A mine rescue robot autonomous following and autonomous navigation method and a mine rescue robot

Abstract

The application discloses a kind of mining rescue robot autonomous following and autonomous navigation method and a kind of mining rescue robot, it is related to intelligent navigation technical field, the method includes obtaining the straight line distance and relative horizontal deflection angle of UWB label and UWB base station, using dynamic obstacle avoidance algorithm, control mining rescue robot autonomous following target to be followed;Obtain the advancing trajectory in the advancing process of mining rescue robot and underground multi-source data, and according to advancing trajectory and underground multi-source data, underground environment map is obtained by construction;Navigation instruction is obtained and after verifying the effectiveness of navigation instruction, using path generation algorithm and dynamic obstacle avoidance algorithm, control mining rescue robot is navigated from navigation starting point to navigation terminal point.The application can realize the accurate autonomous following and efficient autonomous navigation of mining rescue robot in complex underground environment, effectively improve rescue efficiency and safety, reduce the work intensity and risk of rescue personnel, improve the automation and intelligent level of rescue.

Description

Technical Field

[0001] This application relates to the field of intelligent navigation technology, and in particular to a method for autonomous following and autonomous navigation of a mining rescue robot and a mining rescue robot. Background Technology

[0002] Mining rescue robots are key intelligent equipment for emergency rescue and daily safety inspections in coal mines, and their operational performance directly affects the efficiency and safety of underground operations. Currently, the operation and control of mining rescue robots primarily relies on manual wireless remote control. Operators need to control the robot's movement, turning, braking, and other actions in real time via a remote control. However, underground coal mine roadways present challenges such as obstructed visibility, complex terrain, and significant environmental interference. This operating method places stringent demands on the operator's professional skills and sustained attention. Prolonged operation can easily lead to operator fatigue, resulting in misoperation, which not only reduces rescue efficiency but also easily damages the robot equipment, increasing safety risks in underground operations.

[0003] In terms of navigation technology, the navigation and positioning functions of existing mining robots generally rely on pre-deployed underground infrastructure such as communication base stations and positioning anchors to achieve the robot's position perception and path planning through an external positioning system. However, in accident scenarios such as gas explosions and roof collapses in coal mines, the underground communication and positioning infrastructure is easily damaged and rendered ineffective, making existing navigation technologies completely unusable. The robot loses its positioning and navigation capabilities, making it difficult to adapt to the complex working conditions of underground emergency rescue and failing to meet the actual needs of accident rescue.

[0004] Therefore, improving the automation level and environmental adaptability of mining rescue robots has become an urgent technical problem to be solved in this field. Summary of the Invention

[0005] The purpose of this application is to provide a method for autonomous following and autonomous navigation of a mining rescue robot and a mining rescue robot, which can achieve accurate autonomous following and efficient autonomous navigation of the mining rescue robot in complex underground environments.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] In the first aspect, this application provides a method for autonomous following and autonomous navigation of a mining rescue robot, including the following steps.

[0008] The system obtains the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, and uses a dynamic obstacle avoidance algorithm to control the mining rescue robot to autonomously follow the target to be followed; the UWB tag is set on the target to be followed, and the UWB base station is set on the mining rescue robot.

[0009] The system acquires the trajectory and multi-source data from underground sources during the process of a mining rescue robot following a target, and constructs an underground environment map based on the trajectory and multi-source data. The underground multi-source data includes lidar point cloud data, IMU three-axis attitude data, wheel speed meter mileage data, and camera visual semantic data.

[0010] After obtaining navigation instructions and verifying their validity, a path generation algorithm and a dynamic obstacle avoidance algorithm are used to control the mining rescue robot to automatically navigate from the navigation starting point to the navigation endpoint; the navigation instructions include the navigation starting point and the navigation endpoint.

[0011] Secondly, this application provides a mining rescue robot, including: a robot body and a controller, wherein the controller is used to implement the above-mentioned method for autonomous following and autonomous navigation of a mining rescue robot.

[0012] According to the specific embodiments provided in this application, this application has the following technical effects.

