are met.

[0103] Specifically, in an embodiment of the present invention, two clusters of point cloud data of the front radar are combined into a first total point cloud, and two clusters of point cloud data of the rear radar are combined into a second total point cloud, and the ICP (Iterative Closest Point) algorithm in the PCL (Point Cloud Library) point cloud library is used to iteratively align the first total point cloud and the second total point cloud. The ICP algorithm in the PCL is a classic algorithm for point cloud alignment, and the core idea of ​​the algorithm is to iteratively find the closest point pair between the two point clouds and calculate the optimal rigid transformation (including rotation and translation) to align the source point cloud in the coordinate system of the target point cloud. In the embodiment of the present invention, the front radar is used as a reference, that is, the first total point cloud of the front radar is the target point cloud, and the second total point cloud of the rear radar is the source point cloud. The algorithm processing process includes: ① initialization: setting the initial transformation matrix, which is usually a unit matrix. In the embodiment of the present invention, the position difference of the two point clouds is set to 0 and the angle is set to 0; ② nearest point matching: for each point in the source point cloud, the nearest point is found in the target point cloud, and the nearest point is usually defined by the Euclidean distance; ③ transformation matrix estimation: through SVD (singular value decomposition) or least squares method, the rigid transformation matrix that makes the nearest point in the source point cloud coincide with the matching point in the target point cloud as much as possible is estimated; ④ update point cloud: apply the estimated rigid transformation matrix to the source point cloud to make it closer to the target point cloud; ⑤ iteration judgment: check whether the iteration termination condition can meet the preset registration parameters, such as the number of iterations reaches the upper limit, the change of the transformation matrix between two iterations is less than the set threshold, or the error drops below a certain specified value, etc. The preset registration parameters of the embodiment of the present invention are as follows:

[0104] setMaxCorrespondenceDis=0.2

[0105] setMaxIteration=20

[0106] setEpsilon = 0.2

[0107] Among them, setMaxCorrespondenceDis is the maximum distance threshold. The distance between points is less than this value to be considered a matching point pair; setMaxIteration is the maximum number of iterations. If the number of iterations is greater than this value, the iteration is exited; setEpsilon is the maximum difference (error) between two iteration results. If the difference between the two iteration results is less than this value, the alignment is considered successful, the alignment result is returned and the iteration is exited. The specific alignment iteration process is a conventional technical means in this field and will not be repeated here.

[0108] Step S9032: Calculate the error between the current relative position and the calibrated relative position between the first total point cloud and the second total point cloud according to the registration result.

[0109] Specifically, in an embodiment of the present invention, after the registration is completed, the error between the current relative position and the calibrated relative position between the first total point cloud of the front radar and the second total point cloud of the rear radar can be obtained, including the lateral distance error error_x, the longitudinal distance error error_y and the angle error error_angle.

[0110] Step S9033: determine whether the relative position between the first radar and the second radar is shifted according to the error and the preset error threshold; if so, adjust the external parameters of the first radar or the second radar according to the error.

[0111] Specifically, in the embodiment of the present invention, in theory, if the radar is not displaced, error_x, error_y and error_angle are within the factory error range. However, because the factory calibration is in a standard environment, considering that the environment during the actual operation of the AGV is relatively messy, the embodiment of the present invention provides a slightly broader threshold, including: a first preset distance error threshold error_x_thre, a second preset distance error threshold error_y_thre and a preset angle error threshold error_angle_thre. The values ​​of each threshold are as follows, but are not limited to this:

[0112] error_x_thre=0.01

[0113] error_y_thre=0.01

[0114] error_angle_thre=1°

[0115] In some optional implementations, the above step S9033 includes:

[0116] Step b1: If the error is less than a preset error threshold, it is determined that the first radar and / or the second radar is not displaced.

[0117] Step b2: If the error is greater than a preset error threshold and less than a preset multiple of the preset error threshold, it is determined that the first radar and / or the second radar are displaced, a warning signal is generated, and after the condition is satisfied for more than a preset number of times in a row, the external parameters of the first radar or the second radar are adjusted according to the lateral distance error, the longitudinal distance error and / or the angle error.

[0118] Step b3: If the error is greater than a preset multiple of a preset error threshold, it is determined that a fault has occurred and a fault signal is generated.

[0119] Specifically, in the embodiment of the present invention, error_x, error_y and error_angle are compared with corresponding thresholds in turn, and a three-level processing scheme is set according to the comparison result, including a safety level, a warning level and a fault level.

