A control method and device for unmanned intelligent rock drilling trolley
Through the control method of unmanned intelligent rock drilling trolley, using technologies such as lidar, fisheye camera and millimeter wave radar, independent rock drilling is achieved, solving the problem of inefficient operation of traditional rock drilling trolleys, and improving operational efficiency and safety sex.
Patent Information
- Application Number
- CN202310214823.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-07
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2043-03-07
AI Technical Summary
Traditional rock drilling trolleys require manual operation, which causes operators to face dangerous factors such as vibration, noise, dust, rock collapse in harsh environments, making it difficult to accurately analyze the scene and rock types, and are inefficient.
Adopt unmanned intelligent rock drilling trolley control method, use lidar active rock identification, fisheye camera to make up for lidar blind spots, millimeter wave radar to ensure environmental safety, combine high-definition cameras and high-precision navigation system, and realize information processing and remote control technology. Independent rock drilling.
Unmanned autonomous rock drilling is achieved, and the best strategies are generated based on different rocks and conditions, which improves operating efficiency and ensures the safety and accuracy of vehicle operation.
Smart Images

Figure CN116859904B_ABST
Abstract
Description
Technical Field
[0001] The invention discloses an unmanned intelligent rock drilling trolley control method and device, and relates to the field of unmanned intelligent rock drilling trolley control. Background Art
[0002] Rock drilling rigs are generally used in scenes such as mines, railways, lanes and tunnels, which are relatively harsh environments. Traditional rock drilling rigs require manual operation, and operators must be in a dangerous rock drilling environment, which is easily affected by dangerous factors such as vibration, noise, dust, rock collapse, etc. It is difficult to accurately analyze the scene conditions and rock types, and the accuracy and efficiency are limited by manual control, resulting in low operating efficiency. Summary of the invention
[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides an unmanned intelligent rock drilling trolley control method and device, which autonomously controls the vehicle drilling based on the laser radar rock active identification technology, the fisheye camera compensates for the laser radar blind spot, the millimeter wave radar ensures the safety of vehicle operation and working environment, and the camera plus remote control technology compensates for the incorrect operation of the vehicle, so as to realize unmanned and autonomous rock drilling, produce the corresponding optimal strategy according to different rocks and different conditions, and improve work efficiency.
[0004] A method for controlling an unmanned intelligent rock drilling trolley, characterized by comprising:
[0005] Collecting data about the surrounding environment of the unmanned intelligent rock drilling rig based on a sensor unit, wherein the sensor unit includes two laser radars, eight millimeter wave radars, two fisheye cameras, two high-definition cameras and a high-precision integrated navigation system;
[0006] The information processing unit processes the collected data of the surrounding environment of the unmanned intelligent rock drilling rig to obtain obstacle data and a drivable area;
[0007] The remote transmission and control unit performs a security judgment based on the data of the surrounding environment transmitted to the security user, and issues an additional remote control execution command after the judgment. The remote transmission and control unit is composed of a network switch, a router, a 5G CPE, a remote control display screen and a remote control console;
[0008] The strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy;
[0009] The command detection unit detects whether an additional remote control execution command generated by the security user based on the data of the surrounding environment is received;
[0010] If the additional remote control execution command is not received, the unmanned intelligent drilling rig is controlled based on the control strategy;
[0011] If the additional remote control execution command is received, the control strategy obtained by the analysis is modified according to the additional remote control execution command to obtain a modified control strategy, and the unmanned intelligent rock drilling rig is controlled based on the modified control strategy.
[0012] Preferably, the collecting of data about the surrounding environment of the unmanned intelligent drilling rig based on the sensor unit includes:
[0013] The two laser radars respectively acquire first point cloud data and second point cloud data based on real-time scanning of obstacles and terrain around the drilling rig;
[0014] The eight millimeter-wave radars detect the distance, angle and speed of obstacles around the drilling rig to obtain status data of the surrounding obstacles;
[0015] The two high-definition cameras acquire real-time images around the drilling rig;
[0016] The two fisheye cameras obtain real-time images of the blind area of the laser radar signal;
[0017] The integrated navigation system acquires positioning information and navigation routes.
