阅读说明：本技术 基于三维激光雷达的装车物料高度与偏载检测方法及装置 (Three-dimensional laser radar-based method and device for detecting height and unbalance loading of loading material ) 是由 李静 白龙凌 王军政 汪首坤 赵江波 马立玲 沈伟 李金仓 于 2021-07-26 设计创作，主要内容包括：本发明公开了一种基于三维激光雷达的装车物料高度与偏载检测方法和装置,能够解决装车物料实时高度与偏载的检测问题。采用三维激光雷达扫描物料装填区,从三维激光点云中筛选出符合车厢挡板高度的点云,获得车厢挡板位置对应的y轴坐标范围；根据车厢挡板高度的限制和该y轴坐标范围,筛选出物料点云；在物料点云中筛选出第一条、第二条和最后一条激光线束的点云；根据最后一条激光线束的点云高度是否满足车厢挡板形态特征,以确定车辆是否行使到当前车厢的终点；根据第一条和第二条激光线束的点云高度是否分别满足车厢挡板形态特征和物料填充形态特征,以确定车辆是否行使到下一节车厢的起点；最后根据车厢范围内的物料点云计算物料高度与偏载。(The invention discloses a method and a device for detecting the height and the unbalance loading of a loading material based on a three-dimensional laser radar, which can solve the problem of detecting the real-time height and the unbalance loading of the loading material. Scanning a material filling area by using a three-dimensional laser radar, screening out point clouds conforming to the height of the carriage baffle from the three-dimensional laser point clouds, and obtaining a y-axis coordinate range corresponding to the position of the carriage baffle; screening out material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range; screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; determining whether the vehicle runs to the terminal of the current carriage or not according to whether the point cloud height of the last laser beam meets the morphological characteristics of the carriage baffle or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics; and finally, calculating the height and the unbalance loading of the material according to the point cloud of the material in the carriage range.)

1. A three-dimensional laser radar-based method for detecting height and unbalance loading of a loading material is characterized by comprising the following steps:

firstly, fixedly and obliquely installing a three-dimensional laser radar, wherein the three-dimensional laser radar irradiates a material filling area;

step two, self-calibration of the installation attitude angle of the three-dimensional laser radar: calibrating a yaw angle and a pitch angle by taking the horizontal ground as a reference, and calibrating a roll angle by taking baffles on the left side and the right side of the carriage as a reference to obtain a conversion relation between a laser radar coordinate system and a reference coordinate system x, y and z; wherein, the yz plane of the reference coordinate system is parallel to the ground, and the x positive direction is vertical to the ground and faces downwards;

step three, extracting material point cloud: scanning the three-dimensional laser radar to obtain coordinates (x, y, z) of the three-dimensional laser radar point cloud under a reference coordinate system; screening carriage baffle point clouds according with the height of a carriage baffle from the three-dimensional laser point clouds to obtain a y-axis coordinate range corresponding to the position of the carriage baffle; screening out material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle;

step four, determining the position of the carriage: screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; the first laser beam is the beam with the largest three-dimensional laser radar pitch angle, and the last laser beam is the beam with the smallest three-dimensional laser radar pitch angle; determining whether the vehicle runs to the terminal point of the current carriage or not according to whether the point cloud height of the last laser beam, namely the x coordinate, meets the morphological characteristics of the carriage baffle or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics;

and step five, calculating the height and the unbalance loading of the material according to the point cloud of the material in the carriage range.

2. The method according to claim 1, wherein the second step specifically comprises:

coordinate system x of three-dimensional laser radar_Ly_Lz_LIs a right-hand coordinate system; establishing a new right-hand coordinate system which is recorded as a reference coordinate system by taking the ground as a yz plane, the positive direction of an x axis is vertical to the ground downwards, and the running direction of the vehicle is the positive direction of a z axis; the transformation from the lidar coordinate system to the reference coordinate system is performed in the following order:

first winding laser radar coordinate system y_LAxial through rotation matrix R_yTransforming to obtain an intermediate system a, and rewinding z of the intermediate system a₁Axial through rotation matrix R_zTransforming to obtain an intermediate system b, and then winding x of the intermediate system b₂Axial through rotation matrix R_xTransforming to obtain intermediate system c, and translating to transform matrix T_xObtaining a reference coordinate system;

(1) calibration of yaw angle and pitch angle by taking horizontal ground as reference

Scanning the ground by a three-dimensional laser radar, collecting ground point clouds, determining ground points by adopting a plane fitting method to the ground point clouds to obtain ground points under a laser radar coordinate systemGround plane equation Ax_L+By_L+Cz_L+D＝0，A>0, wherein A, B, C, D are parameters of the ground plane equation, respectively; the height of the three-dimensional laser radar coordinate system from the origin to the ground is obtained through calculationThen the corresponding rotation matrix R_z、R_yAnd a translation transformation matrix T_xComprises the following steps:

wherein the yaw anglePitching angle

Point (x) in the lidar coordinate system_L,y_L,z_L) The coordinates (x) in the intermediate system b are obtained by the following rotation transformation₂,y₂,z₂)：

(2) Calibration of roll angle by using baffle plates on two sides of carriage as reference

The baffle of the carriage is vertical to the ground, and the height of the top of the baffle is h_b(ii) a The bottom of the empty carriage is parallel toGround surface with a height h₀(ii) a Under the intermediate system b, the x value of the ground point coordinate is l_groundThe x value of the highest point coordinate of the baffle is l_ground﹣h_b(ii) a In the radar laser point cloud obtained by scanning the carriage area by the three-dimensional laser radar, the coordinate x value and the coordinate l are satisfied_ground﹣h_bThe point with the difference value smaller than the threshold value is recorded as a top baffle threshold value point set, the point set is provided with two areas which respectively belong to the left baffle and the right baffle, and all the points of the two areas are respectively projected to y₂z₂Performing straight line fitting behind the plane to obtain the position of the top of the baffle plate at y₂z₂Equation A' y for two parallel straight lines of a plane₂+B'z₂+C'₁0 and A' y₂+B'z₂+C'₂＝0，A'>0; corresponding rotational transformation matrix R_xIs defined as follows:

wherein the roll angleOrDepending on the direction of travel of the train, the requirement is that the direction of travel of the freight train is z in the intermediate system c₃The shaft is in the positive direction; point (x) under the intermediate system b₂,y₂,z₂) The coordinates (x) in the intermediate system c are obtained by the following rotation transformation₃,y₃,z₃)：

Point (x) under the intermediate system c₃,y₃,z₃) Obtaining coordinates (x, y, z) under a reference coordinate system through translation transformation as follows:

and finally, obtaining a point cloud coordinate under the reference coordinate system, wherein the height information of the point is a negative value of the x coordinate.

