A positioning method and system for adjusting parameters of autonomous driving in underground mining environments

By synchronizing lidar, IMU, and odometer data in underground mining environments, constructing point cloud maps, and adjusting Kalman filter parameters, the positioning error and interruption problems in underground mining environments were solved, achieving higher accuracy and more stable autonomous driving positioning.

CN118670399BActive Publication Date: 2024-11-19LEIKE ZHITU (TAIAN) AUTOMOTIVE TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411146722.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-19
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

Existing technologies for autonomous driving positioning in underground mining environments suffer from problems such as large positioning errors, positioning interruptions, and limitations due to lighting conditions, which affect the safety and reliability of autonomous driving.

Method used

By acquiring data from lidar, IMU, and odometer for temporal and spatial synchronization, a point cloud map of the underground mining environment is constructed. Kalman filter fusion technology is then used, combined with corridor and intersection feature extraction, and Kalman filter parameters are adjusted to output the final positioning data.

Benefits of technology

It improves the positioning accuracy and stability in underground mining environments, enhances the safety and reliability of autonomous driving, and enables vehicles to drive safely and accurately in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118670399B_ABST
    Figure CN118670399B_ABST
Patent Text Reader

Abstract

The present invention proposes a positioning method and system for adjusting the parameters of autonomous driving in an underground mining environment, which belongs to the technical field of autonomous driving vehicle positioning. The method includes: respectively synchronizing the laser radar data, IMU data and odometer data of the acquired target vehicle in time and space; constructing a point cloud map of the underground mining environment; loading the point cloud map of the underground mining environment, using the synchronized data as prediction values, and using the laser radar data as observation values ​​to perform Kalman filter fusion to obtain initial positioning data; extracting features of corridors and forks in the underground mining environment, adjusting the parameters of the Kalman filter according to the extraction results, and outputting the final positioning data. Based on this method, a positioning system for adjusting the parameters of autonomous driving in an underground mining environment is also proposed. The present invention improves the safety and reliability of autonomous driving, allowing autonomous driving vehicles to travel safely and accurately in complex underground environments.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving vehicle positioning, and in particular relates to a positioning method and system for adjusting autonomous driving parameters in an underground mining environment. Background Art

[0002] In a mining environment, the application of autonomous driving technology can greatly improve the efficiency and safety of mining operations. However, the mining environment is complex, including long corridors, forks and other complex terrains, which puts higher requirements on autonomous driving technology. At the same time, positioning technology is also an important part of autonomous driving technology. It determines the position and posture of the autonomous driving vehicle in the environment, thus affecting the safety and accuracy of autonomous driving.

[0003] In existing technologies, laser SLAM (SIMUltaneous Localization and Mapping) is usually used for mapping and positioning. Laser SLAM is a technology that obtains environmental information through laser radar and simultaneously performs positioning and map construction. Secondly, the Kalman fusion of Odom (odometer), IMU (inertial measurement unit) and Lidar (laser radar) is also used to improve the accuracy of positioning. In addition, the fusion of vision and laser is also the mainstream positioning technology underground. However, there are still some problems in the actual application of existing technologies. First, the underground environment is complex, including various complex terrains such as long corridors and forks. This requires the autonomous driving vehicle to adapt the Kalman parameters according to different environments to ensure the accuracy and stability of positioning. Secondly, when facing complex underground environments, existing positioning technologies may have large positioning errors or positioning interruptions, and visual positioning is easily restricted by lighting conditions, which will affect the safety and reliability of autonomous driving. Therefore, how to achieve adaptive parameter adjustment of autonomous driving in underground mining environments and improve the accuracy and stability of positioning is an urgent problem to be solved. Summary of the invention

[0004] The present invention proposes a positioning method and system for adjusting parameters of autonomous driving in an underground mining environment. The present invention can achieve that in an underground mining environment, unmanned driving can adapt Kalman parameters according to different environments, thereby improving the accuracy and stability of positioning.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A positioning method for adjusting parameters of automatic driving in an underground mining environment comprises the following steps:

[0007] Acquire laser radar data, IMU data, and odometer data of the target vehicle, and perform time synchronization and space synchronization on the laser radar data, IMU data, and odometer data, respectively;

[0008] Use LiDAR to scan the underground mining environment, add IMU data and odometer data to the SLAM composition environment, and build a point cloud map of the underground mining environment;

[0009] Load the point cloud map of the underground mining environment, use the synchronized IMU data, odometer data and ultra-wideband positioning data as prediction values, and use the lidar data as observation values ​​to perform Kalman filter fusion to obtain the initial positioning data;

[0010] The features of corridors and intersections in the underground mining environment are extracted respectively, and the parameters of the Kalman filter are adjusted according to the results of feature extraction to output the final positioning data.

