Real-time updating system for unmanned dump retaining wall of open-pit mine

Abstract

The present application relates to open-pit mine unmanned technology field, more particularly to a kind of open-pit mine unmanned dump retaining wall real-time updating system.The present system includes: unmanned vehicle-mounted subsystem, the topography and obstacle information around vehicle are perceived, the operating state of vehicle and the form of retaining wall are sent to unmanned ground subsystem;Image video monitoring subsystem, according to the set unmanned vehicle target, track the operation area of unmanned vehicle, identify and judge the form of retaining wall after the operation of unmanned vehicle;Unmanned ground subsystem, according to the feedback data of unmanned vehicle-mounted subsystem and image video monitoring subsystem, establish dump map and update retaining wall form information.The present application carries out real-time updating to dump retaining wall, increases the accuracy of dump map, and further helps to improve the precision of unmanned vehicle parking retaining wall when dumping, improves the utilization rate of dumping point, reduces the operation amount of bulldozer finishing retaining wall.

Description

Technical Field

[0001] This invention relates to the field of unmanned driving technology in open-pit mines, and more specifically, to a real-time updating system for the retaining wall of an unmanned spoil heap in an open-pit mine. Background Technology

[0002] Currently, unmanned driving technology in open-pit mines is developing rapidly. Many companies, both domestic and international, have conducted research and testing on unmanned driving systems for earthmoving machinery, and some companies have already put their products into initial operation. However, existing unmanned driving systems for earthmoving machinery generally suffer from problems such as insufficient precision in stopping at retaining walls and a lack of intelligence in selecting dumping points.

[0003] A spoil heap, also known as a waste rock dump, is a site where mining waste is centrally disposed of. Spoil disposal refers to the operation of unloading stripped materials at a spoil heap.

[0004] When dumping soil at the spoil heap, unmanned vehicles sometimes fail to properly align with the retaining wall, resulting in the inability to completely dump rocks (or excavated soil, etc.) outside the retaining wall. Some or most of the cargo falls onto the retaining wall and then slides back inside, affecting efficiency and increasing the workload of the bulldozer. The reasons for this are twofold: first, the detection of the spoil heap retaining wall is not accurate enough; and second, the updates to changes in the spoil heap retaining wall are not timely enough. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time updating system for retaining walls of unmanned spoil heaps in open-pit mines, solving the problem of inaccurate spoil disposal by unmanned vehicles in the prior art.

[0006] To achieve the above objectives, the present invention provides a real-time update system for unmanned open-pit mine spoil heap retaining walls, comprising an unmanned vehicle-mounted subsystem, a communication subsystem, an image and video monitoring subsystem, and an unmanned ground subsystem:

[0007] The unmanned vehicle subsystem receives vehicle operation task and path information instructions from the unmanned ground subsystem, analyzes the tasks and target paths to be executed, controls the vehicle to execute tasks, senses surrounding terrain and obstacle information, and sends the vehicle's operating status and the shape of the retaining wall to the unmanned ground subsystem.

[0008] The image and video monitoring subsystem tracks the unmanned vehicle's operating area according to the set unmanned vehicle target, identifies and judges the shape of the retaining wall after the unmanned vehicle has finished operating, and feeds back the identification results to the unmanned ground subsystem.

[0009] The unmanned ground subsystem, based on feedback data from the unmanned vehicle subsystem, the drone map acquisition subsystem, and the image and video monitoring subsystem, establishes a spoil heap map and updates the retaining wall morphology information.

[0010] The communication subsystem establishes a network connection with the vehicle-mounted driving subsystem, the image and video monitoring subsystem, and the unmanned ground subsystem, providing data interaction and data management between the various subsystems.

[0011] In one embodiment, the retaining wall morphology information includes the height, thickness, position, and orientation angle of the retaining wall.

[0012] In one embodiment, the system also includes an unmanned aerial vehicle (UAV) map acquisition subsystem, which completes the cruise and retaining wall shape image acquisition according to the set flight route, and feeds back the acquisition results to the unmanned ground subsystem.

[0013] In one embodiment, when the spoil heap is used for the first time, images of the spoil heap are collected by an unmanned aerial vehicle (UAV) map acquisition subsystem to obtain a preliminary map of the spoil heap.