[0120] 1. If error < error_thre, it is at a safe level, indicating that the radar has not been abnormally displaced and can operate normally. The specific judgment conditions are as follows:

[0121] error_x<error_x_thre

[0122] error_y<error_y_thre

[0123] error_angle<error_angle_thre

[0124] 2. If it is determined that error_thre<error<2*error_thre, it is at the warning level, indicating that the radar is abnormally displaced, and a warning signal is generated at this time, such as Figure 8 As shown, after being judged to be at the warning level for more than 3 consecutive times, the external parameters of the second radar are adjusted directly according to the lateral distance error, longitudinal distance error and angle error. The specific judgment conditions are as follows, and the preset multiple is set to 2 times, but not limited to this:

[0125] error_x_thre<error_x<2*error_x_thre

[0126] error_y_thre<error_y<2*error_y_thre

[0127] error_angle_thre<error_angle<2*error_angle_thre

[0128] judge_warn_times>3

[0129] 3. If error>2*error_thre is determined, it is at the fault level, indicating that a fault may occur. At this time, a fault signal is generated, such as Figure 8 As shown, the user is reminded to intervene, thereby avoiding further expansion of the fault, reducing the vehicle failure rate, and saving unnecessary economic losses. The specific judgment conditions are as follows:

[0130] error_x>2*error_x_thre

[0131] error_y>2*error_y_thre

[0132] error_angle>2*error_angle_thre

[0133] In some optional implementations, in the judgment process of the above three levels, each error needs to meet the judgment conditions in order to determine the current level, and then handle it according to different response strategies. If not, the level cannot be determined and no processing is performed.

[0134] The vehicle radar recalibration detection method provided by the present invention obtains the common view point cloud data of the first radar and the second radar in the common view area on the vehicle, and determines whether the common view point cloud data meets the point verification conditions, distance verification conditions and angle verification conditions. If so, it determines whether the first radar or the second radar is displaced according to the common view point cloud data. If so, the external parameters of the first radar or the second radar are recalibrated. Based on the point cloud data of the radar with the common view area, the present invention can automatically determine whether the current radar is displaced without manual intervention, so as to recalibrate the radar in time, ensure the navigation accuracy of the vehicle, and avoid collision with obstacles.

[0135] In this embodiment, a detection device for vehicle radar recalibration is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated here. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0136] This embodiment provides a detection device for vehicle radar recalibration, such as Fig.13 As shown, including:

[0137] The data acquisition module 1301 is used to acquire the common view point cloud data of the vehicle in the common view area of ​​the first radar and the second radar.

[0138] The condition verification module 1302 is used to determine whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition.

[0139] The displacement determination module 1303 is used to determine whether the first radar or the second radar is displaced according to the common view point cloud data if the conditions are met, and to recalibrate the external parameters of the first radar or the second radar if they are displaced.

[0140] In some optional embodiments, the device further includes: a state determination module, which is used to obtain the rotational angular velocity, vehicle speed and angular velocity of the vehicle, and determine whether the vehicle is in a stationary state based on the rotational angular velocity, vehicle speed and angular velocity.

[0141] In some optional implementations, the data acquisition module 1301 includes:

[0142] The raw data acquisition unit is used to acquire first raw data of the first radar and second raw data of the second radar, and the sensing range of the first radar and the sensing range of the second radar have at least one common view area.

[0143] The raw data processing unit is used to filter out the trailing points of the first raw data and the second raw data, and transform them into a vehicle body coordinate system to obtain the first point cloud data and the second point cloud data.

[0144] The point cloud data screening unit is used to screen the first point cloud data and the second point cloud data according to the common view area range to obtain the first common view point cloud data of the first radar within the common view area and the second common view point cloud data of the second radar within the common view area.

[0145] In some optional implementations, the condition verification module 1302 includes:

[0146] The point cloud clustering unit is used to cluster the first common view point cloud data, select the first cluster of point cloud data and the second cluster of point cloud data with the top two points in terms of point cloud points, and cluster the second common view point cloud data, select the third cluster of point cloud data and the fourth cluster of point cloud data with the top two points in terms of point cloud points.

[0147] The point cloud verification unit is used to determine whether the common view point cloud data meets the verification conditions based on the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data and the fourth cluster of point cloud data.

[0148] In some optional embodiments, the point cloud verification unit includes:

[0149] The point verification subunit is used to obtain the first point number of the first cluster point cloud data, the second point number of the second cluster point cloud data, the third point number of the third cluster point cloud data and the fourth point number of the fourth cluster point cloud data; determine whether the first point number, the second point number, the third point number and the fourth point number are all greater than a preset quantity threshold, and if so, the point verification condition is met.