3. The method of claim 1, wherein in step three, the car baffle point cloud which meets the height of the car baffle is screened from the three-dimensional laser point cloud, and the y-axis coordinate range corresponding to the position of the car baffle is obtained as follows:

under a reference coordinate system, the x value of the ground point coordinate is 0, and the height of the top of the baffle is h_bThe x value of the highest point coordinate of the baffle is-h_b(ii) a Coordinate (x, y, z) screened from three-dimensional laser point cloud meets condition | x- (-h)_b)|<th_rangeProjecting the point cloud to a yz plane to obtain a carriage baffle point set;

the concentrated points of the baffle points of the carriage are provided with two areas which are respectively the projection of the top point of the baffle on the left side of the carriage and the top point of the baffle on the right side of the carriage; selecting a point set of an area corresponding to a baffle on the left side of the carriage, and calculating to obtain the minimum value of y coordinates of all the points, which is recorded as y_max(ii) a Selecting a point set of an area corresponding to a baffle on the right side of the carriage, and calculating to obtain the maximum value of y coordinates of all the points, which is recorded as y_min；y_minAnd y_maxAnd forming a y-axis coordinate range corresponding to the carriage baffle position.

4. The method of claim 3, wherein in step three, the material point cloud is screened out according to the limit of the carriage baffle height and the y-axis coordinate range corresponding to the carriage baffle position as follows:

screening all laser radar points under a reference coordinate system, wherein the coordinates (x, y, z) of the points meet the following conditions, namely the material point cloud:

h₀≤-x<h_max，y_min≤y≤y_max

wherein h is₀Is the height of the bottom of the empty carriage.

5. The method of claim 1, wherein in step four, the step of determining whether the vehicle is driven to the end of the current compartment according to whether the point cloud height (x coordinate) of the last laser beam satisfies the compartment closure morphological feature is as follows:

selecting point clouds belonging to the last laser beam from the material point clouds in the reference coordinate system, and calculating the variance sigma of the x coordinate_end(ii) a If σ_end<th_end，th_endAnd setting a threshold value, wherein the baffle at the rear end of the carriage of the train runs to the material scanning area, and the train runs to the end point of the carriage.

6. The method of claim 1, wherein in step four, the determining whether the vehicle travels to the starting point of the next car according to whether the point cloud heights of the first and second laser beams satisfy the car closure profile and the material filling profile, respectively, is:

selecting point clouds belonging to a first laser beam and a second laser beam from material point clouds under a reference coordinate system, and respectively calculating the variance of an x coordinate and marking as sigma₁And σ₂If σ is₁<th₁And sigma₂>th₂，th₁Threshold value for point cloud change to conform to baffle morphology feature, th₂If the point cloud change meets the threshold value of the material filling form characteristics, the baffle at the front end of the new carriage is judged to be about to enter the material scanning area, and the train runs to the starting point of the new carriage.

7. The method of claim 1, wherein step five is:

step 501: dividing the material point cloud into a left side and a right side relative to the train, and respectively calculating the two parts as follows:

screening out material point clouds of the unilateral carriage according to the y coordinate of the unilateral carriage baffle and the y coordinate of the carriage center line under a reference coordinate system;

dividing the material point cloud of the single-side compartment into M areas along the y axis; the three-dimensional laser radar has N laser beams; for each area, averaging the x coordinates of all point clouds of each laser beam of the N laser beams, and obtaining the material point cloud height corresponding to the N laser beams in each area; averaging the point cloud heights corresponding to the N line bundles to obtain a material point cloud height corresponding to the current area; averaging the point cloud heights of the materials in the M areas to obtain the material height of the carriage on one side;

step 502: calculating the height and the unbalance loading of the material:

the real-time material height h is:

wherein h is₀Height of empty carriage bottom, h_LThe material height of the left carriage, h_RThe material height of the right carriage;

the left and right offset p of the gravity center of the material is as follows:

p>at 0, the right side is heavy; p is a radical of<0, left-side weight, y_minAnd y_maxTwo y-axis coordinate limit values corresponding to the car baffle positions, respectively.

8. A three-dimensional laser radar-based vehicle-loading material height and unbalance loading detection device is characterized by comprising a three-dimensional laser radar and a processing device; the processing device comprises a calibration module, a material point cloud extraction module, a carriage position determination module and a height and unbalance load calculation module;

the three-dimensional laser radar is fixedly and obliquely arranged above the material filling area and irradiates the material filling area;

the calibration module realizes the self-calibration of the installation attitude angle of the three-dimensional laser radar: calibrating a yaw angle and a pitch angle by taking the ground as a reference, and calibrating a roll angle by taking baffles on the left side and the right side of a carriage as a reference to obtain a conversion relation between a laser radar coordinate system and a reference coordinate system x, y and z; wherein, the yz plane of the reference coordinate system is parallel to the ground, and the x positive direction is vertical to the ground and faces downwards; scanning the three-dimensional laser radar to obtain three-dimensional laser radar point cloud, processing the three-dimensional laser radar point cloud by a calibration module to obtain coordinates (x, y, z) of the three-dimensional laser radar point cloud under a reference coordinate system, and sending the coordinates to a material point cloud extraction module;

the material point cloud extraction module is used for screening carriage baffle point clouds conforming to the height of the carriage baffle from the three-dimensional laser point clouds according to the coordinates (x, y, z) of the three-dimensional laser radar point clouds under the reference coordinate system, and obtaining a y-axis coordinate range corresponding to the position of the carriage baffle; screening material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle, and sending the material point clouds to a carriage position determining module;

the carriage position determining module is used for screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; the first laser beam is the beam with the largest three-dimensional laser radar pitch angle, and the last laser beam is the beam with the smallest three-dimensional laser radar pitch angle; determining whether the vehicle runs to the terminal of the current carriage or not according to whether the point cloud height of the last laser beam, namely the x coordinate, meets the morphological characteristics of the carriage or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics;

and the height and unbalance loading calculation module is used for calculating the height and unbalance loading of the material according to the material point cloud in the carriage range.