[0014] In one embodiment, as an unmanned vehicle equipped with an unmanned vehicle-mounted subsystem travels along the edge of the spoil heap, the unmanned ground subsystem receives the vehicle's location information and the detected retaining wall shape information in real time, processes the received location information to form an initial map of the spoil heap.

[0015] In one embodiment, the image and video monitoring subsystem scans and captures images of the planned dumping area one by one in the order of dumping. It calculates and models the shape of the retaining wall in the dumping area using an image recognition algorithm, and calculates the relative coordinates of the retaining wall on the map by combining the positioning information of the image and video monitoring subsystem. It then calculates and obtains the retaining wall shape information and sends it to the unmanned ground subsystem.

[0016] In one embodiment, the image recognition algorithm includes a retaining wall template library, which is obtained through machine vision deep learning training;

[0017] The image and video monitoring subsystem calculates and models the shape of the retaining wall in the spoil heap area using image recognition algorithms, and further includes:

[0018] The distance, height, and width information of the acquired images are calibrated to establish a relative coordinate system for the image information;

[0019] The system calls upon the retaining wall template library to identify retaining walls in the acquired images and marks the identified retaining walls.

[0020] Based on the relative coordinate system established during calibration, the marked retaining wall is calculated to obtain the outline size, direction and relative coordinate position of the marked retaining wall;

[0021] Based on the latitude and longitude coordinate system of the positioning device, a three-dimensional map of the calculated retaining wall is constructed according to the positioning coordinate system to obtain the actual position and shape information of the retaining wall.

[0022] In one embodiment, when the unmanned vehicle is dumping soil, the parking position and orientation of the first dumping are based on the position and azimuth of the retaining wall collected by the image and video monitoring subsystem.

[0023] When the vehicle approaches the retaining wall, the autonomous vehicle subsystem collects the shape information of the retaining wall and adjusts the vehicle trajectory according to the updated shape information of the retaining wall.

[0024] In one embodiment, after the vehicle is unloaded, during the forward lifting process, the autonomous vehicle subsystem collects the shape information of the retaining wall and compares it with the shape information of the retaining wall before unloading. It then determines whether the thickness and orientation angle of the retaining wall have changed. If there is a change, the newly obtained retaining wall shape information is sent to the autonomous ground subsystem for retaining wall shape information update.

[0025] In one embodiment, while the vehicles are dumping soil, the unmanned ground subsystem collects the position and pose information of all vehicles stopping at the dumping points and updates the map boundary and retaining wall information of the dumping site.

[0026] In one embodiment, the autonomous vehicle subsystem determines whether the dumping point corresponding to the retaining wall is available by the rate of change of the retaining wall. If it is unavailable, the system sends the information that the currently used dumping point can no longer dump soil to the autonomous ground subsystem. The autonomous ground subsystem updates the retaining wall information of the dumping site in real time and blocks the dumping point.

[0027] In one embodiment, the unmanned ground subsystem sends a command to the bulldozer to clear the retaining wall of the spoil heap based on the usage of the entire spoil heap and the vehicle transportation situation. The bulldozer then clears the spoil heap and repairs the retaining wall.

[0028] The image and video monitoring subsystem updates the map boundaries and retaining wall information of the spoil heap in real time.

[0029] In one embodiment, the image and video monitoring subsystem determines whether the soil dumping point corresponding to the retaining wall is available. If it is not available, it sends the information that the currently used soil dumping point can no longer dump soil to the unmanned ground subsystem.

[0030] This invention provides a real-time updating system for retaining walls in unmanned open-pit mine spoil heaps. The system updates the retaining walls in real time, increases the accuracy of spoil heap maps, and thus helps improve the accuracy of unmanned vehicles stopping at retaining walls during spoil heaping, increases the utilization rate of spoil heap points, and reduces the workload of bulldozers repairing retaining walls. Attached Figure Description

[0031] The above and other features, properties and advantages of the present invention will become more apparent from the following description taken in conjunction with the accompanying drawings and embodiments, in which the same reference numerals always denote the same features, wherein:

[0032] Figure 1 A schematic diagram of a real-time update system for unmanned spoil heap retaining walls in an open-pit mine, according to an embodiment of the present invention, is disclosed.