[0011] Furthermore, the process of performing time synchronization and space synchronization on the laser radar data, the IMU data and the odometer data respectively includes:

[0012] Calibrate the laser radar coordinate system and the vehicle body coordinate system; and calibrate the IMU coordinate system and the vehicle body coordinate system to achieve spatial synchronization of the laser radar data and the IMU data;

[0013] The ultra-wideband positioning data of the target vehicle is obtained, and the wheel speed meter data is parsed to obtain the odometer data, and the lidar data, IMU data, ultra-wideband positioning data and odometer data are time synchronized.

[0014] Furthermore, in the process of constructing a point cloud map of an underground mining environment: the point cloud data is first preprocessed, and then the preprocessed point cloud data is inter-frame registered and globally optimized to generate a point cloud map in a PCD format.

[0015] Furthermore, the process of preprocessing the point cloud data includes: denoising the point cloud data, downsampling, and filtering out the ground point cloud.

[0016] Furthermore, the specific process of using IMU data, odometer data and ultra-wideband positioning data as prediction values ​​is as follows:

[0017] The IMU factor is calculated based on the IMU three-axis acceleration and the distance the IMU moves on the three axes within the preset time.

[0018] Calculate the speed and acceleration of the wheel odometer, integrate within a preset time to obtain the distance and angle information required for the odometer data as the odometer factor;

[0019] The positioning information of broadband positioning data is used as the UWB optimization factor;

[0020] The IMU factor, odometer factor and uwb optimization factor are used as predicted values.

[0021] Furthermore, the speed and acceleration of the wheel odometer are calculated, and the distance and angle information required for the odometer data are obtained by integration within a preset time. The process as the odometer factor includes:

[0022] Determine the odometer model with the rear axle midpoint as the center, and set the left wheel speed to , the right wheel speed is set to , the odometer model speed is , the angular velocity is ;

[0023] ; (1)

[0024] ; (2)

[0025] in, is the turning radius; is the position of the wheel from the center of the chassis;

[0026] From formula (1) and formula (2), we can get:

[0027] ; (3)

[0028] (4).

[0029] Furthermore, the process of extracting corridor features in the underground mining environment includes: reading in wall point cloud data, dividing the wall point cloud data into blocks by ground projection, and generating a point cloud distribution matrix; then converting the point cloud distribution matrix into a grayscale image and performing binarization processing; finally, using Hough transform to perform straight line detection based on the binary image to complete the corridor feature extraction in the underground mining environment.

[0030] Furthermore, the process of extracting the features of the fork in the road in the underground mining environment includes:

[0031] By calculating the Euclidean distance between two laser points of the acquired target point cloud, it is determined whether the two laser points of the acquired target point cloud belong to the same cluster; if they are in the same cluster, the centroid, size and main direction of the cluster are extracted; the target point cloud is wall point cloud data or a fork in the road;

[0032] By calculating the distances between different clusters and the angles with the main directions of other clusters, it is determined whether there is a fork in the road.

[0033] Furthermore, the parameters of the Kalman filter are adjusted according to the result of feature extraction, and the process of outputting the final positioning data includes:

[0034] According to the characteristics of corridors and intersections in the underground mining environment, the predicted value of the Kalman filter and the weight of the process noise are adjusted to output more accurate final positioning data.

[0035] The present invention also proposes a positioning system for adjusting parameters of automatic driving in an underground mining environment, comprising a preprocessing module, a map building module, a first positioning module and a second positioning module;

[0036] The preprocessing module is used to obtain the laser radar data, IMU data and odometer data of the target vehicle, and synchronize the laser radar data, IMU data and odometer data in time and space respectively;

[0037] The map building module is used to use a laser radar to scan the underground mining environment, add IMU data and odometer data to the SLAM mapping environment, and build a point cloud map of the underground mining environment;

[0038] The first positioning module is used to load the point cloud map of the underground mining environment, use the synchronized IMU data, odometer data and ultra-wideband positioning data as prediction values, and use the lidar data as observation values ​​to perform Kalman filtering fusion to obtain initial positioning data;

[0039] The second positioning module is used to extract features of corridors and intersections in the underground mining environment, adjust the parameters of the Kalman filter according to the results of the feature extraction, and output the final positioning data.