9. The three-dimensional laser radar-based vehicle-loading material height and unbalance loading detection device as claimed in claim 8, wherein the calibration module is calibrated in a manner that:

coordinate system x of three-dimensional laser radar_Ly_Lz_LIs a right-hand coordinate system; establishing a new right-hand coordinate system by taking the ground as a yz plane, the positive direction of an x axis is vertical to the ground downwards, and the running direction of the vehicle is the positive direction of a z axis, and recording the new right-hand coordinate system as a reference coordinateIs a step of; the transformation from the lidar coordinate system to the reference coordinate system is performed in the following order:

first winding laser radar coordinate system y_LAxial through rotation matrix R_yTransforming to obtain an intermediate system a, and rewinding z of the intermediate system a₁Axial through rotation matrix R_zTransforming to obtain an intermediate system b, and then winding x of the intermediate system b₂Axial through rotation matrix R_xTransforming to obtain intermediate system c, and finally performing translation transformation T_xObtaining a reference coordinate system;

(1) calibration of yaw angle and pitch angle by taking horizontal ground as reference

Scanning the ground by a three-dimensional laser radar, collecting ground point clouds, determining ground points by adopting a plane fitting method to the ground point clouds to obtain a ground plane equation Ax under a laser radar coordinate system_L+By_L+Cz_L+D＝0，A>0, wherein A, B, C, D are parameters of the ground plane equation, respectively; the height of the three-dimensional laser radar coordinate system from the origin to the ground is obtained through calculationThen the corresponding rotation matrix R_z、R_yAnd a translation transformation matrix T_xComprises the following steps:

wherein the yaw anglePitching angle

Point (x) in the lidar coordinate system_L,y_L,z_L) The coordinates (x) in the intermediate system b are obtained by the following rotation transformation₂,y₂,z₂)：

(2) Calibration of roll angle by using baffle plates on two sides of carriage as reference

The baffle of the carriage is vertical to the ground, and the height of the top of the baffle is h_b(ii) a The bottom of the empty carriage is parallel to the ground and has the height h₀(ii) a Under the intermediate system b, the x value of the ground point coordinate is l_groundThe x value of the highest point coordinate of the baffle is l_ground﹣h_b(ii) a In the radar laser point cloud obtained by scanning the carriage area by the three-dimensional laser radar, the coordinate x value and the coordinate l are satisfied_ground﹣h_bThe point with the difference value smaller than the threshold value is recorded as a top baffle threshold value point set, the point set is provided with two areas which respectively belong to the left baffle and the right baffle, and all the points of the two areas are respectively projected to y₂z₂Performing straight line fitting behind the plane to obtain the position of the top of the baffle plate at y₂z₂Equation A' y for two parallel straight lines of a plane₂+B'z₂+C'₁0 and A' y₂+B'z₂+C'₂＝0，A'>0; corresponding rotational transformation matrix R_xIs defined as follows:

wherein the roll angleOrDepending on the direction of travel of the train, the requirement is that the direction of travel of the freight train is z in the intermediate system c₃The shaft is in the positive direction; point (x) under the intermediate system b₂,y₂,z₂) The coordinates (x) in the intermediate system c are obtained by the following rotation transformation₃,y₃,z₃)：

Point (x) under the intermediate system c₃,y₃,z₃) Obtaining coordinates (x, y, z) under a reference coordinate system through translation transformation as follows:

and finally, obtaining a point cloud coordinate under the reference coordinate system, wherein the height information of the point is a negative value of the x coordinate.

10. The three-dimensional lidar based loading material height and offset detection device of claim 8, wherein the car position determination module comprises a car end point determination sub-module and a car start point determination sub-module;

the carriage terminal point determining submodule is used for selecting the point cloud belonging to the last laser beam from the material point clouds under the reference coordinate system, and calculating the variance sigma of the x coordinate_end(ii) a If σ_end<th_end，th_endSetting a threshold value, wherein the threshold value indicates that the baffle at the rear end of the carriage of the section runs to a material scanning area and the train runs to the end point of the carriage;

the carriage starting point determining submodule is used for selecting point clouds belonging to a first laser beam and a second laser beam from material point clouds under a reference coordinate system, and calculating the variance of an x coordinate and recording the variance as sigma₁And σ₂If σ is₁<th₁And sigma₂>th₂，th₁Conforming to baffle morphology for point cloud changesThreshold of the feature, th₂If the point cloud change meets the threshold value of the material filling form characteristics, the baffle at the front end of the new carriage is judged to be about to enter the material scanning area, and the train runs to the starting point of the new carriage.

Technical Field

The invention relates to the field of detection, in particular to a method and a device for detecting the height and the unbalance loading of a loading material based on a three-dimensional laser radar.

Background

On a material loading line of a freight train, uniform distribution of materials in a carriage needs to be ensured as much as possible, and full automation of material loading is realized, so that loading control needs to be performed according to the material height and the real-time condition of unbalance loading. The currently adopted method is manual observation and estimation, and the filling of the materials is manually controlled according to the observation result. The method can cause large errors of observed real-time conditions of the materials and insufficient precision of control, and cause uneven distribution of the filled materials in the carriage. When materials are filled, a freight train stably runs on a rail, and the traditional weight measuring method and the wheel weight measuring method according to the stress change of the rail are difficult to accurately obtain the real-time data of the materials in a filling area. The illumination of a filling site is insufficient, the material and a carriage do not have distinct color difference, the detection is difficult to be carried out by adopting an image method, and the data error is extremely large based on the image method, so that accurate data cannot be obtained. The laser radar can detect the surface condition of the material by emitting a laser beam, has higher precision, but dust can be excited on the filling site, so that the radar detects partial error data, and because the data volume which can be obtained by the two-dimensional laser radar is small, if the error data is screened, the residual data volume can not truly reflect the actual condition of the material, and the accuracy of the result is seriously influenced. From the actual situation of the filling line, a suitable detection method is selected. In consideration of the factors, the three-dimensional laser radar is selected for detecting the height and the unbalance loading of the material, so that the advantages of the laser radar measuring method can be kept, and the influence of data errors on results can be reduced. This loading material height detection device simple structure, full automatic operation can judge the height of material fast accurately, as automatic material loading system's input data, realizes the full automatization that the material loaded, improves the precision of control, and convenient and practical uses manpower sparingly.

Disclosure of Invention

In view of the above, the invention provides a method and a device for automatically detecting the height and the unbalance loading of a loading material based on a three-dimensional laser radar, which can solve the problem of detecting the real-time height and the unbalance loading of the loading material.

In order to solve the above-mentioned technical problems, the present invention has been accomplished as described above.