[0033] Figure 2 A flowchart of a method for real-time updating of retaining walls in an unmanned spoil heap of an open-pit mine, according to an embodiment of the present invention, is disclosed. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0035] To address the issues mentioned in existing technologies, this invention proposes a real-time update system for retaining walls in open-pit mines' unmanned spoil heaps, enabling real-time updates to the retaining walls. Real-time updates to the retaining wall positions increase the accuracy of the spoil heap map, thereby improving the precision with which unmanned vehicles stop at the retaining walls during spoil disposal. Improved vehicle stopping accuracy allows all transported stones and other waste to be dumped outside the retaining walls, increasing the utilization rate of spoil disposal points and reducing the workload of bulldozers in maintaining the retaining walls. This not only reduces the fuel costs of bulldozers but also improves the overall operational efficiency of the unmanned transportation system, ultimately increasing the profit margin of mining operations.

[0036] Figure 1 A schematic diagram of a real-time update system for the retaining wall of an unmanned open-pit mine spoil heap, according to an embodiment of the present invention, is disclosed. Figure 1 As shown, the open-pit mine unmanned spoil heap retaining wall real-time update system proposed in this invention includes an unmanned vehicle-mounted subsystem 100, an unmanned aerial vehicle map acquisition subsystem 200, a communication subsystem 300, an unmanned ground subsystem 400, and an image and video monitoring subsystem 500. These five subsystems together constitute the spoil heap retaining wall real-time update system.

[0037] The unmanned ground subsystem 400 (GMS) consists of a database server, a ground control server, a ground control client, a ground communication unit, and a central display screen. It serves as the comprehensive dispatch and command center for unmanned transportation operations of open-pit mine dump trucks, providing users with a clear and intuitive display of comprehensive mining area information and a human-machine interaction window.

[0038] The functions of the unmanned ground subsystem 400 include intelligent vehicle scheduling, map management, operation monitoring, safety protection, and statistical analysis.

[0039] The map management function includes features such as map creation, editing, publishing, dynamic merging, real-time updating, deletion, copying, and exporting.

[0040] The unmanned vehicle-mounted subsystem 100 (VAP) consists of equipment such as an onboard high-definition camera, lidar, millimeter-wave radar, onboard display unit, GPS inertial navigation device with RTK (Real-time kinematic, carrier phase differential) function, onboard industrial control computer, and onboard communication unit.

[0041] The main function of the unmanned vehicle subsystem 100 is to receive the vehicle's work tasks and path information from the unmanned ground subsystem 400, parse the required tasks and target paths according to the instructions, and issue control commands to its own actuators to control the vehicle to perform the tasks.

[0042] During the mission, the unmanned vehicle subsystem 100 uses sensors to perceive the surrounding terrain and obstacles, and sends the vehicle's operating status to the unmanned ground subsystem 100 in real time.

[0043] In some embodiments, using a multi-beam lidar instead of a high-definition camera for retaining wall location and shape recognition is an alternative to this embodiment.

[0044] In some embodiments, using detection devices such as lidar or light field cameras instead of high-definition cameras is an alternative to this embodiment.

[0045] The data communication system 300 utilizes the existing 4G / 5G and other basic communication facilities in the mine to build a vehicle-to-ground communication network through communication equipment such as vehicle-mounted communication units and ground communication units, and undertakes the functions of vehicle-to-ground data and vehicle-to-vehicle data interaction and data management.

[0046] The Drone Map Acquisition Subsystem 200 is a drone equipped with GPS positioning and an ultra-high-definition camera that can automatically complete cruise and image acquisition according to a set flight route.

[0047] The 500 image and video surveillance subsystem integrates an image processing platform, multiple image acquisition devices, and GPS positioning devices. It adopts a mobile design, is powered by solar panels and batteries, and can move its position as the spoil heap changes. It has high-precision RTK GPS positioning capabilities, and the image acquisition devices used have infrared night vision and anti-fog and snow removal functions, making it usable around the clock.

[0048] The image acquisition equipment comes with a pan-tilt unit and a motorized lens. The image acquisition equipment used to monitor the retaining wall of the spoil heap can automatically track the truck unloading area or the bulldozer operation area according to the set truck or bulldozer target. After the truck unloads or the bulldozer operates, it can identify and judge the shape of the retaining wall and feed back the recognition results to the unmanned ground subsystem 400.