[0040] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

[0041] The present invention proposes a positioning method and system for adjusting parameters of autonomous driving in an underground mining environment, which belongs to the technical field of autonomous driving vehicle positioning. The method comprises the following steps: obtaining laser radar data, IMU data and odometer data of a target vehicle, and performing time synchronization and space synchronization on the laser radar data, IMU data and odometer data respectively; scanning the underground mining environment with a laser radar, adding the IMU data and odometer data to a SLAM mapping environment, and constructing a point cloud map of the underground mining environment; loading the point cloud map of the underground mining environment, taking the synchronized IMU data, odometer data and ultra-wideband positioning data as predicted values, and taking the laser radar data as an observed value to perform Kalman filtering fusion to obtain initial positioning data;

[0042] The features of the corridors and forks in the underground mining environment are extracted respectively, and the parameters of the Kalman filter are adjusted according to the results of the feature extraction, and the final positioning data is output. Based on a positioning method for adjusting the parameters of automatic driving in an underground mining environment, a positioning system for adjusting the parameters of automatic driving in an underground mining environment is also proposed. The present invention can achieve that in an underground mining environment, unmanned driving can adapt the parameters of the Kalman according to different environments, thereby improving the accuracy and stability of positioning, thereby improving the safety and reliability of automatic driving, and enabling automatic driving vehicles to drive safely and accurately in complex underground environments.

[0043] The positioning method and system for adjusting automatic driving parameters in underground mining environments proposed in the present invention can also be applied to other complex environments and have strong versatility and practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a general flow chart of a positioning method for adjusting parameters of automatic driving in an underground mining environment proposed in Embodiment 1 of the present invention;

[0045] Figure 2 A schematic diagram of data preprocessing and construction of a point cloud map of an underground mining environment proposed in Embodiment 1 of the present invention;

[0046] Figure 3 This is a schematic diagram of data preprocessing in a positioning method for adjusting parameters of automatic driving in an underground mining environment proposed in Embodiment 1 of the present invention;

[0047] Figure 4 This is a schematic diagram of a corridor proposed in Example 1 of the present invention;

[0048] Figure 5 A schematic diagram of a fork in the road proposed in Embodiment 2 of the present invention;

[0049] Figure 6 This is a schematic diagram of a positioning system for adjusting automatic driving parameters in an underground mining environment proposed in Example 1 of the present invention. DETAILED DESCRIPTION

[0050] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.

[0051] Example 1

[0052] Embodiment 1 of the present invention proposes a positioning method for adjusting parameters of autonomous driving in a mine environment, which is used to solve the technical problems existing in the prior art in autonomous driving positioning in a mine environment. Figure 1 This is a general flow chart of a positioning method for adjusting automatic driving parameters in an underground mining environment proposed in Example 1 of the present invention.

[0053] In step S1, the laser radar data, IMU data and odometer data of the target vehicle are obtained, and the laser radar data, IMU data and odometer data are synchronized in time and space respectively.

[0054] Figure 3 This is a schematic diagram of data preprocessing in a positioning method for adjusting parameters of automatic driving in an underground mining environment proposed in Embodiment 1 of the present invention;

[0055] The specific process includes: calibrating the LiDAR coordinate system and the vehicle coordinate system; calibrating the IMU coordinate system and the vehicle coordinate system to achieve spatial synchronization of LiDAR data and IMU data; obtaining the ultra-wideband positioning data of the target vehicle, and parsing the wheel speed meter data to obtain the odometer data, and time synchronizing the LiDAR data, IMU data, ultra-wideband positioning data and odometer data.

[0056] In this application, the laser radar coordinate system and the vehicle body coordinate system are calibrated to obtain the position and posture relationship between the two, a rear wheel speedometer differential model is constructed, and the speedometer data is analyzed to obtain the odometer data;

[0057] Calibrate the IMU coordinate system and the vehicle coordinate system, obtain the speed and heading angle information of the unmanned vehicle through the odometer, obtain the attitude angle information of the unmanned vehicle through the IMU, and synchronize the obtained speed, heading angle information and attitude angle information;

[0058] At the same time, the positioning information of broadband positioning data is analyzed.

