Path planning optimization method for underground coal mine bracket transfer robot

Abstract

The invention discloses a coal mine underground support transfer robot path planning optimization method, and belongs to the technical field of path planning optimization, and the method comprises the following steps: firstly, obtaining information and data; according to the method, through environment information obtained by a sensor and the relation between a starting point and a target point, an off-line path library generated in advance is traversed in sequence, a path with the maximum probability is found to serve as a feasible path, a path containing an obstacle is provided, and the steps are iterated in planning until the target point is reached. According to the method, the optimal path of the current position is not generated in real time under a mine, the optimal path which is most matched with the current position in the off-line trajectory library is retrieved, the problem that the real-time operation speed of an industrial personal computer in a roadway environment is low in a traditional method is effectively solved, the path planning speed and quality can be improved, and the working safety is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of path planning optimization, and in particular relates to a path planning optimization method for a support transport robot in an underground coal mine. Background Art

[0002] The coal mine support handling robot is a robot device specially used for handling supports in the coal mine environment. During the underground mining process, miners need to move and adjust supports to support the stability of the coal mine tunnel. Traditional support handling work is usually done manually by miners, which not only poses safety risks but also has high labor intensity.

[0003] The support handling robot in coal mines is equipped with various sensors, cameras, and robotic arms, and has autonomous perception and operation capabilities. It can move autonomously in the coal mine environment, accurately locate and grab the support, and then move it to the designated location. The robot can also communicate with other devices to achieve remote control and monitoring of the support.

[0004] However, traditional path planning methods require detailed modeling and calculation of the environment, including map construction, path search, and cost functions. In lanes, accurate modeling and calculation are difficult and time-consuming due to the complexity and uncertainty of the environment. In addition, traditional path planning methods often only consider the shortest path or the fastest path, ignoring safety considerations. This may cause the robot to choose an unsafe path, increase the risk of accidents, and have poor practical application effects. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that traditional path planning methods require detailed modeling and calculation of the environment, including map construction, path search and cost function, etc. In the tunnel, due to the complexity and uncertainty of the environment, accurate modeling and calculation are difficult and time-consuming. In addition, traditional path planning methods often only consider the shortest path or the fastest path, ignoring safety considerations. This may cause the robot to choose an unsafe path and increase the risk of accidents. A path planning optimization method for a support handling robot in an underground coal mine is proposed.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a path planning optimization method for a support handling robot in an underground coal mine, comprising the following steps: S1. Obtain information and data first; S2. Process the acquired information and data; S3, using the acquired information and data, and referring to the state grid algorithm to generate an offline trajectory library; S4. Model the planning environment; S5. Filter paths containing obstacles.

[0007] Furthermore, the method uses the environmental information obtained by the sensor and the relationship between the starting point and the target point to traverse the pre-generated offline path library in sequence, find the path with the highest probability as the feasible path, propose a path containing obstacles, and iterate the above steps in the planning until the target point. That is, it does not generate the optimal path for the current position in real time underground, but retrieves the optimal path that best matches the current position in the offline trajectory library. This effectively solves the problem of slow real-time computing speed of industrial computers in tunnel environments in traditional methods, and can also improve the speed and quality of path planning to ensure work safety.

[0008] As a further description of the above technical solution: In S1, information and data are first acquired, and the acquired information and data include map information of the coal mine, specification parameters of the transport robot, and operation parameters of the transport robot.

[0009] As a further description of the above technical solution: In S2, the acquired information and data are processed, and the specific steps of the processing are: first, the data and information are flattened, the data and information are screened one by one, abnormal information and data are screened out and their positions are marked, the screened abnormal information and data are deleted, the deleted abnormal information and data are backtracked, and the deleted abnormal information is filled in, and finally the processed information and data are classified and grouped and numbered, and the group numbering adopts one or more of digital numbering and alphabetical numbering. The software required for data and information processing is one or more of FoneLabDataRetriever software, WiseDataRecovery software, and WindowsFileRecovery software.