[0049] Image acquisition equipment used for monitoring spoil heaps can also be remotely controlled, allowing for remote camera rotation and focus adjustment to capture panoramic and detailed images of the spoil heap.

[0050] In some embodiments, installing one or more cameras at or near the spoil heap to monitor overall changes at the spoil heap is an alternative to the image and video monitoring subsystem 500 in this embodiment.

[0051] In some embodiments, using a lidar, line scan camera, 3D camera, binocular vision camera, or light field camera to replace the image acquisition device in the image and video monitoring subsystem 500 of this embodiment for acquiring retaining wall morphology information is an alternative to this solution.

[0052] It should be noted that not all five subsystems in this embodiment are required. They can be combined and matched according to the actual production situation of the mine, as long as they can realize the real-time updating function of the spoil heap retaining wall.

[0053] The present invention proposes an unmanned open-pit mine spoil heap retaining wall real-time update system. After the unmanned vehicle unloads spoil, it uses a rear-facing high-definition camera to scan changes in the retaining wall and update the height, thickness, position and orientation information of the retaining wall in real time. The system uses drones or video surveillance to update the height, thickness, position and orientation information of the retaining wall after the truck unloads spoil and after the bulldozer operates, through image recognition technology.

[0054] Figure 2 A flowchart of a method for real-time updating of retaining walls in an unmanned open-pit mine spoil heap according to an embodiment of the present invention is disclosed below. Figure 2 Explain the scheme for creating a spoil heap map and collecting information on retaining wall morphology.

[0055] When the spoil heap is used for the first time, the UAV map acquisition subsystem 200 is used to acquire high-definition images of the spoil heap, and the image processing algorithm is used to obtain a preliminary map of the spoil heap.

[0056] Then, a vehicle equipped with the autonomous vehicle subsystem 100 is remotely controlled to drive along the edge of the spoil heap. During the drive, the autonomous vehicle subsystem 100 receives the vehicle's GPS location information in real time. After the vehicle completes one lap, the autonomous vehicle subsystem 100 processes the received location information according to the map creation algorithm to form an initial map of the spoil heap.

[0057] Then, the image and video monitoring subsystem 500 scans and captures high-resolution images of the planned dumping area one by one in the order of dumping. The image recognition algorithm calculates and models the shape of the retaining wall in the dumping area, and calculates the relative coordinates of the retaining wall on the map by combining the GPS location information of the image and video monitoring subsystem 500. The position, shape, direction and thickness of the retaining wall are obtained, and this information is sent to the unmanned vehicle subsystem 100.

[0058] After the image and video monitoring subsystem 500 has finished processing the images of the spoil disposal area, the unmanned ground subsystem will obtain a series of retaining wall outlines, location information and azimuth information. Then, through the map management module, the obtained single retaining wall segments are stitched together with GPS location coordinate information to obtain the outline, location information and azimuth of the entire spoil disposal line. The spoil disposal line information is then integrated with the initial map to generate the official map of the spoil disposal site.

[0059] At this point, the creation of the spoil heap map and the collection of retaining wall location information are complete.

[0060] The aforementioned image recognition algorithm includes a retaining wall template library, which is trained using a large number of retaining wall images taken at different shapes, distances, and angles through machine vision deep learning technology.

[0061] The process of calculating and modeling the shape of the retaining wall is as follows:

[0062] Before use, image acquisition equipment is calibrated, mainly to calibrate the distance, height, and width information of the acquired image and establish a relative coordinate system for the image information.

[0063] After receiving the image from the image acquisition device, the image processing platform automatically calls the retaining wall template library to identify the retaining walls in the image and marks the identified retaining walls. Then, based on the relative coordinate system established during calibration, it calculates the outline size, direction, and relative coordinate position of the marked retaining walls.

[0064] The modeling is based on the latitude and longitude coordinate system of the GPS positioning device. The calculated retaining wall is then used to create a three-dimensional map based on the GPS coordinate system to obtain the actual location and shape information of the retaining wall.

[0065] The following is combined Figure 2 Explain the usage of the spoil heap and the real-time update plan for the retaining wall.