[0059] The data output by all sensors is time-synchronized, that is, the data is time-stamped.

[0060] In step S2, a laser radar is used to scan the underground mining environment, and IMU data and odometer data are added to the SLAM mapping environment to construct a point cloud map of the underground mining environment.

[0061] In the process of constructing the point cloud map of the underground mining environment: the point cloud data is first preprocessed, and then the inter-frame registration and global optimization of the preprocessed point cloud data are performed to generate a point cloud map in PCD format.

[0062] In this application, the autonomous driving vehicle is equipped with IMU and lidar to scan the underground environment, record and save the topic data (bag package) of the sensor, and use the cartographer algorithm to adapt and build the map according to the lidar, IMU, and odom. The point cloud data is pre-processed by voxel downsampling and denoising, and the inter-frame registration and global optimization of the point cloud data are performed to generate a point cloud map in PCD format.

[0063] The process of denoising point cloud data includes: the ranging accuracy of the laser radar varies with the distance. When the distance is 0.3-1m, the ranging accuracy is ±0.7cm, when the distance is 1-50m, the ranging accuracy is about 2cm, and after 50m, the ranging accuracy drops to about 5cm. At the same time, points that are too close will also have certain errors due to the influence of reflection. Therefore, before using the point cloud data, it is necessary to eliminate points that are too close and too far.

[0064] The downsampling process includes: the laser radar returns millions of points per second. If each point is processed, it will consume a lot of computing power. Therefore, it is necessary to downsample the returned laser points, that is, divide the three-dimensional space around the unmanned vehicle into cubes with a side length of a set value. At this time, all points projected into the cube are replaced by the coordinates of the center of the cube. In this way, the amount of data is greatly reduced and the computing efficiency is improved.

[0065] The process of filtering out the ground point cloud includes: ground point cloud filtering. When the unmanned vehicle is driving in a tunnel, due to factors such as water accumulation on the ground, the ground point cloud needs to be removed to eliminate the influence of the ground on the laser point cloud. This ground elimination algorithm first reduces the three-dimensional space of the point cloud to a two-dimensional plane, calculates the plane angle from each point to the positive x direction of the vehicle, and distinguishes ground points from non-ground points based on the angle, thereby eliminating ground points and retaining non-ground points.

[0066] Figure 2 This is a schematic diagram of data preprocessing and construction of a point cloud map of an underground mining environment proposed in Example 1 of the present invention; the laser radar data LIDAR, IMU data and odom data are respectively synchronized in time and space, the underground mining environment is scanned by laser radar, and the IMU data and odometer data are added to the SLAM composition environment to construct a point cloud map of the underground mining environment.

[0067] In step S3, the point cloud map of the underground mining environment is loaded, and the synchronized IMU data, odometer data and ultra-wideband positioning data are used as prediction values, and the lidar data is used as the observation value to perform Kalman filtering fusion to obtain the initial positioning data;

[0068] The specific process of using IMU data, odometer data and ultra-wideband positioning data as prediction values ​​is as follows: the moving distance of the IMU on the three axes within the preset time is calculated based on the IMU three-axis acceleration as the IMU factor;

[0069] Calculate the speed and acceleration of the wheel odometer, integrate the distance and angle information required for the odometer data within the preset time, and use it as the odometer factor; use the positioning information of the broadband positioning data as the uwb optimization factor; use the IMU factor, odometer factor, and uwb optimization factor as the predicted value.

[0070] In this application, the speed and acceleration of the wheel odometer are calculated, and the distance and angle information required for the odometer data are obtained by integration within a preset time. The process as the odometer factor includes: determining the odometer model with the midpoint of the rear axle as the center, and the left wheel speed is set to , the right wheel speed is set to , the odometer model speed is , the angular velocity is ;

[0071] ; (1)

[0072] ; (2)

[0073] in, is the turning radius; is the position of the wheel from the center of the chassis;

[0074] From formula (1) and formula (2), we can get:

[0075] ; (3)

[0076] (4).

[0077] Through the above formula, the speed and angular velocity of the wheel odometer can be calculated, and then the distance and angle information required for the odometer data can be obtained by accumulating the time integral in delta_t as the odometer factor.

[0078] The data of IMU, odom and uwb sensors are used as predicted values ​​(measurement noise), and lidar is used as observed values ​​(observation noise) for unscented Kalman filtering (UKF) fusion processing to calculate the output pose.