[0013] This application provides a method for autonomous following and navigation of a mining rescue robot, and a mining rescue robot itself. By acquiring the straight-line distance and relative horizontal deflection angle between a UWB tag and a UWB base station, and employing a dynamic obstacle avoidance algorithm, the mining rescue robot is controlled to autonomously follow the target to be followed, achieving autonomous following in a signal-free underground environment. By analyzing the movement trajectory of the mining rescue robot while following the target and using multi-source data from underground sources, an underground environment map is constructed based on the movement trajectory and multi-source data, enabling map construction and updating in complex underground environments, providing a foundation for autonomous navigation of the rescue robot. By receiving navigation commands and employing path generation and dynamic obstacle avoidance algorithms, the mining rescue robot is controlled to automatically navigate from the navigation starting point to the navigation endpoint, achieving autonomous navigation of the rescue robot in complex underground environments. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is an application environment diagram of an autonomous following and autonomous navigation method for a mining rescue robot according to an embodiment of this application.

[0016] Figure 2 This is a flowchart illustrating a method for autonomous following and navigation of a mining rescue robot, provided as an embodiment of this application.

[0017] Figure 3 This is a schematic diagram of the autonomous following and autonomous navigation application process of a mining rescue robot, provided as another embodiment of this application.

[0018] Figure 4 This is a schematic diagram of the functional modules of a mining rescue robot provided in one embodiment of this application.

[0019] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] This application provides an embodiment of a method for autonomous following and navigation of a mining rescue robot, which can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up separately, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the acquired straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station to server 104. After receiving the acquired straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, server 104 uses a dynamic obstacle avoidance algorithm to control the mining rescue robot to autonomously follow the target. It acquires the movement trajectory and underground multi-source data of the mining rescue robot during its movement while following the target, and constructs an underground environment map based on the movement trajectory and underground multi-source data. After acquiring navigation commands and verifying their validity, it uses a path generation algorithm and a dynamic obstacle avoidance algorithm to control the mining rescue robot to automatically navigate from the navigation starting point to the navigation endpoint. Server 104 can feed back the data generated by the mining rescue robot during autonomous following and navigation to terminal 102. Furthermore, in some embodiments, the autonomous following and navigation method of the mining rescue robot can also be implemented independently by server 104 or terminal 102. For example, terminal 102 can directly process the straight-line distance and relative horizontal deflection angle between the acquired UWB tag and the UWB base station, or server 104 can retrieve the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station from the data storage system and process the acquired straight-line distance and relative horizontal deflection angle.

[0023] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0024] In one exemplary embodiment, such as Figure 2 As shown, a method for autonomous following and navigation of a mining rescue robot is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps A1 to A3.

[0025] Step A1: Obtain the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, and use a dynamic obstacle avoidance algorithm to control the mining rescue robot to autonomously follow the target to be followed; the UWB tag is set on the target to be followed, and the UWB base station is set on the mining rescue robot.

[0026] In this embodiment, the straight-line distance between the UWB tag and the UWB base station is measured using the Time-of-Flight (TOF) method; the relative horizontal deflection angle between the UWB tag and the UWB base station is measured using the phase difference method. The target to be followed establishes point-to-point wireless communication with the UWB tag on the robot's UWB base station, thereby achieving clock synchronization. After detecting a valid ID signal from the same UWB tag for five consecutive frames, it is identified as the target to be followed, and the autonomous following mode is triggered.

[0027] The mining rescue robot specifically includes: an explosion-proof main body, a power battery and battery management system, a drive-by-wire system, a steering-by-wire system, a braking-by-wire system, and a chassis controller. It also includes headlights, an audible and visual alarm device, a remote control, environmental sensors, wheel speedometers, a UWB base station, an intrinsically safe lidar, an intrinsically safe industrial night vision camera, a nine-axis IMU, an intrinsically safe touchscreen, and ultrasonic radar.