[0066] When the autonomous vehicle is dumping soil, the initial parking position and orientation will be based on the position and azimuth of the retaining wall collected by the image and video monitoring subsystem 500. When the vehicle approaches the retaining wall, the autonomous vehicle's rear-facing high-definition camera will take a high-definition photo of the retaining wall's position. Then, the shape of the retaining wall will be recalculated and modeled by the image recognition algorithm integrated into the autonomous vehicle subsystem 100. Combined with the vehicle's GPS position information, the relative coordinates of the retaining wall on the map will be calculated to obtain information such as the position, shape, direction, and thickness of the retaining wall. The autonomous vehicle subsystem 100 will then adjust the vehicle's reversing trajectory appropriately based on the obtained retaining wall information to ensure that the vehicle accurately approaches the retaining wall.

[0067] After the vehicle is unloaded, during the forward lifting process, the unmanned vehicle subsystem 100 uses the rear-facing high-definition camera again to take multiple pictures of the retaining wall. After the pictures are taken, the image data is processed in a timely manner and compared with the retaining wall information before unloading to confirm whether there is material piled on the retaining wall and whether the thickness and orientation angle of the retaining wall have changed after the material is piled on.

[0068] If there are any changes, the new retaining wall outline and orientation angle information will be sent to the unmanned ground subsystem 400, and the unmanned ground subsystem 400 will update the retaining wall information in a timely manner.

[0069] The unmanned vehicle-mounted subsystem 100 also determines whether the soil dumping point is usable by the rate of change of the retaining wall (including the height of the retaining wall, the thickness of the retaining wall, the change of the base of the retaining wall, whether there are large blocks, etc.).

[0070] If unavailable, the information that the currently used dumping point can no longer dump soil is sent to the unmanned ground subsystem 400. The unmanned ground subsystem 400 then updates the retaining wall information of the dumping site in real time and blocks the dumping point. No vehicles will be arranged to unload soil at the dumping point until the bulldozer clears the dumping point.

[0071] While the vehicles are dumping soil, the unmanned ground subsystem 400 collects the position and pose information of all vehicles stopping at the dumping points, and continuously updates the map boundaries and retaining wall information of the dumping site.

[0072] The above process enables real-time updates of the retaining wall at the spoil heap during the spoil unloading process.

[0073] After receiving information that the spoil disposal point is unavailable, the unmanned ground subsystem 400 will make a comprehensive intelligent judgment based on the overall usage of the spoil disposal site and the vehicle transportation situation. If it is necessary to clean the spoil disposal site retaining wall, the request will be sent to the on-board display unit on the bulldozer, prompting the bulldozer to clean the spoil disposal site and repair the retaining wall.

[0074] After the bulldozer clears and repairs the retaining wall of the spoil heap, it uses the 500 image and video monitoring subsystem to take high-resolution photos of the cleared and repaired spoil heap area. Through image recognition technology, the map boundary and retaining wall information are updated in real time.

[0075] The image and video monitoring subsystem 500 serves as a real-time video image monitoring device for the spoil heap. It uses image recognition technology to monitor changes in the retaining wall after trucks dump soil and after bulldozers operate, identifying the shape and location of the retaining wall.

[0076] The specific process is as follows:

[0077] 1) After the truck enters the spoil heap, the truck's position is automatically identified and tracked through a pre-set truck model. After the truck finishes unloading and leaves the spoil heap retaining wall, the image data of the retaining wall is captured and compared with the retaining wall information before unloading. The shape of the retaining wall is calculated and modeled through image recognition technology to confirm whether there is material piled on the retaining wall, whether the thickness and orientation angle of the retaining wall have changed after the material is piled, and whether the spoil heap point is usable.

[0078] If a dumping site is unavailable, the information that the currently used dumping site can no longer dump soil is sent to the unmanned ground subsystem 400. The unmanned ground subsystem 400 then updates the retaining wall information of the dumping site in real time and blocks the dumping site, preventing vehicles from unloading soil at the site until a bulldozer clears it. This process ensures that the retaining wall information of the dumping site is updated after soil unloading, and this process is redundant with the identification process of the unmanned vehicle subsystem 100.

[0079] 2) After the bulldozer operation, high-resolution image data of the retaining wall is captured. The shape of the retaining wall is calculated and modeled using image recognition technology. Combined with the GPS location information of the image and video monitoring subsystem, the relative coordinate information of the retaining wall on the map is calculated to obtain information such as the position, shape, direction and thickness of the retaining wall. This information is then sent to the unmanned ground subsystem 400 to realize the update of the retaining wall information of the spoil heap after the bulldozer operation.