A three-dimensional laser radar-based method for detecting height and unbalance loading of a loading material comprises the following steps:

firstly, fixedly and obliquely installing a three-dimensional laser radar, wherein the three-dimensional laser radar irradiates a material filling area;

step two, self-calibration of the installation attitude angle of the three-dimensional laser radar: calibrating a yaw angle and a pitch angle by taking the horizontal ground as a reference, and calibrating a roll angle by taking baffles on the left side and the right side of the carriage as a reference to obtain a conversion relation between a laser radar coordinate system and a reference coordinate system x, y and z; wherein, the yz plane of the reference coordinate system is parallel to the ground, and the x positive direction is vertical to the ground and faces downwards;

step three, extracting material point cloud: scanning the three-dimensional laser radar to obtain coordinates (x, y, z) of the three-dimensional laser radar point cloud under a reference coordinate system; screening carriage baffle point clouds according with the height of a carriage baffle from the three-dimensional laser point clouds to obtain a y-axis coordinate range corresponding to the position of the carriage baffle; screening out material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle;

step four, determining the position of the carriage: screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; the first laser beam is the beam with the largest three-dimensional laser radar pitch angle, and the last laser beam is the beam with the smallest three-dimensional laser radar pitch angle; determining whether the vehicle runs to the terminal point of the current carriage or not according to whether the point cloud height of the last laser beam, namely the x coordinate, meets the morphological characteristics of the carriage baffle or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics;

and step five, calculating the height and the unbalance loading of the material according to the point cloud of the material in the carriage range.

Preferably, the second step specifically includes:

coordinate system x of three-dimensional laser radar_Ly_Lz_LIs a right-hand coordinate system; establishing a new right-hand coordinate system which is recorded as a reference coordinate system by taking the ground as a yz plane, the positive direction of an x axis is vertical to the ground downwards, and the running direction of the vehicle is the positive direction of a z axis; the transformation from the lidar coordinate system to the reference coordinate system is performed in the following order:

first winding laser radar coordinate system y_LAxial through rotation matrix R_yTransforming to obtain an intermediate system a, and rewinding z of the intermediate system a₁Axial through rotation matrix R_zTransforming to obtain an intermediate system b, and then winding x of the intermediate system b₂Axial through rotation matrix R_xTransforming to obtain intermediate system c, and translating to transform matrix T_xObtaining a reference coordinate system;

(1) calibration of yaw angle and pitch angle by taking horizontal ground as reference

Scanning the ground by a three-dimensional laser radar, collecting ground point clouds, determining ground points by adopting a plane fitting method to the ground point clouds to obtain a ground plane equation Ax under a laser radar coordinate system_L+By_L+Cz_L+D＝0，A>0, wherein A, B, C, D are parameters of the ground plane equation, respectively; the height of the three-dimensional laser radar coordinate system from the origin to the ground is obtained through calculationThen the corresponding rotation matrix R_z、R_yAnd a translation transformation matrix T_xComprises the following steps:

wherein the yaw anglePitching angle

Point (x) in the lidar coordinate system_L,y_L,z_L) The coordinates (x) in the intermediate system b are obtained by the following rotation transformation₂,y₂,z₂)：

(2) Calibration of roll angle by using baffle plates on two sides of carriage as reference

The baffle of the carriage is vertical to the ground, and the height of the top of the baffle is h_b(ii) a The bottom of the empty carriage is parallel to the ground and has the height h₀(ii) a Under the intermediate system b, the x value of the ground point coordinate is l_groundThe x value of the highest point coordinate of the baffle is l_ground﹣h_b(ii) a In the radar laser point cloud obtained by scanning the carriage area by the three-dimensional laser radar, the coordinate x value and the coordinate l are satisfied_ground﹣h_bThe point with the difference value smaller than the threshold value is recorded as a top baffle threshold value point set, the point set is provided with two areas which respectively belong to the left baffle and the right baffle, and all the points of the two areas are respectively projected to y₂z₂Performing straight line fitting behind the plane to obtain the position of the top of the baffle plate at y₂z₂Equation A' y for two parallel straight lines of a plane₂+B'z₂+C'₁0 and A' y₂+B'z₂+C'₂＝0，A'>0; corresponding rotational transformation matrix R_xIs defined as follows:

wherein the roll angleOrDepending on the direction of travel of the train, the requirement is that the direction of travel of the freight train is z in the intermediate system c₃The shaft is in the positive direction; point (x) under the intermediate system b₂,y₂,z₂) The coordinates (x) in the intermediate system c are obtained by the following rotation transformation₃,y₃,z₃)：

Point (x) under the intermediate system c₃,y₃,z₃) Obtaining coordinates (x, y, z) under a reference coordinate system through translation transformation as follows:

and finally, obtaining a point cloud coordinate under the reference coordinate system, wherein the height information of the point is a negative value of the x coordinate.

Preferably, in the third step, the car baffle point cloud conforming to the height of the car baffle is screened from the three-dimensional laser point cloud, and the y-axis coordinate range corresponding to the car baffle position is obtained as follows:

under a reference coordinate system, the x value of the ground point coordinate is 0, and the height of the top of the baffle is h_bThe x value of the highest point coordinate of the baffle is-h_b(ii) a Screening out coordinates (x, y, z) from three-dimensional laser point cloud to meet conditions|x-(-h_b)|<th_rangeProjecting the point cloud to a yz plane to obtain a carriage baffle point set;

the concentrated points of the baffle points of the carriage are provided with two areas which are respectively the projection of the top point of the baffle on the left side of the carriage and the top point of the baffle on the right side of the carriage; selecting a point set of an area corresponding to a baffle on the left side of the carriage, and calculating to obtain the minimum value of y coordinates of all the points, which is recorded as y_max(ii) a Selecting a point set of an area corresponding to a baffle on the right side of the carriage, and calculating to obtain the maximum value of y coordinates of all the points, which is recorded as y_min。y_minAnd y_maxAnd forming a y-axis coordinate range corresponding to the carriage baffle position.

Preferably, in step three, screening out the material point cloud according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle:

screening all laser radar points under a reference coordinate system, wherein the coordinates (x, y, z) of the points meet the following conditions, namely the material point cloud:

h₀≤-x<h_max，y_min≤y≤y_max

wherein h is₀Is the height of the bottom of the empty carriage.

Preferably, in step four, the determining whether the vehicle drives to the end point of the current car according to whether the point cloud height of the last laser beam, i.e. the x coordinate, meets the car baffle morphological characteristics is as follows:

selecting point clouds belonging to the last laser beam from the material point clouds in the reference coordinate system, and calculating the variance sigma of the x coordinate_end(ii) a If σ_end<th_end，th_endAnd setting a threshold value, wherein the baffle at the rear end of the carriage of the train runs to the material scanning area, and the train runs to the end point of the carriage.

Preferably, in the fourth step, the determining whether the vehicle travels to the starting point of the next car according to whether the point cloud heights of the first and second laser beams respectively satisfy the car baffle morphological characteristic and the material filling morphological characteristic is as follows:

selecting the material point cloud belonging to the first bar in the reference coordinate systemRespectively calculating the variance of x coordinates of the point clouds of the laser beam and the second laser beam and recording the variance as sigma₁And σ₂If σ is₁<th₁And sigma₂>th₂，th₁Threshold value for point cloud change to conform to baffle morphology feature, th₂If the point cloud change meets the threshold value of the material filling form characteristics, the baffle at the front end of the new carriage is judged to be about to enter the material scanning area, and the train runs to the starting point of the new carriage.