[0010] As a further description of the above technical solution: In the S3, the acquired information and data are used to generate an offline trajectory library with reference to the state grid algorithm, and the specific steps are as follows: according to the prior map, the continuous two-dimensional space is divided into square grids of equal size, each grid represents a discrete state, and is uniquely identified by coordinates (x, y); according to the environment in which the grid is located, a state value is assigned to each grid to represent the accessibility or cost of the grid; the A* algorithm is used to find the optimal path from the starting point to the end point on the grid map; the search algorithm will select the next direction of advance according to the state value of the current grid and the heuristic function; and the search space is continuously expanded until the optimal path from the starting point to the end point is found.

[0011] As a further description of the above technical solution: In S3, for the grid where the obstacle is located, its state value is set to positive infinity, indicating that it is unreachable; for the idle passing grid, its state value is set to a positive number, reflecting the cost of passing through the grid.

[0012] As a further description of the above technical solution: In S4, the planning environment is modeled as follows: R is the global environment, Q R is the environment occupied by the robot, M Q is the robot's starting position, N Q is the end point position, is the possible direction that the robot can drive to at the starting position M, is the possible direction that the robot can drive to at the edge position L, S Q is the sensing range of the sensor, the obstacle is O, and L is the edge of the sensor's sensing range. is a set of trajectories with the largest probability value, is the probability of reaching the end position N from the starting position, | is the probability of the edge position reaching the end position N, is the probability of the edge position reaching the end position N, is the probability of the sample edge position reaching the end position N.

[0013] As a further description of the above technical solution: In S4, the formula used in modeling the planning environment is as follows: |

[0014] according to | Specify n The sample is And the Monte Carlo sampling theorem yields: |

[0015] So we get the simplified formula: .

[0016] As a further description of the above technical solution: In S5, the paths containing obstacles are filtered: Assume the following formula to determine whether there is an obstacle on the current path: ; Introduce the shown In, we get: .

[0017] As a further description of the above technical solution: In S5, if a certain path is occupied by an obstacle, the probability of the path is 0 and it is not considered during the search. The path with the highest value is the robot execution path.

[0018] As a further description of the above technical solution: In S5, the environmental information obtained by the sensor and the relationship between the starting point and the target point are used to traverse the pre-generated offline path library in turn, find the path with the highest probability as a feasible path, propose a path containing obstacles, and iterate the above steps in the planning until the target point. That is, the optimal path for the current position is not generated in real time under the mine, but the optimal path that best matches the current position in the offline trajectory library is retrieved.

[0019] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: In the present invention, the method uses the environmental information obtained by the sensor and the relationship between the starting point and the target point to traverse the offline path library generated in advance in sequence, find the path with the highest probability as a feasible path, propose a path containing obstacles, and iterate the above steps in the planning until the target point. That is, it does not generate the optimal path of the current position in real time under the mine, but retrieves the optimal path that best matches the current position in the offline trajectory library. This effectively solves the problem of slow real-time computing speed of industrial computers in tunnel environments in traditional methods, and can also improve the speed and quality of path planning to ensure work safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The present invention is a flow chart of a path planning optimization method for a support handling robot in underground coal mines. DETAILED DESCRIPTION

[0021] 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.