[0018] Preferably, the information processing unit processes the collected data about the surrounding environment of the unmanned intelligent drilling rig to obtain obstacle data and a drivable area, including:
[0019] The first point cloud data and the second point cloud data are fused by using a data fusion algorithm to synthesize the initial point cloud information;
[0020] Perform ground point cloud removal processing according to the depth data to obtain processed point cloud information;
[0021] Comprehensively analyze point cloud information, status data, real-time images, positioning information and navigation routes to obtain obstacle data and drivable areas.
[0022] Preferably, the remote control unit performs a safety judgment based on the data of the surrounding environment transmitted to the safety user, and issues an additional remote control execution command after the judgment, including:
[0023] Extract the surrounding environment data to the network switch, and transmit the surrounding environment data to the remote control display screen based on the router and 5G CPE;
[0024] The security officer at the remote end judges the displayed data of the surrounding environment and issues additional remote control execution commands.
[0025] Preferably, the strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy, including:
[0026] Input obstacle data I, and obtain the image height and width of the obstacle in I, respectively, h and w; generate a set of candidate object boxes based on the deep learning framework RPN in Faster R-CNN, and the candidate object boxes represent the area containing rock obstacles; for each candidate object box b i , the RPN network outputs two values: p i Indicates the probability of whether the candidate object box contains an object, Indicates b i The four parameters to be adjusted, the output of the RPN network is expressed as:
[0027]
[0028] Where n represents the number of candidate object boxes;
[0029] Classify and regress candidate object boxes. Use a network based on Faster R-CNN to classify and regress candidate object boxes. For each candidate object box b i , the Fast R-CNN network takes it as input and outputs a probability vector S i =(S i,1 ,S i,2 ,…,S i,C ) and a regression vector B i =(B i,x ,B i,y ,B i,w ,B i,h ), where C represents the number of rock obstacle categories; S i,C Indicates b i The probability of belonging to the Cth category, B i,x Indicates b i The regression value of the central horizontal axis, B i,y Indicates b i The regression value of the central ordinate, B i,w Indicates b i The regression value of width, B i, h means b i Height of regression value;
[0030] The output of the Fast R-CNN network is expressed as:
[0031]
[0032] The Faster R-CNN algorithm screens and performs non-maximum suppression on the candidate object boxes generated by the RPN network to obtain the final rock obstacle detection results.
[0033] Preferably, the generation control strategy includes:
[0034] Based on the path planning algorithm and motion control algorithm, the rock obstacle detection results and the drivable area are analyzed to generate the optimal path and speed parameters and obtain the control strategy.
[0035] Preferably, the control strategy for controlling the unmanned intelligent drilling rig includes:
[0036] Calculate and decide the control strategy to obtain control instructions and execution action parameters;
[0037] The control instructions and execution action parameters are encoded and decoded, and transmitted to the actuator to control the action of the unmanned intelligent rock drilling rig.
[0038] Preferably, the control instructions and execution action parameters include throttle control, steering control, brake control, oil pump control, water pump control, air pump control, drill arm control, thrust beam control, rock drill control and communication control.
[0039] The control method of an unmanned intelligent rock drilling rig is characterized in that the actuator is provided with a component monitoring, and the component monitoring detects the status information of the oil pressure status, the water temperature status, the air pressure status, the steering system working status and the braking system working status to the remote control center.