Preferably, the step five is:

step 501: dividing the material point cloud into a left side and a right side relative to the train, and respectively calculating the two parts as follows:

screening out material point clouds of the unilateral carriage according to the y coordinate of the unilateral carriage baffle and the y coordinate of the carriage center line under a reference coordinate system;

dividing the material point cloud of the single-side compartment into M areas along the y axis; the three-dimensional laser radar has N laser beams; for each area, averaging the x coordinates of all point clouds of each laser beam of the N laser beams, and obtaining the material point cloud height corresponding to the N laser beams in each area; averaging the point cloud heights corresponding to the N line bundles to obtain a material point cloud height corresponding to the current area; averaging the point cloud heights of the materials in the M areas to obtain the material height of the carriage on one side;

step 502: calculating the height and the unbalance loading of the material:

the real-time material height h is:

wherein h is₀Height of empty carriage bottom, h_LThe material height of the left carriage, h_RThe material height of the right carriage;

the left and right offset p of the gravity center of the material is as follows:

p>at 0, the right side is heavy; p is a radical of<0, left-side weight, y_minAnd y_maxTwo y-axis coordinate limit values corresponding to the car baffle positions, respectively.

The invention provides a three-dimensional laser radar-based vehicle-loading material height and unbalance loading detection device, which comprises a three-dimensional laser radar and a processing device, wherein the three-dimensional laser radar is used for detecting the height of a vehicle-loading material; the processing device comprises a calibration module, a material point cloud extraction module, a carriage position determination module and a height and unbalance load calculation module;

the three-dimensional laser radar is fixedly and obliquely arranged above the material filling area and irradiates the material filling area;

the calibration module realizes the self-calibration of the installation attitude angle of the three-dimensional laser radar: calibrating a yaw angle and a pitch angle by taking the ground as a reference, and calibrating a roll angle by taking baffles on the left side and the right side of a carriage as a reference to obtain a conversion relation between a laser radar coordinate system and a reference coordinate system x, y and z; wherein, the yz plane of the reference coordinate system is parallel to the ground, and the x positive direction is vertical to the ground and faces downwards; scanning the three-dimensional laser radar to obtain three-dimensional laser radar point cloud, processing the three-dimensional laser radar point cloud by a calibration module to obtain coordinates (x, y, z) of the three-dimensional laser radar point cloud under a reference coordinate system, and sending the coordinates to a material point cloud extraction module;

the material point cloud extraction module is used for screening carriage baffle point clouds conforming to the height of the carriage baffle from the three-dimensional laser point clouds according to the coordinates (x, y, z) of the three-dimensional laser radar point clouds under the reference coordinate system, and obtaining a y-axis coordinate range corresponding to the position of the carriage baffle; screening material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle, and sending the material point clouds to a carriage position determining module;

the carriage position determining module is used for screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; the first laser beam is the beam with the largest three-dimensional laser radar pitch angle, and the last laser beam is the beam with the smallest three-dimensional laser radar pitch angle; determining whether the vehicle runs to the terminal of the current carriage or not according to whether the point cloud height of the last laser beam, namely the x coordinate, meets the morphological characteristics of the carriage or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics;

and the height and unbalance loading calculation module is used for calculating the height and unbalance loading of the material according to the material point cloud in the carriage range.

Preferably, the calibration mode of the calibration module is as follows:

coordinate system x of three-dimensional laser radar_Ly_Lz_LIs a right-hand coordinate system; establishing a new right-hand coordinate system which is recorded as a reference coordinate system by taking the ground as a yz plane, the positive direction of an x axis is vertical to the ground downwards, and the running direction of the vehicle is the positive direction of a z axis; the transformation from the lidar coordinate system to the reference coordinate system is performed in the following order:

first winding laser radar coordinate system y_LAxial through rotation matrix R_yTransforming to obtain an intermediate system a, and rewinding z of the intermediate system a₁Axial through rotation matrix R_zTransforming to obtain an intermediate system b, and then winding x of the intermediate system b₂Axial through rotation matrix R_xTransforming to obtain intermediate system c, and finally performing translation transformation T_xObtaining a reference coordinate system;

(1) calibration of yaw angle and pitch angle by taking horizontal ground as reference

Scanning the ground by a three-dimensional laser radar, collecting ground point clouds, determining ground points by adopting a plane fitting method to the ground point clouds to obtain a ground plane equation Ax under a laser radar coordinate system_L+By_L+Cz_L+D＝0，A>0, wherein A, B, C, D are parameters of the ground plane equation, respectively; the height of the three-dimensional laser radar coordinate system from the origin to the ground is obtained through calculationThen the corresponding rotation matrix R_z、R_yAnd a translation transformation matrix T_xComprises the following steps:

wherein the yaw anglePitching angle

Point (x) in the lidar coordinate system_L,y_L,z_L) The coordinates (x) in the intermediate system b are obtained by the following rotation transformation₂,y₂,z₂)：

(2) Calibration of roll angle by using baffle plates on two sides of carriage as reference

The baffle of the carriage is vertical to the ground, and the height of the top of the baffle is h_b(ii) a The bottom of the empty carriage is parallel to the ground and has the height h₀(ii) a Under the intermediate system b, the x value of the ground point coordinate is l_groundThe x value of the highest point coordinate of the baffle is l_ground﹣h_b(ii) a In the radar laser point cloud obtained by scanning the carriage area by the three-dimensional laser radar, the coordinate x value and the coordinate l are satisfied_ground﹣h_bThe point with the difference value smaller than the threshold value is recorded as a top baffle threshold value point set, the point set is provided with two areas which respectively belong to the left baffle and the right baffle, and all the points of the two areas are respectively projected to y₂z₂Performing straight line fitting behind the plane to obtain the position of the top of the baffle plate at y₂z₂Equation A' y for two parallel straight lines of a plane₂+B'z₂+C'₁0 and A' y₂+B'z₂+C'₂＝0，A'>0; corresponding rotational transformation matrix R_xIs defined as follows:

wherein the roll angleOrDepending on the direction of travel of the train, the requirement is that the direction of travel of the freight train is z in the intermediate system c₃The shaft is in the positive direction; point (x) under the intermediate system b₂,y₂,z₂) The coordinates (x) in the intermediate system c are obtained by the following rotation transformation₃,y₃,z₃)：

Point (x) under the intermediate system c₃,y₃,z₃) Obtaining coordinates (x, y, z) under a reference coordinate system through translation transformation as follows:

and finally, obtaining a point cloud coordinate under the reference coordinate system, wherein the height information of the point is a negative value of the x coordinate.