[0028] The robot's main structure provides the mounting foundation for all functional components and reserves standardized mounting interfaces for sensors and actuators. The explosion-proof power battery and battery management system (BMS) provide stable power support for the entire robot system. The BMS monitors battery status, provides overcharge and over-discharge protection, and controls explosion-proof heat dissipation to ensure safe battery use underground. The drive-by-wire / steering / braking system enables the robot's forward and backward movement, stepless steering, and emergency / smooth braking functions, respectively, receiving and executing commands from the chassis controller. The headlights and audible / visual alarm devices provide underground working lighting, obstacle warnings, and robot malfunction alarms. The remote control... An intrinsically safe wireless remote control module is used to provide an operating terminal for manual remote control mode, enabling emergency operations when autonomous mode fails. Environmental sensors are explosion-proof gas sensors that monitor the concentration of gases such as oxygen, methane, and carbon monoxide in the well in real time, triggering audible and visual alarms when concentrations exceed limits. The chassis controller integrates underlying control software, responsible for executing and controlling the robot's walking, steering, braking, lighting, and audible and visual alarms. It transmits commands and status information bidirectionally with the controller via CAN / Ethernet. Wheel speed meters calculate the robot's speed, distance traveled, and mileage by measuring wheel speeds, providing basic mileage data for positioning and navigation. The controller, as the core computing unit of the system, uses a high-performance embedded controller responsible for the acquisition, preprocessing, fusion, and processing of all environmental perception data. It runs core algorithms such as laser and vision fusion, path planning, and decision control, sending action commands to the chassis controller via CAN / Ethernet and receiving execution status feedback from the chassis controller to achieve closed-loop control.

[0029] UWB Following Module: Composed of a UWB base station on the robot and a UWB tag on the target to be followed, it identifies and measures the distance and relative angle between the target and the robot in real time, transmitting the raw data to the controller; LiDAR: Employs explosion-proof solid-state LiDAR to collect point cloud data of the robot's surrounding environment, acquiring the position, distance, and shape information of obstacles and tunnel contours, participating in high-precision map construction and navigation positioning matching; Camera: Uses intrinsically safe industrial night vision cameras to recognize visual features such as underground traffic lights, traffic signs, and tunnel corners, providing environmental semantic information for path planning; IMU (Inertial Measurement Unit) Measurement Unit: A nine-axis inertial measurement unit (three-axis gyroscope + three-axis accelerometer + three-axis magnetometer) to measure the robot's three-axis attitude angles, acceleration, and angular velocity, enabling robot attitude correction and suppressing the cumulative error of the wheel speedometer; Intrinsically Safe Touchscreen: The robot's human-machine interface, displaying the robot's operating status, environmental data, path information, and high-precision map in real time, supporting manual touch input of navigation start and end points to trigger autonomous navigation mode; Ultrasonic Radar: Enables the robot to detect obstacles at close range (≤5m), complementing the LiDAR in perception and solving the problem of insufficient LiDAR perception of low and transparent obstacles.

[0030] Step A2: Acquire the movement trajectory and underground multi-source data of the mining rescue robot during its journey to follow the target, and construct an underground environment map based on the movement trajectory and underground multi-source data; the underground multi-source data includes lidar point cloud data, IMU three-axis attitude data, wheel speed meter mileage data, and camera visual semantic data.

[0031] Step A3: After obtaining navigation instructions and verifying their validity, a path generation algorithm and a dynamic obstacle avoidance algorithm are used to control the mining rescue robot to automatically navigate from the navigation starting point to the navigation endpoint. The navigation instructions include the navigation starting point and the navigation endpoint. The path generation algorithm is based on a high-precision raster map, introducing cost factors such as tunnel width, corner curvature, and obstacle distance to plan the globally optimal path from the starting point to the endpoint, and outputting the global path represented by a sequence of waypoint coordinates. The specific expression is as follows.

[0032] .

[0033] in, Let n be the total cost. The actual cost from the starting point to node n. Let n be the heuristic cost of the Manhattan distance from the endpoint. , , These are the costs associated with the tunnel width, corner curvature, and obstacle distance, respectively. , and The cost weight coefficient is used. Based on the global path, the dynamic obstacle avoidance algorithm uses the DWA dynamic window method to construct a dynamic speed window for the mining rescue robot. It evaluates the cost of each speed sample within the window and selects the optimal speed sample as the real-time driving command. The formula for the dynamic speed window of the mining rescue robot is as follows.