[0040] An unmanned intelligent rock drilling trolley device, characterized in that the unmanned intelligent rock drilling trolley device comprises:
[0041] Sensor unit: used to collect data about the environment around the unmanned intelligent rock drilling rig;
[0042] Information processing unit: used to process the collected data of the surrounding environment of the unmanned intelligent rock drilling rig to obtain obstacle data and drivable area;
[0043] Remote control unit: used to transmit the data of the surrounding environment to the remote security user for security judgment, and issue additional remote control execution commands after the judgment;
[0044] Strategy generation unit: used to actively identify rocks based on the obstacle data and generate control strategies.
[0045] A command detection unit: configured to detect whether an additional remote control execution command generated by a security user based on the data of the surrounding environment is received;
[0046] Output port: used to transmit the final control strategy to the actuator of the unmanned intelligent rock drilling rig, and the actuator executes the control strategy;
[0047] Component monitoring: used to detect oil pressure status, water temperature status, air pressure status, steering system working status, brake system working status and other status information to the remote control center.
[0048] In the embodiment of the present invention, the unmanned intelligent rock drilling trolley autonomously controls the vehicle's drilling advance based on the laser radar active rock recognition technology, the fisheye camera compensates for the laser radar blind spot, the millimeter wave radar ensures the safety of the vehicle operation and working environment, and the camera plus remote control technology compensates for the incorrect operation of the vehicle, thereby realizing unmanned autonomous rock drilling, producing the corresponding optimal strategy according to different rocks and different conditions, and improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0050] Figure 1 It is a flow chart of an unmanned intelligent rock drilling trolley and a control method in an embodiment of the present invention;
[0051] Figure 2 It is a schematic diagram of the structural composition of an unmanned intelligent rock drilling trolley device in an embodiment of the present invention. DETAILED DESCRIPTION
[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0053] Figure 1 A schematic diagram showing a flow chart of an unmanned intelligent rock drilling rig and a control method in an embodiment of the present invention
[0054] like Figure 1 As shown, a method for controlling an unmanned intelligent rock drilling rig comprises:
[0055] S11: Collecting data about the surrounding environment of the unmanned intelligent drilling rig based on the sensor unit;
[0056] In the specific implementation process of the present invention, the sensor unit includes two laser radars, eight millimeter wave radars, two fisheye cameras, two high-definition cameras and a high-precision combined navigation system;
[0057] Preferably, the collecting of data about the surrounding environment of the unmanned intelligent rock drilling rig based on the sensor unit comprises: the two laser radars respectively obtain the first point cloud data and the second point cloud data based on real-time scanning of obstacles and terrain around the rock drilling rig;
[0058] The eight millimeter-wave radars detect the distance, angle and speed of obstacles around the drilling rig and obtain the status data of the surrounding obstacles. The millimeter-wave radar can provide high-resolution environmental perception capabilities under adverse weather conditions, detect static and dynamic obstacles around the drilling rig, and provide corresponding distance and speed information to support autonomous navigation and obstacle avoidance;
[0059] The two high-definition cameras acquire real-time images of the surroundings of the drilling rig, providing the operator with more intuitive environmental awareness, and can also be used for autonomous navigation and environmental recognition;
[0060] The two fisheye cameras obtain real-time images of the blind spots of the laser radar signal, providing the operator with a more comprehensive environmental perception capability, and can also be used for autonomous navigation and environmental recognition;
[0061] The combined navigation system can utilize data from multiple sensors (such as GPS, IMU, compass, etc.) to achieve high-precision positioning and navigation, and provide it to the autonomous navigation and path planning modules to help the drilling rig accurately reach the target location.
[0062] S12: The information processing unit processes the collected data of the surrounding environment of the unmanned intelligent drilling rig to obtain obstacle data and a drivable area;
[0063] Preferably, the information processing unit processes the collected data about the surrounding environment of the unmanned intelligent drilling rig to obtain obstacle data and a drivable area, including:
[0064] The first point cloud data and the second point cloud data are fused by using a data fusion algorithm to synthesize the initial point cloud information;
[0065] Perform ground point cloud removal processing according to the depth data to obtain processed point cloud information;
[0066] Comprehensively analyze point cloud information, status data, real-time images, positioning information and navigation routes to obtain obstacle data and drivable areas.