Preferably, the car position determination module comprises a car end point determination submodule and a car start point determination submodule;

the carriage terminal point determining submodule is used for selecting the point cloud belonging to the last laser beam from the material point clouds under the reference coordinate system, and calculating the variance sigma of the x coordinate_end(ii) a If σ_end<th_end，th_endIn order to set a threshold value, a baffle plate at the rear end of the carriage of the train runs to a material scanning area, and the train runs toAnd (4) finishing the carriage.

The carriage starting point determining submodule is used for selecting point clouds belonging to a first laser beam and a second laser beam from material point clouds under a reference coordinate system, and calculating the variance of an x coordinate and recording the variance as sigma₁And σ₂If σ is₁<th₁And sigma₂>th₂，th₁Threshold value for point cloud change to conform to baffle morphology feature, th₂If the point cloud change meets the threshold value of the material filling form characteristics, the baffle at the front end of the new carriage is judged to be about to enter the material scanning area, and the train runs to the starting point of the new carriage.

Has the advantages that:

(1) according to the method for detecting the height and the unbalance loading of the material, the height and the unbalance loading information of the material can be automatically calculated through the point cloud of the laser radar, and the data are sent to the controller, so that the full automation of material filling is realized. Manual judgment and control are omitted, and efficiency and precision are greatly improved. Compared with a two-dimensional laser radar method, the point cloud data volume obtained by the three-dimensional laser radar detection technology is larger, the calculated material height is more accurate, and the precision is higher. Compared with a weight measuring method and a stress change measuring method, the method is lower in cost and higher in efficiency.

(2) According to the method, the positions of left and right baffle plates of the carriage are screened according to the height of the baffle plate of the carriage, and then material point cloud is determined; and simultaneously, determining the material point cloud in the carriage position by utilizing the carriage baffle plate morphological characteristics and the material filling morphological characteristics, and further calculating the material height and the unbalance loading degree. According to the scheme, the height information of the carriage, the shape characteristics of the carriage baffle and the material filling shape characteristics are fully utilized, so that the automatic extraction of the material point cloud in the carriage can be realized, and the extraction scheme is reliable and effective.

(3) The invention provides a method for calibrating the installation angle of a three-dimensional laser radar. The algorithm can perform angle calibration by referring to any known plane, is not limited by the installation angle and whether the plane is a horizontal ground or not, and has wider application range; in addition, the angle calibration based on the self-calibration algorithm does not need to install an attitude sensor, so that the use of hardware is reduced, and the cost is saved.

(4) When the height and the unbalance loading of the material are calculated, the method divides areas, and each area is independently calculated and then combined. The material point cloud is obtained by scanning the three-dimensional laser radar due to uneven distribution of the material, the material point cloud is divided into areas when the point cloud densities of different areas are different, and each area is independently calculated and then combined, so that the actual heights of the materials at each position have the same weight when calculated, the influence of uneven distribution of the point cloud densities is eliminated, and the detection precision is improved.

Drawings

FIG. 1 is a schematic view of a three-dimensional lidar mounting and carriage;

FIG. 2 is a schematic diagram of three-dimensional lidar mounting attitude yaw and pitch angle calibration;

FIG. 3 is a plan view of a train carriage and a schematic diagram illustrating calibration of a roll angle of a three-dimensional lidar mounting attitude;

FIG. 4 is a schematic diagram of a three-dimensional point cloud of carriage materials;

fig. 5 is a schematic diagram of the device for detecting the height and the unbalance loading of the loading material based on the three-dimensional laser radar.

In the figure, 1-three-dimensional laser radar; 2-a material filling port; 3-front and rear baffle plates of the carriage; 4-left and right baffle plates of the carriage; 5-freight train carriages; 6-material detection area; 7-train running direction; 8-a three-dimensional lidar coordinate system; 9-a reference coordinate system; 10-the normal of the ground under the three-dimensional laser radar coordinate system; 11-installing a posture yaw angle of the three-dimensional laser radar; 12-three-dimensional laser radar installation attitude elevation angle; 13-three-dimensional laser radar installation attitude roll angle; 14-material surface point cloud in the carriage; 15-baffle area and gap between the cars.

Detailed Description

The invention is described in detail below with reference to the figures and examples. The unit of measurement for this embodiment is the international unit meter (m).

The invention relates to a three-dimensional laser radar-based vehicle-loading material height and unbalance loading detection scheme. After installation, the yaw, pitch and roll angles of the three-dimensional laser radar installation are measured by using the horizontal ground information and the carriage characteristic information and using a self-calibration algorithm, so that the three-dimensional laser radar point cloud is corrected. And calculating the relative distance between the laser radar and the ground through the point cloud reflected by the ground, thereby obtaining the installation height of the laser radar. When the loading operation is carried out, the laser radar point clouds are collected in real time, the boundaries of the two sides of the carriage are automatically detected, the point clouds belonging to materials are screened out, and whether the material loading of the carriage of the section is finished or not and whether the material loading of a new carriage of the section is started or not are judged. And finally, processing the point cloud of the material, calculating to obtain the height and unbalance loading information of the material, and sending the height and unbalance loading information to an automatic material filling system.

The invention provides a three-dimensional laser radar-based method for detecting the height and the unbalance loading of a loading material, which comprises the following steps of:

the method comprises the following steps: the three-dimensional laser radar 1 is fixedly and obliquely arranged in front of the material filling opening 2. In front of the lidar, i.e. x of the lidar coordinate system_LThe positive direction of the axis irradiates the material detection area 6.

Step two: three-dimensional laser radar installation attitude angle self-calibration

The three-dimensional lidar coordinate system 8 is a right-hand coordinate system. And establishing a new right-hand coordinate system right below the three-dimensional laser radar coordinate system 8 by taking the ground as a yz plane, the positive direction of the x axis is vertical to the ground downwards, and the running direction 7 of the freight train is the positive direction of the z axis, and recording the new right-hand coordinate system as a reference coordinate system 9. The transformation from the lidar coordinate system 8 to the reference coordinate system 9 is performed in the following order:

y around lidar coordinate system 8_LAxial through rotation matrix R_yTransforming to obtain an intermediate system a, and rewinding z of the intermediate system a₁Axial through rotation matrix R_zTransforming to obtain an intermediate system b, and then winding x of the intermediate system b₂Axial through rotation matrix R_xTransforming to obtain intermediate system c, and finally performing translation transformation T_xA reference coordinate system 9 is obtained.