[0034] .

[0035] in, V This is a dynamic speed window for mining rescue robots. This is the minimum linear velocity. This represents the maximum linear velocity. This is the minimum value of the angular velocity. This represents the maximum value of the angular velocity.

[0036] In this embodiment, when the distance between the robot's precise location and the navigation endpoint is less than or equal to the navigation accuracy threshold, the controller determines that the endpoint has been reached, issues a braking command to bring the robot to a smooth stop, triggers an audible and visual alarm, and displays "Navigation Complete" on the touchscreen, thus completing the autonomous navigation task. If abnormalities such as excessive gas levels or sensor malfunctions are detected during navigation, the system automatically switches to remote control mode and issues an alarm. Preferably, the navigation accuracy threshold is 0.2 meters.

[0037] Implementing steps A1 to A3 above enables the mining rescue robot to achieve precise autonomous following and efficient autonomous navigation in complex underground environments, effectively improving rescue efficiency and safety, reducing the workload and risks for rescue personnel, and enhancing the automation and intelligence level of rescue operations. Furthermore, this application can also improve environmental perception capabilities by fusing multi-source sensor data. LiDAR provides three-dimensional spatial information, IMU reflects motion status, wheel speedometer calculates distance traveled, and camera identifies environmental features. Multi-source data fusion can construct an accurate underground map, providing a data foundation for the robot's autonomous following and navigation. The dynamic obstacle avoidance algorithm analyzes obstacle characteristics and robot motion characteristics to implement intelligent obstacle avoidance strategies, ensuring safe and efficient obstacle avoidance. The path generation algorithm combines dynamic environmental factors to generate safe and fast paths.

[0038] In another exemplary embodiment of this application, in order to automatically plan the real-time following path of the mining rescue robot, the original ranging and angular measurement data can be obtained by acquiring the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, thereby obtaining the relative coordinates of the UWB tag in the robot coordinate system. Then, based on the relative coordinates of the UWB tag in the robot coordinate system, the coordinates of the UWB tag in the global coordinate system are obtained. Finally, based on the coordinates of the UWB tag in the global coordinate system and obstacle perception data, a dynamic obstacle avoidance algorithm is used to automatically plan the real-time following path of the mining rescue robot. In this case, step A1 is replaced by steps A11 to A16.

[0039] Step A11: Obtain the straight-line distance between the UWB tag and the UWB base station to obtain the raw ranging data.

[0040] Step A12: Obtain the relative horizontal deflection angle between the UWB tag and the UWB base station to obtain the raw angle measurement data.

[0041] Step A13: Calculate the relative coordinates of the UWB tag in the robot coordinate system based on the original ranging data and the original angle measurement data; the origin of the robot coordinate system is the geometric center of the mining rescue robot.

[0042] Step A14: Based on the IMU attitude data and rotation transformation matrix, convert the relative coordinates of the UWB tag in the robot coordinate system to the coordinates of the UWB tag in the global coordinate system.

[0043] Step A15: Based on the coordinates of the UWB tag in the global coordinate system and obstacle perception data, a dynamic obstacle avoidance algorithm is used to automatically plan the real-time following path of the mining rescue robot; the obstacle perception data is acquired through intrinsically safe lidar and ultrasonic radar.

[0044] In this embodiment, the real-time position of the UWB tag in the robot coordinate system is converted into the coordinates of the UWB tag in the global coordinate system based on the IMU attitude data and the rotation transformation matrix. The rotation transformation matrix is ​​shown in the following formula.

[0045] .

[0046] in, Let be the rotation transformation matrix. γ The heading angle of the mining rescue robot as measured by the IMU.

[0047] The relative coordinates of the UWB tag in the robot coordinate system are calculated using the following formula.

[0048] .

[0049] The relative coordinates of the UWB tag in the robot coordinate system are: d This is the original distance measurement; θ This is angular measurement data.

[0050] In another exemplary embodiment of this application, step A14 is replaced by steps B1 to B3.

[0051] Step B1: By combining the coordinates of UWB tags in the global coordinate system and obstacle perception data with a dynamic obstacle avoidance algorithm, a real-time following path for the mining rescue robot is planned.