[0080] The image and video monitoring subsystem 500 can be used as a supplementary or auxiliary monitoring method for the detection of retaining walls at spoil heaps in the fully unmanned truck group operation mode.

[0081] In mixed operations of bulldozers, including those with manned / unmanned operation and those without sensors such as positioning devices, angle, and displacement sensors to match unmanned transportation operations, it serves as the primary monitoring method for detecting retaining walls at spoil heaps.

[0082] The 500 image and video monitoring subsystem also provides real-time video image data of the spoil heap to ground operators, making it easier for them to understand the actual situation of the spoil heap.

[0083] The UAV map acquisition subsystem 200 serves as an auxiliary means, periodically or immediately taking high-definition photos of the spoil heap via pre-set flight routes and intervals or according to instructions from ground dispatchers. It also uses image recognition to obtain relevant information about the retaining wall. This compensates for situations where rear-view cameras on some vehicles are obstructed by dust or malfunction, or where the image acquisition equipment of the image and video monitoring subsystem 500 is obstructed or malfunctioning, preventing timely acquisition of retaining wall images.

[0084] Through the above four-process cycle, the position, shape, thickness and orientation angle of the retaining wall can be updated in real time throughout the use and maintenance of the spoil heap, so as to achieve the purpose of automatic updating of the spoil heap retaining wall.

[0085] By updating the retaining wall information of the spoil heap in real time, the unmanned ground subsystem can keep track of changes in the retaining wall, which helps the intelligent scheduling and decision-making of the entire unmanned transportation system.

[0086] Real-time information about the retaining wall helps the unmanned ground subsystem plan a more reasonable reversing trajectory for the dumping vehicles, enabling them to park more accurately next to the retaining wall. This allows all the transported stones and other debris to be dumped outside the retaining wall, improving the utilization rate of the dumping point and reducing the workload of bulldozers in repairing the retaining wall. This not only reduces the fuel consumption cost of bulldozers but also improves the overall operational efficiency of the unmanned transportation system.

[0087] Real-time updates of the retaining wall information at the spoil heap are a crucial part of the entire unmanned transportation system, playing a decisive role in the smooth operation of the system. The implementation of this real-time update scheme will greatly improve the efficiency of unmanned transportation operations.

[0088] This invention provides a real-time updating system for retaining walls in unmanned open-pit mine spoil heaps. It is applicable not only to purely unmanned transportation operations but also to transportation operations involving a mix of manned and unmanned vehicles sharing a spoil heap. This eliminates the need to divide the spoil heap into manned and unmanned sections, improving site utilization efficiency. In mining environments with limited space, it reduces construction organization difficulty and increases production efficiency.

[0089] Although the methods described above are illustrated and depicted as a series of actions for the sake of simplicity, it should be understood and appreciated that these methods are not limited by the order of the actions, as some actions may occur in a different order and / or concurrently with other actions from the illustrations and descriptions herein or not illustrated and described herein but which may be understood by those skilled in the art, according to one or more embodiments.

[0090] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0091] The above embodiments are provided for those skilled in the art to implement or use the present invention. Those skilled in the art can make various modifications or changes to the above embodiments without departing from the inventive concept of the present invention. Therefore, the protection scope of the present invention is not limited to the above embodiments, but should be the maximum scope that conforms to the innovative features mentioned in the claims.