[0067] S13: The remote control unit performs a security judgment based on the data of the surrounding environment transmitted to the security user, and issues an additional remote control execution command after the judgment; the remote control unit is composed of a network switch, a router, a 5G CPE, a remote control display screen and a remote control console;
[0068] The network switch is used to connect different devices together to form a local area network, which facilitates communication and data transmission between devices; the router is responsible for connecting the local area network to the external network, such as connecting to the Internet; the router can also control network traffic and protect network security; the 5G CPE (Customer Premises Equipment) is a device that accesses the 5G network and can provide high-speed wireless network connection. 5G CPE is usually used to ensure the stability and high speed of remote transmission and control; the remote display screen is used to display the status and content of the remotely controlled device. The remote display screen is usually a high-resolution screen. The remote control console is a device used to control the remotely controlled device; the remote control console usually provides some special interfaces and controllers, such as keyboard, mouse, touch screen, etc., to remotely control the operation of the device. These components can work together so that users can easily remotely control various devices and systems through the remote transmission and control unit.
[0069] Preferably, the remote control unit performs a safety judgment based on the data of the surrounding environment transmitted to the safety user, and issues an additional remote control execution command after the judgment, including:
[0070] Extract the surrounding environment data to the network switch, and transmit the surrounding environment data to the remote control display screen based on the router and 5G CPE;
[0071] The security officer at the remote end judges the displayed data of the surrounding environment and issues additional remote control execution commands.
[0072] S14: The strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy;
[0073] Preferably, the strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy, including: using a feature extraction algorithm to utilize edge detection, corner detection and other technologies to extract rock obstacles from the background to form significant features, and then using a feature matching algorithm to achieve obstacle identification; using a deep learning-based algorithm to use a convolutional neural network (CNN) to train a classification model to distinguish rock obstacles from the background, and using the model for classification to achieve rock obstacle identification; using a target detection algorithm to treat the rock obstacle as a target, and using the target detection algorithm Faster R-CNN to detect the position and size of the rock obstacle from the image; finally, based on an optical flow algorithm, using the optical flow information between two consecutive frames captured by the camera to determine the position and movement direction of the obstacle, and then identifying the rock obstacle.
[0074] Preferably, the process of the target detection algorithm Faster R-CNN is as follows: input obstacle data I, obtain the image height and width of the obstacle in I, respectively h and w; generate a set of candidate object frames based on the deep learning framework Region Proposal Network (RPN) in Faster R-CNN, wherein the candidate object frames represent the regions containing rock obstacles; for each candidate object frame b i , the RPN network outputs two values: p i Indicates the probability of whether the candidate object box contains an object, Indicates b i The four parameters to be adjusted, the output of the RPN network is expressed as:
[0075]
[0076] Where n represents the number of candidate object boxes;
[0077] Classify and regress candidate object boxes. Use a network based on Faster R-CNN to classify and regress candidate object boxes. For each candidate object box b i , the Fast R-CNN network takes it as input and outputs a probability vector S i =(S i,1 ,S i,2 ,…,S i,C ) and a regression vector B i =(B i,x ,B i,y ,B i,w ,B i,h ), where C represents the number of rock obstacle categories; S i,C Indicates b i The probability of belonging to the Cth category, B i,x Indicates b i The regression value of the central horizontal axis, B i,y Indicates b i The regression value of the central ordinate, B i,w Indicates b i The regression value of width, B i, h means b i Height of regression value;
[0078] The output of the Fast R-CNN network is expressed as:
[0079]
[0080] The Faster R-CNN algorithm screens and performs non-maximum suppression on the candidate object boxes generated by the RPN network to obtain the final rock obstacle detection results.
[0081] Preferably, the generation control strategy includes:
[0082] Based on the path planning algorithm and motion control algorithm, the rock obstacle detection results and the drivable area are analyzed to generate the optimal path and speed parameters and obtain the control strategy.