[0052] Step B2: Send speed and steering angle commands to the chassis controller based on the real-time following path of the mining rescue robot.

[0053] Step B3: Receive instructions from the chassis controller to control the drive-by-wire and steering system to perform following operations.

[0054] In this embodiment, the real-time following path of the mining rescue robot is set with a safe following distance of 1.5-3m and a following deviation angle threshold of ±10°. Speed ​​and steering angle commands are sent to the chassis controller. The chassis controller receives the commands and controls the drive-by-wire and steering system to perform following actions. If the target exceeds the deviation angle threshold, a steering action is performed. If an obstacle is detected, a temporary obstacle avoidance path is planned. After obstacle avoidance, the robot returns to the original following path. The chassis controller controls the drive-by-wire system to perform navigation actions and simultaneously feeds back the actual execution status of the mining rescue robot to the controller. The controller calculates the command deviation. If the deviation exceeds a first set deviation threshold, the commands are adjusted in real time. The actual execution status includes actual speed and actual angular velocity. The command deviation includes speed command deviation and angular velocity command deviation. The formula for calculating the command deviation is as follows.

[0055] .

[0056] .

[0057] in, For speed command deviation, For actual speed, To achieve the optimal speed; For angular velocity command deviation, This is the actual angular velocity. This is the optimal angular velocity.

[0058] In another exemplary embodiment of this application, step A2 is replaced by steps A21 to A23.

[0059] Step A21: Based on the multi-source data from the integrated intrinsically safe lidar, nine-axis IMU, and wheel speedometer, and the real-time following path of the mining rescue robot, acquire and record in real time the trajectory of the mining rescue robot as it follows the target to be followed; the trajectory is stored in the form of waypoint coordinate sequence, including waypoint global coordinates, travel speed, IMU attitude angle, and acquisition timestamp.

[0060] Step A22: Acquire multi-source data underground during the process of the mining rescue robot following the target being followed by an intrinsically safe lidar, a nine-axis IMU, a wheel speed gauge, and an intrinsically safe industrial night vision camera.

[0061] Step A23: Based on the trajectory of the mining rescue robot and multi-source data underground, construct an underground environment map and store the underground environment map in binary format.

[0062] In another exemplary embodiment of this application, step A21 is replaced by steps C1 to C3.

[0063] Step C1: The particle filter algorithm is used to fuse data from the lidar, nine-axis IMU, and wheel speedometer to achieve precise positioning. The fused odometer data from the wheel speedometer and nine-axis IMU is used as the coarse positioning result and serves as the initial particle set for the particle filter.

[0064] Step C2: The real-time point cloud of the LiDAR is registered with the fused high-precision map point cloud using the ICP algorithm to obtain the pose observation value.

[0065] Step C3 involves sampling, weighting, and resampling the particle set based on the observed values ​​to output the precise position of the mining rescue robot in the global coordinate system. The trajectory is obtained. The origin of the global coordinate system is the robot's initial starting position.

[0066] In this embodiment, the objective function of the ICP iterative nearest point algorithm is:

[0067] .

[0068] in, For point cloud registration error, For the current frame's point cloud data, For the previous frame of point cloud data, R Let be a rotation matrix. t It is a translation matrix.

[0069] The particle weight update formula is as follows.

[0070] .

[0071] in, For the i-th particle in t Weight of time, For the i-th particle in t The weight at time -1 Let be the likelihood probability of the observation model.

[0072] In another embodiment of this application, step A22 is replaced by steps D1 to D5.

[0073] Step D1: Collect point cloud data of the environment surrounding the mining rescue robot using an intrinsically safe lidar to obtain lidar point cloud data.

[0074] Step D2: Measure the three-axis attitude angles, accelerations, and angular velocities of the mining rescue robot using a nine-axis IMU to obtain the IMU's three-axis attitude data.

[0075] Step D3: Measure the wheel speed using a wheel speed meter, and calculate the driving speed, driving distance, and mileage information of the mining rescue robot based on the wheel speed to obtain the mileage data from the wheel speed meter.