Claims

1. A real-time updating system for retaining walls of unmanned spoil heaps in open-pit mines, characterized in that, It includes an unmanned vehicle-mounted subsystem, a communication subsystem, an image and video surveillance subsystem, and an unmanned ground subsystem: The unmanned vehicle subsystem receives vehicle operation task and path information instructions from the unmanned ground subsystem, analyzes the tasks and target paths to be executed, controls the vehicle to execute tasks, senses surrounding terrain and obstacle information, and sends the vehicle's operating status and the shape of the retaining wall to the unmanned ground subsystem. The image and video monitoring subsystem tracks the unmanned vehicle's operating area according to the set unmanned vehicle target, identifies and judges the shape of the retaining wall after the unmanned vehicle has finished operating, and feeds back the identification results to the unmanned ground subsystem. The unmanned ground subsystem establishes a spoil heap map and updates the retaining wall morphology information based on feedback data from the unmanned vehicle subsystem and the image and video monitoring subsystem. The communication subsystem establishes a network connection with the unmanned vehicle subsystem, the image and video monitoring subsystem, and the unmanned ground subsystem, providing data interaction and data management between the various subsystems; The image and video monitoring subsystem scans and captures images of the planned dumping areas one by one in the order of dumping. It calculates and models the shape of the retaining wall in the dumping area using an image recognition algorithm, and calculates the relative coordinates of the retaining wall on the map by combining the positioning information of the image and video monitoring subsystem. It then calculates and obtains the shape information of the retaining wall and sends it to the unmanned ground subsystem.

2. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 1, characterized in that, It also includes a drone map acquisition subsystem, which completes the cruise and retaining wall shape image acquisition according to the set flight route, and feeds back the acquisition results to the unmanned ground subsystem.

3. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 2, characterized in that, When the spoil heap was used for the first time, images of the spoil heap were collected through the drone map acquisition subsystem to obtain a preliminary map of the spoil heap.

4. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 1, characterized in that, As the unmanned vehicle equipped with the unmanned vehicle-mounted subsystem travels along the edge of the spoil heap, the unmanned ground subsystem receives the vehicle's location information in real time, processes the received location information, and generates an initial map of the spoil heap.

5. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 1, characterized in that, The image recognition algorithm includes a retaining wall template library, which is obtained through machine vision deep learning training. The image and video monitoring subsystem calculates and models the shape of the retaining wall in the spoil heap area using image recognition algorithms, and further includes: The distance, height, and width information of the acquired images are calibrated to establish a relative coordinate system for the image information; The system calls upon the retaining wall template library to identify retaining walls in the acquired images and marks the identified retaining walls. Based on the relative coordinate system established during calibration, the marked retaining wall is calculated to obtain the outline size, direction and relative coordinate position of the marked retaining wall; Based on the latitude and longitude coordinate system of the positioning device, a three-dimensional map of the calculated retaining wall is constructed according to the positioning coordinate system to obtain the actual position and shape information of the retaining wall.

6. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 1, characterized in that, When unmanned vehicles are dumping soil, the initial dumping location and orientation are based on the position and azimuth of the retaining wall collected by the image and video monitoring subsystem. When the vehicle approaches the retaining wall, the autonomous vehicle subsystem collects the shape information of the retaining wall and adjusts the vehicle trajectory according to the updated shape information of the retaining wall.

7. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 6, characterized in that, After the vehicle is unloaded, during the forward lifting process, the autonomous vehicle subsystem collects the shape information of the retaining wall and compares it with the shape information of the retaining wall before unloading. It then determines whether the thickness and orientation angle of the retaining wall have changed. If there are changes, the newly obtained retaining wall shape information is sent to the autonomous ground subsystem for retaining wall shape information update.

8. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 6, characterized in that, While the vehicles are dumping soil, the unmanned ground subsystem collects the position and pose information of all vehicles at the dumping points and updates the map boundaries and retaining wall information of the dumping site.

9. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 7, characterized in that, The autonomous vehicle subsystem determines whether the dumping point corresponding to the retaining wall is available by measuring the rate of change of the retaining wall. If it is unavailable, it sends the information that the currently used dumping point can no longer dump soil to the autonomous ground subsystem. The autonomous ground subsystem updates the retaining wall information of the dumping site in real time and blocks the dumping point.

10. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 7, characterized in that, The unmanned ground subsystem sends a command to the bulldozer to clean the retaining wall of the spoil heap based on the usage and vehicle transportation status of the entire spoil heap. The bulldozer then cleans the spoil heap and repairs the retaining wall. The image and video monitoring subsystem updates the map boundaries and retaining wall information of the spoil heap in real time.

11. The real-time updating system for retaining walls of unmanned open-pit mine spoil heaps according to claim 10, characterized in that, The image and video monitoring subsystem determines whether the soil dumping point corresponding to the retaining wall is available. If it is not available, it sends the information that the currently used soil dumping point can no longer dump soil to the unmanned ground subsystem.

Images