(1) Yaw and pitch angle calibration

Using horizontal ground as referenceAnd (6) calibrating. The three-dimensional laser radar 1 scans the ground, collects ground point clouds, determines ground points for the point clouds by adopting a plane fitting method, and obtains a ground plane equation Ax under a laser radar coordinate system 8_L+By_L+Cz_L+D＝0(A>0). The height of the three-dimensional laser radar coordinate system 8 origin from the ground is obtained through calculationAnd calculating a yaw angle 11 and a pitch angle 12 of the installation attitude of the three-dimensional laser radar 1 according to a normal 10 of the ground plane. Then the corresponding rotation matrix R_z、R_yAnd a translation transformation matrix T_xComprises the following steps:

wherein the yaw anglePitching angle

Point (x) under lidar coordinate system 8_L,y_L,z_L) The coordinates (x) in the intermediate system b are obtained by the following rotation transformation₂,y₂,z₂)：

(2) Roll angle calibration

Using the left and right sides of the carriageThe baffle 4 is calibrated for reference. The baffle of the carriage is vertical to the ground, and the height of the top of the baffle is h_b(ii) a The bottom of the empty carriage is parallel to the ground and has the height h₀. Under the intermediate system b, the x value of the ground point coordinate is l_groundThe x value of the highest point coordinate of the baffle is l_ground﹣h_b. The three-dimensional laser radar 1 scans the radar point cloud obtained by the carriage area and meets the coordinate x value and the coordinate l_ground﹣h_bThe point with the difference value smaller than the threshold value is recorded as a top baffle threshold value point set, the point set is provided with two areas which respectively belong to the left baffle and the right baffle, and all the points of the two areas are respectively projected to y₂z₂Performing straight line fitting behind the plane to obtain the position of the top of the baffle plate at y₂z₂Equation A' y for two parallel straight lines of a plane₂+B'z₂+C'₁0 and A' y₂+B'z₂+C'₂＝0，A'>0. And calculating the roll angle 13 of the installation posture of the three-dimensional laser radar 1 according to a linear equation. The corresponding rotational transformation is defined as follows:

wherein the roll angleOrDepending on the train direction of travel 7, it is satisfied that in the intermediate system c the direction of travel 7 of the freight train is z₃The shaft is forward. Point (x) under the intermediate system b₂,y₂,z₂) The coordinates (x) in the intermediate system c are obtained by the following rotation transformation₃,y₃,z₃)：

Point (x) under the intermediate system c₃,y₃,z₃) Is translated byThe transformation yields the coordinates (x, y, z) under the reference coordinate system 9:

and finally, obtaining a point cloud coordinate under the reference coordinate system 9, wherein the height information of the point is a negative value of the x coordinate.

Step three: automatic determination of a material point cloud

The left baffle plate and the right baffle plate 4 of the carriage are vertical to the ground, and the height of the top of the baffle plate is h_b. The bottom of the empty carriage is parallel to the ground and has the height h₀. The height threshold of the material filling is h_maxThe height of any part of the material cannot be greater than the threshold value, and h_max<h_b. The above height information of the present embodiment is h_b＝2.8、h₀1.1 and h_max＝2.3。

The baffles 4 on two sides of the carriage are completely parallel, under a reference coordinate system 9, the x value of the ground point coordinate is 0, and the x value of the highest point coordinate of the baffles is-h_b. In the radar point cloud obtained by scanning the carriage area by the three-dimensional laser radar 1, the coordinates (x, y, z) of the points satisfy the following conditions:

|x-(-h_b)|<th_range，th_rangeis a threshold value

Threshold th selected in the present embodiment_range0.05. Projection onto the yz plane yields the point (y)_{b_0},z_{b_0}) And is marked as a carriage baffle point set:

the concentrated points of the baffle points of the carriage are provided with two areas which are respectively the projection of the top point of the baffle on the left side and the top point of the baffle on the right side of the carriage. Selecting a point set of an area corresponding to a baffle on the left side of the carriage, and calculating to obtain the minimum value of y coordinates of all the points, which is recorded as y_max(ii) a Selecting a point set of an area corresponding to a baffle on the right side of the carriage, and calculating to obtain the maximum value of y coordinates of all the points, which is recorded as y_min。

Screening all laser radar points under a reference coordinate system, wherein the coordinates (x, y, z) of the points satisfy the following conditions:

h₀≤-x<h_max，y_min≤y≤y_max

namely the three-dimensional laser radar point cloud of the material.

Step four: car start and end point detection

The train for filling materials consists of a plurality of train carriages 5, and a certain gap is formed between the two train carriages. When the materials are filled, the freight train carriage 5 runs stably. And detecting the front and rear baffle plates 3 of the carriage by using the characteristics of the materials and the carriage baffle plates, and determining the starting point and the ending point of the carriage. The height of the material is uneven, and the material has the characteristics of low left and right sides and high middle part; the barriers of the railway carriage are approximately in a plane and the height h of the top of the barrier_bHigher than the maximum height h of the material_max. The above height information of the present embodiment is h_b2.8 and h_max＝2.3。

The three-dimensional laser radar 1 measures the distance and orientation information of the target in the environment by emitting laser beams in a rotating manner at a plurality of pitch angles of 360 °, receiving the signal reflected by the target, and comparing the signal with the emitted signal. The first line beam with the largest pitch angle is a radar and irradiates the forefront of a material scanning area in the carriage; the last beam with the smallest pitch angle is the radar beam that illuminates the back of the material scanning zone in the car.

Selecting point clouds belonging to the last wire harness from the material point clouds in the reference coordinate system 9, wherein the coordinates are (x) respectively_i,y_i,z_i)(i＝1,2,…,n_end)，n_endThe number of point clouds belonging to the last line bundle in the material point clouds. Calculate the variance of the x coordinate:

wherein the content of the first and second substances,representing the mean of the x coordinates.

If σ_end<th_end，th_endAnd setting a threshold value, wherein the threshold value indicates that the baffle at the rear end of the carriage of the section runs to a material scanning area, the train runs to the end point of the carriage, and the calculation of the material related data is stopped.

Similarly, in the material point clouds in the reference coordinate system 9, the point clouds belonging to the first line beam are selected, and the coordinates are (x) respectively_i,y_i,z_i)(i＝1,2,…,n₁)，n₁The number of the point clouds belonging to the first line beam in the material point clouds is determined. Calculate the variance of the x coordinate:

selecting point clouds with coordinates (x) belonging to the second beam_i,y_i,z_i)(i＝1,2,…,n₂)，n₂The number of the point clouds belonging to the second line beam in the material point clouds is shown. Calculate the variance of the x coordinate:

if σ₁<th₁And sigma₂>th₂And indicating that the baffle at the front end of the new carriage is about to completely leave the material scanning area, and the train runs to the starting point of the new carriage, and the calculation of the material related data is restarted. th (h)₁And th₂To set the threshold. The material point cloud 14 in the carriage and the front and back baffle areas and the gap 15 between the carriages can be obtained through the processing. The threshold selected in this embodiment is th_end＝th₁＝th₂＝0.03。

Step five: real-time material height and offset calculation

The material point cloud is divided into a left side and a right side, and the two parts are respectively calculated.