[0076] Step D4: Visual features are identified using an intrinsically safe industrial night vision camera to obtain camera visual semantic data; the visual features include underground traffic lights, traffic signs, and tunnel corners.

[0077] Step D5 involves using statistical filtering and voxel downsampling to denoise the lidar point cloud data, using complementary filtering to suppress noise in the IMU three-axis attitude data, and using timestamp interpolation to achieve time synchronization of multi-sensor data to obtain downhole multi-source data. The multi-sensor data specifically includes lidar point cloud data, IMU three-axis attitude data, wheel speed meter mileage data, and camera visual semantic data.

[0078] In another exemplary embodiment of this application, step A23 is replaced by steps E1 to E2.

[0079] Step E1: Graph optimization algorithm is used to optimize the backend of the map. The waypoints of the mining rescue robot are used as nodes of the graph, and the pose constraints between waypoints are used as edges of the map. The global cumulative deviation of the map is corrected by minimizing the overall error function through the Gauss-Newton method. At the same time, the camera visual semantic data is associated with the corresponding location on the map to realize semantic annotation of the map.

[0080] Step E2 involves fusing the LiDAR point cloud map and the grid map, and using the occupancy grid method to calculate the obstacle occupancy probability of each grid cell. The grid cells are then divided into three categories: occupied, vacant, and unknown, resulting in a downhole environment map. This downhole environment map is stored in binary format in the controller's solid-state storage, supporting real-time updates and visualization on an intrinsically safe touchscreen. The calculation formula for the occupancy grid method is as follows.

[0081] .

[0082] in, Let be the posterior probability that a grid cell is occupied by an obstacle. For the probability of the observation model, For prior probability, This represents the global probability of the observed data.

[0083] In this embodiment, the grid is divided into three categories according to probability: occupied (≥0.7), idle (≤0.3), and unknown (0.3~0.7).

[0084] In another embodiment of this application, such as Figure 3 As shown, a process for autonomous following and autonomous navigation of a mining rescue robot includes steps F1 to F6.

[0085] Step F1: The operator puts UWB tags on the rescuers and places the mining rescue robot at the starting point of the underground operation.

[0086] Step F2: After a valid target is detected, the autonomous following mode is triggered.

[0087] In step F3, the mining rescue robot follows the rescuers along the underground tunnel, simultaneously recording the movement trajectory and constructing an underground environment map, which is then displayed in real time on the touchscreen.

[0088] Step F4: After arriving at the designated work point, the rescue personnel use the touchscreen to select the work point as the navigation starting point and the well exit as the navigation endpoint on the downhole environment map, and then input navigation commands.

[0089] Step F5: After verifying the validity of the command, the autonomous navigation mode is triggered, and a path is planned to control the mining rescue robot to autonomously travel along the planned path to the underground exit.

[0090] In step F6, after the mining rescue robot reaches the exit, it comes to a smooth stop and triggers an alarm, completing its autonomous navigation task.

[0091] This application also provides an application scenario in which the above-mentioned autonomous following and autonomous navigation method for mining rescue robots is applied. Specifically, the autonomous following and autonomous navigation method for mining rescue robots provided in this embodiment can be applied in emergency rescue operations for coal mine accidents. A complete rescue process typically includes the following steps: First, the rescue robot is deployed underground from the ground command center and navigates to the accident area through a pre-set map or autonomous exploration; second, in the accident area, the robot needs to follow the rescue personnel or autonomously search for survivors, while simultaneously avoiding obstacles and planning paths in real time; finally, the robot transmits the collected environmental data, video images, and other information back to the ground command center to support rescue decision-making. Among these, the autonomous following and autonomous navigation method for mining rescue robots provided in this application belongs to the second step of the above process, that is, the robot autonomously follows and navigates in the accident area to ensure that it can accurately follow the rescue personnel or autonomously plan paths, thereby improving rescue efficiency and safety. In addition, this method is also applicable to daily safety inspection scenarios, such as when a robot departs from a charging station, autonomously navigates to various inspection points, completes data collection, and returns; autonomous navigation is a key link in this process.