[0083] Preferably, the control strategy for controlling the unmanned intelligent drilling rig includes:
[0084] Calculate and decide the control strategy to obtain control instructions and execution action parameters;
[0085] The control instructions and execution action parameters are encoded and decoded, and transmitted to the actuator to control the action of the unmanned intelligent rock drilling rig.
[0086] S15: The command detection unit detects whether an additional remote control execution command generated by the security user based on the data of the surrounding environment is received. If the additional remote control execution command is not received, the process proceeds to step S16. If the additional remote control execution command is received, the process proceeds to step S17.
[0087] S16: Controlling the unmanned intelligent drilling rig based on the control strategy;
[0088] S17: modifying the control strategy obtained by the analysis according to the additional remote control execution command to obtain a modified control strategy, and controlling the unmanned intelligent rock drilling rig based on the modified control strategy.
[0089] Preferably, the unmanned intelligent drilling rig is provided with an output port, which is used to transmit the final control strategy to the actuator of the unmanned intelligent drilling rig, and the actuator executes the control strategy; the output port transmits the final control strategy to the actuator of the unmanned intelligent drilling rig. The main purpose of this step is to transmit the final control strategy that has been calculated and decided to the actuator, and the actuator executes the control action, thereby achieving precise control of the drilling rig.
[0090] Specifically, the output port can be an interface on the electronic device, through which the digital signal is transmitted to the actuator. The transmitted digital signal contains information such as control instructions and execution action parameters, and the actuator can perform rock drilling operations based on this information. During the transmission process, the transmitted information also needs to be encoded and decoded to ensure the accuracy and reliability of the transmission.
[0091] In the process of realizing the output port, many factors need to be considered, such as data transmission rate, transmission distance, communication protocol, etc. It is also necessary to consider the actual situation of the actuator, such as the actuator's response speed, action accuracy, etc., to ensure that the output control instructions can be accurately executed.
[0092] Preferably, the control instructions and execution action parameters include throttle control, steering control, brake control, oil pump control, water pump control, air pump control, drill arm control, thrust beam control, rock drill control and communication control.
[0093] Preferably, the actuator is provided with a component monitoring, and the component monitoring detects the status information of the oil pressure status, the water temperature status, the air pressure status, the steering system working status and the braking system working status to the remote control center.
[0094] Figure 2 A schematic diagram of the structure of an unmanned intelligent rock drilling trolley device in an embodiment of the present invention is shown, and the unmanned intelligent rock drilling trolley comprises:
[0095] Sensor unit: used to collect data about the environment around the unmanned intelligent rock drilling rig;
[0096] Information processing unit: used to process the collected data of the surrounding environment of the unmanned intelligent rock drilling rig to obtain obstacle data and drivable area;
[0097] Remote control unit: used to transmit the data of the surrounding environment to the remote security user for security judgment, and issue additional remote control execution commands after the judgment;
[0098] Strategy generation unit: used for actively identifying rocks based on the obstacle data and generating a control strategy;
[0099] A command detection unit: configured to detect whether an additional remote control execution command generated by a security user based on the data of the surrounding environment is received;
[0100] Output port: used to transmit the final control strategy to the actuator of the unmanned intelligent rock drilling rig, and the actuator executes the control strategy;
[0101] Component monitoring: used to detect oil pressure status, water temperature status, air pressure status, steering system working status, brake system working status and other status information to the remote control center.
[0102] In the embodiment of the present invention, the unmanned intelligent rock drilling trolley autonomously controls the vehicle's drilling advance based on the laser radar active rock recognition technology, the fisheye camera compensates for the laser radar blind spot, the millimeter wave radar ensures the safety of the vehicle operation and working environment, and the camera plus remote control technology compensates for the incorrect operation of the vehicle to achieve unmanned autonomous rock drilling, produce the corresponding optimal strategy according to different rocks and different conditions, and improve work efficiency.