Under the reference coordinate system 9, the point coordinates (x, y, z) of the left portion satisfy the following condition:

the point coordinates (x, y, z) of the right part satisfy the following condition:

for any part, the material heights appear to be unevenly distributed, but in the actual processing of the material, the heights throughout should be considered evenly. Therefore, the material point cloud is processed in a partitioning mode.

(1) Left hand portion material point cloud processing

For the material point cloud on the left side, the material point cloud is divided into M regions along the y axis, where M is 20 in this embodiment, and the M regions are:

the three-dimensional laser radar 1 is set to have N lines. For the ith area, n of the jth line beam of the three-dimensional laser radar_{Li_j}Processing the points, and calculating to obtain the height average value of the points of the jth wire harness of the ith area:

wherein, x_kRepresenting the height value of the kth point in the jth line.

Calculating each region to obtain N height values, and calculating the average value of the point cloud heights of the materials in the ith region:

the material height of the left part is:

the three-dimensional laser radar selected in the present embodiment is a 32-line laser radar, and N is 32.

(2) Material point cloud processing of right part

For the material point cloud on the right side, the material point cloud is divided into M regions along the y axis, where M is 20 in this embodiment, and the M regions are:

assume that the three-dimensional lidar 1 has a total of N beams. For the ith area, n of the jth line beam of the three-dimensional laser radar_{Ri_j}Processing the points, and calculating to obtain the height average value of the points of the jth wire harness of the ith area:

calculating each region to obtain N height values, and calculating the average value of the point cloud heights of the materials in the ith region:

the material height of the right part is:

the three-dimensional laser radar selected in the present embodiment is a 32-line laser radar, and N is 32.

(3) Calculation of material height and unbalance loading

The real-time material height is:

wherein h is₀The height of the empty car bottom, h in this embodiment₀＝1.1。

The offset about the material focus is:

when p >0, right-side heavy; when p <0, the left side is emphasized.

And sending the height and the unbalance loading data of the materials to a material filling system server to provide input data for the automatic material filling system.

In summary, the invention aims to achieve automatic detection of the height and the unbalance loading of the loading material, and provides an automatic detection device for the height and the unbalance loading of the loading material based on a three-dimensional laser radar and a working method thereof. The three-dimensional laser radar for detection can scan the surface of a material at a higher frequency by utilizing the self-calibration installation attitude angle of the ground and the carriage baffle. The coordinate transformation relation of the three-dimensional laser radar coordinate system is solved through the installation angle, and then the height and the unbalance loading condition of the material are calculated through the point cloud of the material, so that the automatic detection of the height and the unbalance loading of the loading material is realized. High efficiency and high precision. The detected material height and unbalance loading information are used as input data of the automatic material filling system, and full automation of material filling is realized.

The invention also provides a device for detecting the height and the unbalance loading of the loading material based on the three-dimensional laser radar, as shown in figure 5, the device comprises the three-dimensional laser radar and a processing device; the processing device comprises a calibration module, a material point cloud extraction module, a carriage position determination module and a height and unbalance load calculation module;

the three-dimensional laser radar is fixedly and obliquely arranged above the material filling area and irradiates the material filling area;

the calibration module realizes the self-calibration of the installation attitude angle of the three-dimensional laser radar: calibrating a yaw angle and a pitch angle by taking the ground as a reference, and calibrating a roll angle by taking baffles on the left side and the right side of a carriage as a reference to obtain a conversion relation between a laser radar coordinate system and a reference coordinate system x, y and z; wherein, the yz plane of the reference coordinate system is parallel to the ground, and the x positive direction is vertical to the ground and faces downwards; scanning the three-dimensional laser radar to obtain three-dimensional laser radar point cloud, processing the three-dimensional laser radar point cloud by a calibration module to obtain coordinates (x, y, z) of the three-dimensional laser radar point cloud under a reference coordinate system, and sending the coordinates to a material point cloud extraction module;

the material point cloud extraction module is used for screening carriage baffle point clouds which accord with the height and the aggregation degree of a carriage baffle from the three-dimensional laser point clouds according to the coordinates (x, y, z) of the three-dimensional laser radar point clouds under a reference coordinate system, and obtaining a y-axis coordinate range corresponding to the position of the carriage baffle; screening material point clouds according to the limit of the height of the carriage baffle and the y-axis coordinate range corresponding to the position of the carriage baffle, and sending the material point clouds to a carriage position determining module;

the carriage position determining module is used for screening out point clouds of a first laser beam, a second laser beam and a last laser beam from the material point clouds; the first laser beam is the beam with the largest three-dimensional laser radar pitch angle, and the last laser beam is the beam with the smallest three-dimensional laser radar pitch angle; determining whether the vehicle runs to the terminal point of the current carriage or not according to whether the point cloud height of the last laser beam, namely the x coordinate, meets the morphological characteristics of the carriage baffle or not; determining whether the vehicle drives to the starting point of the next carriage or not according to whether the point cloud heights of the first laser beam and the second laser beam respectively meet the carriage baffle morphological characteristics and the material filling morphological characteristics;

and the height and unbalance loading calculation module is used for calculating the height and unbalance loading of the material according to the material point cloud in the carriage range.

The calibration mode of the calibration module has already been described above, and is not described herein again.

The carriage position determining module comprises a carriage terminal point determining submodule and a carriage starting point determining submodule; the carriage end point determining submodule is used for selecting point clouds belonging to the last laser beam from the material point clouds under the reference coordinate system, and calculating the variance sigma of the x coordinate_end(ii) a If σ_end<th_end，th_endTo set the thresholdThe value indicates that the baffle at the rear end of the carriage of the section runs to the material scanning area and the train runs to the end point of the carriage;

a carriage starting point determining submodule for selecting point clouds belonging to the first laser beam and the second laser beam from the material point clouds under the reference coordinate system, and calculating the variance of the x coordinate and recording the variance as sigma₁And σ₂If σ is₁<th₁And sigma₂>th₂，th₁Threshold for point cloud changes to conform to baffle features, th₂If the point cloud change meets the threshold value of the material characteristics, the baffle at the front end of the new carriage is judged to be about to enter the material scanning area, and the train runs to the starting point of the new carriage.

The above embodiments only describe the design principle of the present invention, and the shapes and names of the components in the description may be different without limitation. Therefore, a person skilled in the art of the present invention can modify or substitute the technical solutions described in the foregoing embodiments; such modifications and substitutions do not depart from the spirit and scope of the present invention.