[0150] The distance verification subunit is used to calculate the first distance interval between the first centroid of the first cluster point cloud data and the second centroid of the second cluster point cloud data, and the second distance interval between the third centroid of the third cluster point cloud data and the fourth centroid of the fourth cluster point cloud data; determine whether the first distance interval and the second distance interval are both less than a preset distance threshold, and if both are less than, the distance verification condition is met.

[0151] The angle verification subunit is used to perform straight line fitting on the first cluster point cloud data and the third cluster point cloud data to obtain a first straight line, and to perform straight line fitting on the second cluster point cloud data and the fourth cluster point cloud data to obtain a second straight line; and to determine whether the angle between the first straight line and the second straight line is less than a preset angle threshold. If so, the angle verification condition is met.

[0152] In some optional implementations, the shift determination module 1303 includes:

[0153] The point cloud registration unit is used to iteratively register the first total point cloud composed of the first cluster point cloud data and the second cluster point cloud data, and the second total point cloud composed of the third cluster point cloud data and the fourth cluster point cloud data using a preset point cloud registration algorithm until the preset registration parameters are met.

[0154] The error calculation unit is used to calculate the error between the current relative position and the calibrated relative position between the first total point cloud and the second total point cloud according to the registration result.

[0155] The recalibration detection unit is used to determine whether the relative position between the first radar and the second radar is shifted according to the error and a preset error threshold, and if shifted, adjust the external parameters of the first radar or the second radar according to the error.

[0156] In some optional embodiments, the recalibration detection unit includes:

[0157] The first-level detection subunit is used to determine that the first radar and / or the second radar are not displaced if the error is less than a preset error threshold, the error includes: lateral distance error, longitudinal distance error and angle error, and the preset error threshold includes a first preset distance error threshold, a second preset distance error threshold and a preset angle error threshold.

[0158] The second-level detection subunit is used to determine that the first radar and / or the second radar are displaced if the error is greater than a preset error threshold and less than a preset multiple of the preset error threshold, generate a warning signal, and adjust the external parameters of the first radar or the second radar according to the lateral distance error, the longitudinal distance error and / or the angle error after the condition is met continuously for more than a preset number of times.

[0159] The third-level detection subunit is used to determine that a fault occurs and generate a fault signal if the error is greater than a preset multiple of a preset error threshold.

[0160] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0161] The vehicle radar recalibration detection device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0162] The embodiment of the present invention also provides a computer device having the above Fig.13 The detection device for vehicle radar recalibration is shown.

[0163] See also Fig.14 , Fig.14 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig.14 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig.14 A processor 10 is taken as an example.

[0164] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0165] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0166] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0167] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0168] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Fig.14 The example of connecting through bus is taken in the following.

[0169] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0170] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0171] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0172] The embodiment of the present invention also provides a vehicle having Fig.14 The computer device and vehicle shown include AGV, but are not limited thereto.

[0173] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A vehicle radar recalibration detection method, characterized in that: The method comprises: Acquire common view point cloud data on the vehicle in the common view area of ​​the first radar and the second radar; Determining whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition; If it is consistent, it is determined whether the first radar or the second radar is displaced according to the common view point cloud data; if it is displaced, the external parameters of the first radar or the second radar are recalibrated.

2. The method according to claim 1, characterized in that After acquiring the common view point cloud data on the vehicle in the common view area of ​​the first radar and the second radar, the method further includes: The rotational angular velocity, vehicle speed and angular velocity of the vehicle are obtained, and whether the vehicle is in a stationary state is determined based on the rotational angular velocity, the vehicle speed and the angular velocity. If so, the step of determining whether the common view point cloud data meets a preset verification condition is executed.

3. The method according to claim 1, characterized in that The acquiring of the common view point cloud data of the vehicle in the common view area of ​​the first radar and the second radar comprises: Acquire first raw data of the first radar and second raw data of the second radar, wherein the sensing range of the first radar and the sensing range of the second radar have at least one common view area; Filtering out trailing points from the first original data and the second original data, and transforming them into a vehicle body coordinate system to obtain first point cloud data and second point cloud data; The first point cloud data and the second point cloud data are screened according to the common view area range to obtain the first common view point cloud data of the first radar within the common view area range and the second common view point cloud data of the second radar within the common view area range.