[0103] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.
[0104] In addition, the above is a detailed introduction to an unmanned intelligent rock drilling trolley control method and device provided in an embodiment of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. A method for controlling an unmanned intelligent rock drilling rig, characterized in that: include: Collecting data about the surrounding environment of the unmanned intelligent rock drilling rig based on a sensor unit, wherein the sensor unit includes two laser radars, eight millimeter wave radars, two fisheye cameras, two high-definition cameras and a high-precision integrated navigation system; The information processing unit processes the collected data of the surrounding environment of the unmanned intelligent rock drilling rig to obtain obstacle data and a drivable area; The remote transmission and control unit performs a security judgment based on the data of the surrounding environment transmitted to the security user, and issues an additional remote control execution command after the judgment. The remote transmission and control unit is composed of a network switch, a router, a 5G CPE, a remote control display screen and a remote control console; The strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy; The command detection unit detects whether an additional remote control execution command generated by the security user based on the data of the surrounding environment is received; If the additional remote control execution command is not received, the unmanned intelligent drilling rig is controlled based on the control strategy; If the additional remote control execution command is received, the control strategy obtained by analysis is modified according to the additional remote control execution command to obtain a modified control strategy, and the unmanned intelligent rock drilling rig is controlled based on the modified control strategy; The strategy generation unit performs active rock identification based on the obstacle data and generates a control strategy, including: Input obstacle data I, and obtain the image height and width of the obstacle in I, respectively, h and w; generate a set of candidate object boxes based on the deep learning framework RPN in Faster R-CNN, and the candidate object boxes represent the area containing rock obstacles; for each candidate object box b i , the RPN network outputs two values: p i Indicates the probability of whether the candidate object box contains an object, Indicates b i The four parameters to be adjusted, the output of the RPN network is expressed as: Where n represents the number of candidate object boxes; Classify and regress candidate object boxes. Use a network based on Faster R-CNN to classify and regress candidate object boxes. For each candidate object box b i , the Faster R-CNN network takes it as input and outputs a probability vector S i =(S i,1 , S i,2 , ..., S i,C ) and a regression vector B i =(B i,x , B i,y , B i,w , B i,h ), where C represents the number of rock obstacle categories; S i,C Indicates b i The probability of belonging to the Cth category, B i,x Indicates b i The regression value of the central horizontal axis, B i,y Indicates b i The regression value of the central ordinate, B i,w Indicates b i The regression value of width, B i,h Indicates b i Height of regression value; The output of the Faster R-CNN network is expressed as: The Faster R-CNN algorithm screens and performs non-maximum suppression on the candidate object boxes generated by the RPN network to obtain the final rock obstacle detection results.
2. The unmanned intelligent drilling trolley control method according to claim 1, characterized in that: The data of the surrounding environment of the unmanned intelligent rock drilling rig is collected based on the sensor unit, including: The two laser radars respectively acquire first point cloud data and second point cloud data based on real-time scanning of obstacles and terrain around the drilling rig; The eight millimeter-wave radars detect the distance, angle and speed of obstacles around the drilling rig to obtain status data of the surrounding obstacles; The two high-definition cameras acquire real-time images around the drilling rig; The two fisheye cameras obtain real-time images of the blind area of the laser radar signal; The integrated navigation system acquires positioning information and navigation routes.
3. The unmanned intelligent drilling rig control method according to claim 2, characterized in that: The information processing unit processes the collected data about the surrounding environment of the unmanned intelligent drilling rig to obtain obstacle data and a drivable area, including: The first point cloud data and the second point cloud data are fused by using a data fusion algorithm to synthesize the initial point cloud information; Perform ground point cloud removal processing according to the depth data to obtain processed point cloud information; Comprehensively analyze point cloud information, status data, real-time images, positioning information and navigation routes to obtain obstacle data and drivable areas.