[0079] The Kalman filtering process includes:

[0080] First, define the state vector and observation value

[0081] ①、Define state vector and observation value

[0082] State Vector :

[0083] ;in, and is the position coordinate; is the attitude angle;

[0084] ②. Control input

[0085] ;in, is displacement; It’s a corner;

[0086] ③. Observation value

[0087] LiDAR scan matching: can provide observations of relative pose.

[0088] Odometer / IMU: Provides prediction data for motion models

[0089] Second, initialization

[0090] Initialize state estimate and the covariance matrix ;

[0091] Initialize the process noise covariance Q and the observation noise covariance R.

[0092] Generate sigma points and adjust the state vector Perform the transformation to generate a set of sigma points.

[0093] Prediction step: predict each sigma point; use the motion model (IMU / odometer / UWB) to update the state of each sigma point to obtain the predicted sigma point; calculate the predicted mean and variance.

[0094] Update step, use the LiDAR scan results as observations to generate observation sigma points; apply the observation model to the predicted sigma points to obtain the observed sigma points. Calculate the mean and covariance; Calculate the cross covariance between the state and observations: Calculate the Kalman gain: Update the state estimate and covariance matrix.

[0095] In each iteration, the prediction step and the update step are updated, and the odometer / IMU data / uwb and lidar data are continuously integrated to continuously improve the accuracy of pose estimation.

[0096] In step S4, features of corridors and intersections in the underground mining environment are extracted respectively, and the parameters of the Kalman filter are adjusted according to the results of the feature extraction to output the final positioning data.

[0097] The vertical surface features in the point cloud are used to detect the walls on the left and right sides. In a long corridor environment, the walls are usually parallel and continuous, while in a fork in the road environment, the walls may be bifurcated or interrupted.

[0098] Figure 4 This is a schematic diagram of the corridor proposed in Example 1 of the present invention; the process of extracting corridor features in an underground mining environment includes: reading in wall point cloud data, dividing the wall point cloud data into blocks by ground projection, and generating a point cloud distribution matrix; then converting the point cloud distribution matrix into a grayscale image and performing binarization processing; finally, using Hough transform based on the binary image to perform straight line detection to complete corridor feature extraction in the underground mining environment.

[0099] Read the wall point cloud data and store it as a matrix A with N rows and two columns. The first and second columns store the X and Y coordinates of the point cloud, and N represents the number of point clouds.

[0100] Using the X and Y coordinates of the N points stored in the matrix A, calculate the maximum and minimum values ​​in the X and Y directions, and determine the number of blocks;

[0101] Count the number of point clouds in each block area, generate a point cloud distribution matrix according to the number of point clouds in each block area, and obtain a point cloud distribution map;

[0102] The point cloud distribution matrix is ​​converted into a grayscale image. To obtain a more concise description of the point cloud distribution, a threshold is set to perform binary processing on the grayscale image. For areas above the threshold, the value is stored as 1, and the rest are stored as 0 to generate a point cloud distribution binary map.

[0103] The Hough transform algorithm is used to detect straight lines in binary images. The selection of the Hough transform peak point threshold is particularly important. Only points greater than the threshold are considered to represent straight lines, otherwise they are forks. The Hough matrix H is a matrix composed of peak points as elements. The peak point threshold defaults to 1 / 2 of the largest element in the matrix, that is, 0.5Hmax, and above, the corridor feature is obtained.

[0104] In a fork-in-the-road environment, there are usually multiple openings, i.e., directions in which corridors extend, and detecting these openings can help identify forks in the road.

[0105] Figure 5 A schematic diagram of a fork in the road proposed in Example 2 of the present invention; a process for extracting features of a fork in the road in an underground mining environment comprises: determining whether the two laser points of the acquired target point cloud belong to the same cluster by calculating the Euclidean distance between the two laser points of the acquired target point cloud; if they are located in the same cluster, extracting the centroid of the cluster, the size of the cluster, and the main direction of the cluster; the target point cloud is wall point cloud data or fork in the road point cloud data; determining whether there is a fork in the road by calculating the distance between different clusters and the angle with the main directions of other clusters.

[0106] Use the Euclidean clustering algorithm to cluster wall point cloud data or intersections and identify multiple opening areas.