[0092] Based on the same inventive concept, this application also provides a mine rescue robot applying the methods described above. The solution provided by this mine rescue robot is similar to the solution described in the above methods. In an exemplary embodiment, such as... Figure 4 As shown, the mining rescue robot includes: a robot body and a controller. The controller is used to realize the autonomous following and autonomous navigation method of the mining rescue robot.

[0093] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data generated during autonomous following and navigation. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements an autonomous following and navigation method for a mining rescue robot.

[0094] Those skilled in the art will understand that Figure 5 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0095] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0096] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0097] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0098] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0099] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for autonomous following and autonomous navigation of a mining rescue robot, characterized in that, include: The system obtains the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, and uses a dynamic obstacle avoidance algorithm to control the mining rescue robot to autonomously follow the target to be followed; the UWB tag is set on the target to be followed, and the UWB base station is set on the mining rescue robot. The system acquires the movement trajectory and multi-source data from underground sources during the process of a mining rescue robot following a target, and constructs an underground environment map based on the movement trajectory and multi-source data. The underground multi-source data includes lidar point cloud data, IMU three-axis attitude data, wheel speed meter mileage data, and camera visual semantic data. After obtaining navigation instructions and verifying their validity, a path generation algorithm and a dynamic obstacle avoidance algorithm are used to control the mining rescue robot to automatically navigate from the navigation starting point to the navigation endpoint. The navigation instructions include a navigation start point and a navigation destination.

2. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 1, characterized in that, The mining rescue robot specifically includes: an explosion-proof main body, a power battery and battery management system, a drive-by-wire system, a steering-by-wire system, a braking-by-wire system, and a chassis controller. It also includes headlights, an audible and visual alarm device, a remote control, environmental sensors, wheel speedometers, a UWB base station, an intrinsically safe lidar, an intrinsically safe industrial night vision camera, a nine-axis IMU, an intrinsically safe touchscreen, and ultrasonic radar.

3. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 2, characterized in that, The system obtains the straight-line distance and relative horizontal deflection angle between the UWB tag and the UWB base station, and uses a dynamic obstacle avoidance algorithm to control the mining rescue robot to autonomously follow the target. Specifically, this includes: Obtain the straight-line distance between the UWB tag and the UWB base station to obtain the raw ranging data; Obtain the relative horizontal deflection angle between the UWB tag and the UWB base station to obtain the raw angle measurement data; Based on the original ranging data and original angle measurement data, the relative coordinates of the UWB tag in the robot coordinate system are calculated; the origin of the robot coordinate system is the geometric center of the mining rescue robot. Based on the IMU attitude data and rotation transformation matrix, the relative coordinates of the UWB tag in the robot coordinate system are converted to the coordinates of the UWB tag in the global coordinate system; Based on the coordinates of the UWB tag in the global coordinate system and obstacle perception data, a dynamic obstacle avoidance algorithm is used to automatically plan the real-time following path of the mining rescue robot; the obstacle perception data is acquired through intrinsically safe lidar and ultrasonic radar.

4. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 3, characterized in that, The relative coordinates of the UWB tag in the robot coordinate system are calculated using the following formula: ； The relative coordinates of the UWB tag in the robot coordinate system are: ; d This is the original distance measurement; θ This is angular measurement data.

5. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 3, characterized in that, Based on the coordinates of UWB tags in the global coordinate system and obstacle perception data, a dynamic obstacle avoidance algorithm is used to automatically plan the real-time following path of the mining rescue robot, specifically including: By combining the coordinates of UWB tags in the global coordinate system and obstacle perception data with a dynamic obstacle avoidance algorithm, a real-time following path for a mining rescue robot is planned. Based on the real-time following path of the mining rescue robot, speed and steering angle commands are sent to the chassis controller; The chassis controller receives commands and controls the drive-by-wire and steering system to perform following operations.

6. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 3, characterized in that, Acquire the movement trajectory and multi-source underground data of the mining rescue robot as it follows the target, and construct an underground environment map based on the trajectory and multi-source underground data, specifically including: Based on multi-source data from an intrinsically safe lidar, a nine-axis IMU, and a wheel speedometer, and the real-time following path of the mining rescue robot, the robot's trajectory during its journey following the target is acquired and recorded in real time. The trajectory is stored in the form of a waypoint coordinate sequence, including the waypoint global coordinates, speed, IMU attitude angle, and acquisition timestamp. The system uses an intrinsically safe lidar, a nine-axis IMU, a wheel speedometer, and an intrinsically safe industrial night vision camera to acquire multi-source data from underground sources during the process of a mining rescue robot following a target. Based on the movement trajectory of the mining rescue robot and multi-source data underground, an underground environment map is constructed and stored in binary format.

7. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 6, characterized in that, Based on multi-source data from integrated intrinsically safe lidar, nine-axis IMU, and wheel speedometer, and the real-time following path of the mining rescue robot, the robot's trajectory during its pursuit of the target is acquired and recorded in real time, specifically including: A particle filter algorithm is used to fuse data from lidar, nine-axis IMU, and wheel speed meter to achieve precise positioning. The fusion of wheel speed meter and nine-axis IMU odometer data is used as the coarse positioning result and as the initial particle set for particle filtering. The pose observation value is obtained by registering the real-time point cloud of the lidar with the fused high-precision map point cloud using the ICP algorithm. Based on the observed values, the particle set is sampled, its weights are updated, and it is resampled to output the precise position of the mining rescue robot in the global coordinate system. The trajectory is obtained; the origin of the global coordinate system is the robot's initial starting position. The particle weight update formula is: ； in, For the i-th particle in t Weight of time, For the i-th particle in t The weight at time -1 Let be the likelihood probability of the observation model.

8. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 7, characterized in that, Using intrinsically safe lidar, a nine-axis IMU, wheel speedometers, and intrinsically safe industrial night vision cameras, multi-source underground data is acquired during the process of a mining rescue robot following a target. Specifically, this includes: Point cloud data of the environment surrounding the mining rescue robot is obtained by collecting point cloud data of the intrinsically safe lidar. The three-axis attitude angles, accelerations, and angular velocities of the mining rescue robot were measured using a nine-axis IMU to obtain the IMU's three-axis attitude data. The wheel speed is measured by a wheel speed meter, and the driving speed, driving distance and mileage information of the mining rescue robot are calculated based on the wheel speed to obtain the mileage wheel speed meter mileage data. Visual features are identified using an intrinsically safe industrial night vision camera to obtain camera visual semantic data; the visual features include underground traffic lights, traffic signs, and tunnel corners; Statistical filtering and voxel downsampling are used to denoise the lidar point cloud data, complementary filtering is used to suppress the IMU three-axis attitude data, and time synchronization of multi-sensor data is achieved through timestamp interpolation to obtain downhole multi-source data; the multi-sensor data specifically includes lidar point cloud data, IMU three-axis attitude data, wheel speed meter mileage data, and camera visual semantic data.

9. The autonomous following and autonomous navigation method for a mining rescue robot according to claim 6, characterized in that, Based on the trajectory of the mining rescue robot and multi-source underground data, an underground environment map is constructed and stored in binary format, specifically including: A graph optimization algorithm is used to optimize the backend of the map. The waypoints of the mining rescue robot are used as nodes of the graph, and the pose constraints between waypoints are used as edges of the map. The Gauss-Newton method is used to minimize the overall error function and correct the global cumulative deviation of the map. At the same time, the camera visual semantic data is associated with the corresponding location on the map to achieve semantic map annotation. By fusing lidar point cloud maps and grid maps, and employing the occupancy grid method to calculate the obstacle occupancy probability of each grid cell, the grid cells are divided into three categories: occupied, vacant, and unknown, resulting in an underground environment map. This underground environment map is stored in binary format in the controller's solid-state storage, supporting real-time updates and intrinsically safe touchscreen visualization. The calculation formula for the occupancy grid method is as follows: ； in, Let be the posterior probability that a grid cell is occupied by an obstacle. For the probability of the observation model, For prior probability, This represents the global probability of the observed data.

10. A mining rescue robot, comprising: The robot body and controller are characterized in that the controller is used to implement the autonomous following and autonomous navigation method of the mining rescue robot according to any one of claims 1-9.

Images