4. The unmanned intelligent drilling trolley control method according to claim 1, characterized in that: The remote control unit performs a safety judgment based on the data of the surrounding environment transmitted to the safety user, and issues an additional remote control execution command after the judgment, including: Extract the surrounding environment data to the network switch, and transmit the surrounding environment data to the remote control display screen based on the router and 5G CPE; The security officer at the remote end judges the displayed data of the surrounding environment and issues additional remote control execution commands.
5. The unmanned intelligent drilling rig control method according to claim 1, characterized in that: The generation control strategy includes: Based on the path planning algorithm and motion control algorithm, the rock obstacle detection results and the drivable area are analyzed to generate the optimal path and speed parameters and obtain the control strategy.
6. The unmanned intelligent drilling rig control method according to claim 1, characterized in that The control strategy for controlling the unmanned intelligent rock drilling rig includes: Calculate and decide the control strategy to obtain control instructions and execution action parameters; The control instructions and execution action parameters are encoded and decoded, and transmitted to the actuator to control the action of the unmanned intelligent rock drilling rig.
7. The unmanned intelligent drilling rig control method according to claim 6, characterized in that: The control instructions and execution action parameters include throttle control, steering control, brake control, oil pump control, water pump control, air pump control, drill arm control, propulsion beam control, rock drill control and communication control.
8. The unmanned intelligent drilling rig control method according to claim 6, characterized in that: The actuator is provided with a component monitoring, and the component monitoring detects the status information of the oil pressure status, the water temperature status, the air pressure status, the steering system working status and the braking system working status to the remote control center.
9. An unmanned intelligent rock drilling trolley device, characterized in that: The unmanned intelligent rock drilling trolley device comprises: Sensor unit: used to collect data about the environment around the unmanned intelligent rock drilling rig; Information processing unit: used to process the collected data of the surrounding environment of the unmanned intelligent rock drilling rig to obtain obstacle data and drivable area; Remote transmission and control unit: used to transmit the data of the surrounding environment to the remote security user for security judgment, and issue additional remote control execution commands after the judgment; Strategy generation unit: used for actively identifying rocks based on the obstacle data and generating a control strategy; A command detection unit: configured to detect whether an additional remote control execution command generated by a security user based on the data of the surrounding environment is received; Output port: used to transmit the final control strategy to the actuator of the unmanned intelligent rock drilling rig, and the actuator executes the control strategy; Component monitoring: used to detect the status information of oil pressure, water temperature, air pressure, steering system working status, and brake system working status to the remote control center; The active rock identification based on the obstacle data and the generation of a control strategy include: Input obstacle data I, and obtain the image height and width of the obstacle in I, respectively, h and w; generate a set of candidate object boxes based on the deep learning framework RPN in Faster R-CNN, and the candidate object boxes represent the area containing rock obstacles; for each candidate object box b i , the RPN network outputs two values: p i Indicates the probability of whether the candidate object box contains an object, Indicates b i The four parameters to be adjusted, the output of the RPN network is expressed as: Where n represents the number of candidate object boxes; Classify and regress candidate object boxes. Based on Faster R-CNN, a network is used to classify and regress candidate object boxes. For each candidate object box b i , the Faster R-CNN network takes it as input and outputs a probability vector S i =(S i,1 , S i,2 , ..., S i,C ) and a regression vector B i =(B i,x , B i,y , B i,w , B i,h ), where C represents the number of rock obstacle categories; S i,C Indicates b i The probability of belonging to the Cth category, B i,x Indicates b i The regression value of the central horizontal axis, B i,y Indicates b i The regression value of the central ordinate, B i,w Indicates b i The regression value of width, B i,h Indicates b i Height of regression value; The output of the Faster R-CNN network is expressed as: The Faster R-CNN algorithm screens and performs non-maximum suppression on the candidate object boxes generated by the RPN network to obtain the final rock obstacle detection results.