[0107] Determine whether two laser points belong to the same cluster. and Point The Euclidean distance between is:

[0108]

[0109] Clustering parameters: Distance threshold ϵ: Set a maximum distance threshold. When the distance between two points is less than ϵ, they are considered to be in the same cluster. Minimum cluster size : Minimum number of points per cluster.

[0110] The process of extracting the centroid of a cluster is:

[0111] The formula for calculating the center of mass C is:

[0112] ;

[0113] Extract the width of the cluster ,high and depth They are:

[0114] ;

[0115] ;

[0116] ;

[0117] The main directions of the clusters are calculated by principal component analysis (PCA), which calculates the principal component direction matrix V and eigenvalue λ;

[0118] ;

[0119]

[0120] Where Cov(X) is the covariance matrix, Λ is the eigenvalue diagonal matrix, and V is the eigenvector matrix.

[0121] A fork in the road usually has multiple branches and can be detected by the following criteria: Inter-cluster distance: Detect the distance between different clusters to identify whether there is an intersection.

[0122] Set Cluster and The distance between them is:

[0123]

[0124] The angle with the main direction of other clusters, the angle θ is calculated as:

[0125] ;

[0126] in, and It is a cluster and Cluster The main direction vector of .

[0127] The presence of a fork in the road is determined based on the distance and angle changes. If there are N (N is 4) clusters within a small distance and their main directions are significantly different, it may indicate a fork in the road.

[0128] If the environmental feature is a corridor, the prediction noise is reduced and the process noise is increased, that is, the data of odom / IMU is trusted more; conversely, the prediction noise is increased and the process noise is reduced, that is, the posture obtained by lidar matching is trusted more.

[0129] The positioning method for adjusting the parameters of autonomous driving in an underground mining environment proposed in Example 1 of the present invention can realize that in an underground mining environment, unmanned driving can adapt the Kalman parameters according to different environments, thereby improving the positioning accuracy and stability, thereby improving the safety and reliability of autonomous driving, and enabling autonomous driving vehicles to drive safely and accurately in complex underground environments.

[0130] The positioning method for adjusting automatic driving parameters in underground mining environments proposed in Example 1 of the present invention can also be applied to other complex environments and has strong versatility and practicality.

[0131] Example 2

[0132] Based on the positioning method for adjusting the parameters of autonomous driving in a mine environment proposed in Example 1 of the present invention, Example 2 of the present invention further proposes a positioning system for adjusting the parameters of autonomous driving in a mine environment. Figure 6 A schematic diagram of a positioning system for adjusting parameters of autonomous driving in an underground mining environment proposed in Embodiment 1 of the present invention, comprising a preprocessing module, a map building module, a first positioning module and a second positioning module;

[0133] The preprocessing module is used to obtain the laser radar data, IMU data and odometer data of the target vehicle, and synchronize the laser radar data, IMU data and odometer data in time and space respectively;

[0134] The map building module is used to scan the underground mining environment using LiDAR, add IMU data and odometer data to the SLAM mapping environment, and build a point cloud map of the underground mining environment;

[0135] The first positioning module is used to load the point cloud map of the underground mining environment, use the synchronized IMU data, odometer data and ultra-wideband positioning data as prediction values, and use the lidar data as observation values ​​to perform Kalman filtering fusion to obtain the initial positioning data;

[0136] The second positioning module is used to extract features of corridors and intersections in the underground mining environment, adjust the parameters of the Kalman filter according to the results of the feature extraction, and output the final positioning data.

[0137] In the preprocessing module, the process of time synchronization and space synchronization of the lidar data, IMU data and odometer data respectively includes: calibrating the lidar coordinate system and the vehicle coordinate system; and calibrating the IMU coordinate system and the vehicle coordinate system to achieve spatial synchronization of the lidar data and the IMU data; obtaining the ultra-wideband positioning data of the target vehicle, and parsing the wheel speed meter data to obtain the odometer data, and time synchronization of the lidar data, IMU data, ultra-wideband positioning data and odometer data.

[0138] In the map building module, the point cloud data is first preprocessed, and then the inter-frame registration and global optimization of the preprocessed point cloud data are performed to generate a point cloud map in PCD format. The process of preprocessing the point cloud data includes: denoising, downsampling, and filtering out the ground point cloud.