4. The method according to claim 3, characterized in that The determining whether the common view point cloud data meets a preset verification condition includes: Clustering the first common view point cloud data, selecting the first cluster of point cloud data and the second cluster of point cloud data with the top two points in terms of point cloud points, and clustering the second common view point cloud data, selecting the third cluster of point cloud data and the fourth cluster of point cloud data with the top two points in terms of point cloud points; Whether the common view point cloud data meets the verification condition is determined according to the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data, and the fourth cluster of point cloud data.

5. The method according to claim 4, characterized in that The determining whether the common view point cloud data meets the point count verification condition according to the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data, and the fourth cluster of point cloud data includes: Acquire a first number of points of the first cluster of point cloud data, a second number of points of the second cluster of point cloud data, a third number of points of the third cluster of point cloud data, and a fourth number of points of the fourth cluster of point cloud data; Determine whether the first point number, the second point number, the third point number, and the fourth point number are all greater than a preset number threshold, and if so, the point verification condition is satisfied; The determining whether the common view point cloud data meets the distance verification condition according to the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data, and the fourth cluster of point cloud data includes: Calculating a first distance interval between a first centroid of the first cluster of point cloud data and a second centroid of the second cluster of point cloud data, and a second distance interval between a third centroid of the third cluster of point cloud data and a fourth centroid of the fourth cluster of point cloud data; Determine whether the first distance interval and the second distance interval are both smaller than a preset distance threshold, and if both are smaller than, the distance verification condition is met; The determining whether the common view point cloud data meets the angle verification condition according to the first cluster of point cloud data, the second cluster of point cloud data, the third cluster of point cloud data, and the fourth cluster of point cloud data includes: Performing straight line fitting on the first cluster of point cloud data and the third cluster of point cloud data to obtain a first straight line, and performing straight line fitting on the second cluster of point cloud data and the fourth cluster of point cloud data to obtain a second straight line; It is determined whether the angle between the first straight line and the second straight line is less than a preset angle threshold. If so, the angle verification condition is met.

6. The method according to claim 4, characterized in that The determining whether the first radar or the second radar is displaced according to the common view point cloud data, and if displaced, recalibrating the external parameters of the first radar or the second radar, includes: Iteratively registering a first total point cloud composed of the first cluster of point cloud data and the second cluster of point cloud data, and a second total point cloud composed of the third cluster of point cloud data and the fourth cluster of point cloud data using a preset point cloud registration algorithm until preset registration parameters are met; Calculate the error between the current relative position and the calibrated relative position between the first total point cloud and the second total point cloud according to the registration result; Determine whether the relative position between the first radar and the second radar is displaced according to the error and a preset error threshold; if displaced, adjust the external parameters of the first radar or the second radar according to the error.

7. The method according to claim 6, characterized in that The determining whether the relative position between the first radar and the second radar is displaced according to the error and a preset error threshold, and if displaced, adjusting the external parameters of the first radar or the second radar according to the error, includes: If the error is less than the preset error threshold, it is determined that the first radar and / or the second radar is not displaced, the error includes: a lateral distance error, a longitudinal distance error and an angle error, and the preset error threshold includes a first preset distance error threshold, a second preset distance error threshold and a preset angle error threshold; If the error is greater than the preset error threshold and less than a preset multiple of the preset error threshold, it is determined that the first radar and / or the second radar are displaced, a warning signal is generated, and after the error is continuously satisfied for more than a preset number of times, an external parameter of the first radar or the second radar is adjusted according to the lateral distance error, the longitudinal distance error and / or the angle error; If the error is greater than the preset multiple of the preset error threshold, it is determined that a fault occurs and a fault signal is generated.

8. A detection device for vehicle radar recalibration, characterized in that: The device comprises: A data acquisition module, used to acquire common view point cloud data on the vehicle in a common view area of ​​the first radar and the second radar; A condition verification module, used to determine whether the common view point cloud data meets a preset verification condition, wherein the verification condition includes at least one of the following: a point number verification condition, a distance verification condition, and an angle verification condition; The displacement determination module is used to determine whether the first radar or the second radar is displaced according to the common view point cloud data if the conditions are met, and to recalibrate the external parameters of the first radar or the second radar if they are displaced.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle radar recalibration detection method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the vehicle radar recalibration detection method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions are used to cause a computer to execute the vehicle radar recalibration detection method according to any one of claims 1 to 7.

12. A vehicle, characterized in that: A computer device comprising the method of claim 9.

Images