[0139] In the first positioning module, the specific process of using IMU data, odometer data and ultra-wideband positioning data as prediction values ​​is as follows: the moving distance of the IMU on the three axes within the preset time is calculated according to the IMU three-axis acceleration as the IMU factor; the speed and acceleration of the wheel odometer are calculated, and the distance and angle information required for the odometer data are obtained by integrating within the preset time as the odometer factor; the positioning information of the broadband positioning data is used as the uwb optimization factor; the IMU factor, odometer factor and uwb optimization factor are used as prediction values.

[0140] The process of calculating the speed and acceleration of the wheel odometer and integrating the distance and angle information required for the odometer data within a preset time as the odometer factor includes:

[0141] Determine the odometer model with the rear axle midpoint as the center, and set the left wheel speed to , the right wheel speed is set to , the odometer model speed is , the angular velocity is ;

[0142] ; (1)

[0143] ; (2)

[0144] in, is the turning radius; is the position of the wheel from the center of the chassis;

[0145] From formula (1) and formula (2), we can get:

[0146] ; (3)

[0147] (4).

[0148] In the second positioning module, the process of extracting corridor features in the underground mining environment includes: reading in wall point cloud data, dividing the wall point cloud data into blocks by ground projection, and generating a point cloud distribution matrix; then converting the point cloud distribution matrix into a grayscale image and performing binarization processing; finally, using Hough transform based on the binary image to perform straight line detection to complete the corridor feature extraction in the underground mining environment.

[0149] The process of extracting the features of forks in underground mining environments includes: determining whether the two laser points of the acquired target point cloud belong to the same cluster by calculating the Euclidean distance between the two laser points of the acquired target point cloud; if they are in the same cluster, extracting the centroid, cluster size, and main direction of the cluster; the target point cloud is wall point cloud data or a fork in the road; and determining whether there is a fork in the road by calculating the distance between different clusters and the angle with the main direction of other clusters.

[0150] The process of adjusting the parameters of the Kalman filter according to the result of feature extraction and outputting the final positioning data includes: adjusting the predicted value of the Kalman filter and the weight of the process noise according to the corridor characteristics and the intersection characteristics in the underground mining environment, and outputting more accurate final positioning data.

[0151] A positioning system for adjusting automatic driving parameters in an underground mining environment proposed in Example 2 of the present invention can realize that in an underground mining environment, unmanned driving can adapt the Kalman parameters according to different environments, thereby improving the positioning accuracy and stability, thereby improving the safety and reliability of automatic driving, and enabling automatic driving vehicles to drive safely and accurately in complex underground environments.

[0152] The positioning system for adjusting automatic driving parameters in underground mining environments proposed in Example 2 of the present invention can also be applied to other complex environments and has strong versatility and practicality.

[0153] The description of the relevant parts of a positioning system for adjusting automatic driving parameters in an underground mining environment provided in Example 2 of the present application can be found in the detailed description of the corresponding parts of a positioning method for adjusting automatic driving parameters in an underground mining environment provided in Example 1 of the present application, and will not be repeated here.

[0154] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment that includes a series of elements are inherent to the elements. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment that includes the elements. In addition, the above-mentioned technical solution provided in the embodiment of the present application is consistent with the corresponding technical solution in the prior art in principle, and the part is not described in detail, so as not to repeat too much.

[0155] Although the above describes the specific implementation of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. For those skilled in the art, other different forms of modifications or deformations can be made on the basis of the above description. It is not necessary and impossible to list all the implementation methods here. On the basis of the technical solution of the present invention, various modifications or deformations that can be made by those skilled in the art without creative work are still within the scope of protection of the present invention.

Claims

1. A positioning method for adjusting parameters of automatic driving in an underground mining environment, characterized in that: The following steps are involved: Acquire laser radar data, IMU data, and odometer data of the target vehicle, and perform time synchronization and space synchronization on the laser radar data, IMU data, and odometer data, respectively; Use LiDAR to scan the underground mining environment, add IMU data and odometer data to the SLAM composition environment, and build a point cloud map of the underground mining environment; Load the point cloud map of the underground mining environment, use the synchronized IMU data, odometer data and ultra-wideband positioning data as prediction values, and use the lidar data as observation values ​​to perform Kalman filter fusion to obtain the initial positioning data; Extract features of corridors and intersections in underground mining environments, adjust Kalman filter parameters based on the results of feature extraction, and output final positioning data; The process of extracting corridor features in the underground mining environment includes: reading in wall point cloud data, dividing the wall point cloud data into blocks by ground projection, and generating a point cloud distribution matrix; then converting the point cloud distribution matrix into a grayscale image and performing binarization processing; finally, using Hough transform to perform line detection based on the binary image to complete corridor feature extraction in the underground mining environment; The process of extracting the features of forks in underground mining environments includes: determining whether the two laser points of the acquired target point cloud belong to the same cluster by calculating the Euclidean distance between the two laser points of the acquired target point cloud; if they are in the same cluster, extracting the centroid, size and main direction of the cluster; the target point cloud is wall point cloud data or fork point cloud data; determining whether there is a fork by calculating the distance between different clusters and the angle with the main direction of other clusters; The process of adjusting the parameters of the Kalman filter according to the result of feature extraction and outputting the final positioning data includes: adjusting the predicted value of the Kalman filter and the weight of the process noise according to the corridor characteristics and the fork characteristics in the underground mining environment, and outputting more accurate final positioning data; If the environmental feature is a corridor, the prediction noise is reduced and the process noise is increased, that is, the data of odom / IMU is trusted more; conversely, the prediction noise is increased and the process noise is reduced, that is, the posture obtained by lidar matching is trusted more.

2. A positioning method for adjusting parameters of automatic driving in an underground mining environment according to claim 1, characterized in that: The process of respectively performing time synchronization and space synchronization on the laser radar data, the IMU data, and the odometer data includes: Calibrate the laser radar coordinate system and the vehicle body coordinate system; and calibrate the IMU coordinate system and the vehicle body coordinate system to achieve spatial synchronization of the laser radar data and the IMU data; The ultra-wideband positioning data of the target vehicle is obtained, and the wheel speed meter data is parsed to obtain the odometer data, and the lidar data, IMU data, ultra-wideband positioning data and odometer data are time synchronized.

3. The positioning method for adjusting parameters of automatic driving in an underground mining environment according to claim 1, characterized in that: In the process of constructing the point cloud map of the underground mining environment: the point cloud data is first preprocessed, and then the inter-frame registration and global optimization of the preprocessed point cloud data are performed to generate a point cloud map in PCD format.

4. A positioning method for adjusting parameters of automatic driving in an underground mining environment according to claim 3, characterized in that: The process of preprocessing point cloud data includes: denoising, downsampling and filtering out ground point clouds.

5. The positioning method for adjusting parameters of automatic driving in an underground mining environment according to claim 1, characterized in that: The specific process of using IMU data, odometer data, and ultra-wideband positioning data as prediction values ​​is as follows: The IMU factor is calculated based on the IMU three-axis acceleration and the distance the IMU moves on the three axes within the preset time. Calculate the speed and acceleration of the wheel odometer, integrate within a preset time to obtain the distance and angle information required for the odometer data as the odometer factor; The positioning information of broadband positioning data is used as the UWB optimization factor; The IMU factor, odometer factor and uwb optimization factor are used as predicted values.

6. A positioning method for adjusting parameters of automatic driving in an underground mining environment according to claim 5, characterized in that: The process of calculating the speed and acceleration of the wheel odometer and integrating the distance and angle information required for the odometer data within a preset time as the odometer factor includes: Determine the odometer model with the rear axle midpoint as the center, and set the left wheel speed to , the right wheel speed is set to , the odometer model speed is , the angular velocity is ; ;(1) ;(2) in, is the turning radius; is the position of the wheel from the center of the chassis; From formula (1) and formula (2), we can get: ;(3) (4)。 7. A positioning system for adjusting parameters of automatic driving in an underground mining environment, used to execute a positioning method for adjusting parameters of automatic driving in an underground mining environment as claimed in any one of claims 1 to 6, characterized in that: It includes a pre-processing module, a map building module, a first positioning module and a second positioning module; The preprocessing module is used to obtain the laser radar data, IMU data and odometer data of the target vehicle, and synchronize the laser radar data, IMU data and odometer data in time and space respectively; The map building module is used to use a laser radar to scan the underground mining environment, add IMU data and odometer data to the SLAM mapping environment, and build a point cloud map of the underground mining environment; The first positioning module is used to load the point cloud map of the underground mining environment, use the synchronized IMU data, odometer data and ultra-wideband positioning data as prediction values, and use the lidar data as observation values ​​to perform Kalman filtering fusion to obtain initial positioning data; The second positioning module is used to extract features of corridors and intersections in the underground mining environment, adjust the parameters of the Kalman filter according to the results of the feature extraction, and output the final positioning data.