[0022] Example 1: Please refer to Figure 1 The present invention provides a technical solution: a path planning optimization method for a support handling robot in a coal mine, comprising the following steps: S1. First, information and data are acquired, including map information of the coal mine, specification parameters of the handling robot, and operation parameters of the handling robot; S2. Process the acquired information and data. The specific steps of the processing are: first, flatten the data and information, screen the data and information one by one, screen out abnormal information and data and mark their positions, delete the screened abnormal information and data, trace back the deleted abnormal information and data, fill in the deleted abnormal information, and finally classify the processed information and data and group them. The group numbers are digital numbers. The software required for data and information processing is FoneLabDataRetriever software; S3. Using the acquired information and data, refer to the state grid algorithm to generate an offline trajectory library. The specific steps are as follows: according to the prior map, the continuous two-dimensional space is divided into square grids of equal size. Each grid represents a discrete state and is uniquely identified by the coordinates (x, y); according to the environment in which the grid is located, a state value is assigned to each grid to represent the accessibility or cost of the grid; the A* algorithm is used to find the optimal path from the starting point to the end point on the grid map; the search algorithm will select the next direction of advance based on the state value of the current grid and the heuristic function; the search space is continuously expanded until the optimal path from the starting point to the end point is found. For the grid where the obstacle is located, its state value is set to a positive infinity, indicating that it is unreachable; for the idle passing grid, its state value is set to a positive number, reflecting the cost of passing the grid; S4. Model the planning environment as follows: R is the global environment, Q R is the environment occupied by the robot, M Q is the robot's starting position, N Q is the end point position, is the possible direction that the robot can drive to at the starting position M, is the possible direction that the robot can drive to at the edge position L, S Q is the sensing range of the sensor, the obstacle is O, and L is the edge of the sensor's sensing range. is a set of trajectories with the largest probability value, is the probability of reaching the end position N from the starting position, | is the probability of the edge position reaching the end position N, is the probability of the edge position reaching the end position N, is the probability of the sample edge position reaching the end position N; The formulas used in modeling the planning environment are as follows: |

[0023] according to | Specify n The sample is And the Monte Carlo sampling theorem yields: |

[0024] So we get the simplified formula: ; S5. Filter paths containing obstacles: Assume the following formula to determine whether there is an obstacle on the current path: ; Introduce the shown In, we get: ; If one of the paths is occupied by an obstacle, the probability of the path is 0 and it will not be considered during the search. The path with the highest value is the robot's execution path. The environmental information obtained by the sensor and the relationship between the starting point and the target point are used to traverse the pre-generated offline path library in turn to find the path with the highest probability as the feasible path. A path containing obstacles is proposed, and the above steps are iterated in the planning until the target point. That is, the optimal path for the current position is not generated in real time by the mine, but the optimal path that best fits the current position in the offline trajectory library is retrieved.

[0025] Example 2: Please refer to Figure 1 The present invention provides a technical solution: a path planning optimization method for a support handling robot in a coal mine, comprising the following steps: S1. First, information and data are acquired, including map information of the coal mine, specification parameters of the handling robot, and operation parameters of the handling robot; S2. Process the acquired information and data. The specific steps of the processing are: first, flatten the data and information, screen the data and information one by one, screen out abnormal information and data and mark their positions, delete the screened abnormal information and data, trace back the deleted abnormal information and data, fill in the deleted abnormal information, and finally classify the processed information and data and group them. The group numbers are alphabetical. The software required for data and information processing is WiseDataRecovery software; S3. Using the acquired information and data, refer to the state grid algorithm to generate an offline trajectory library. The specific steps are as follows: according to the prior map, the continuous two-dimensional space is divided into square grids of equal size. Each grid represents a discrete state and is uniquely identified by the coordinates (x, y); according to the environment in which the grid is located, a state value is assigned to each grid to represent the accessibility or cost of the grid; the A* algorithm is used to find the optimal path from the starting point to the end point on the grid map; the search algorithm will select the next direction of advance based on the state value of the current grid and the heuristic function; the search space is continuously expanded until the optimal path from the starting point to the end point is found. For the grid where the obstacle is located, its state value is set to a positive infinity, indicating that it is unreachable; for the idle passing grid, its state value is set to a positive number, reflecting the cost of passing the grid; S4. Model the planning environment as follows: R is the global environment, Q R is the environment occupied by the robot, M Q is the robot's starting position, N Q is the end point position, is the possible direction that the robot can drive to at the starting position M, is the possible direction that the robot can drive to at the edge position L, S Q is the sensing range of the sensor, the obstacle is O, and L is the edge of the sensor's sensing range. is a set of trajectories with the largest probability value, is the probability of reaching the end position N from the starting position, | is the probability of the edge position reaching the end position N, is the probability of the edge position reaching the end position N, is the probability of the sample edge position reaching the end position N; The formulas used in modeling the planning environment are as follows: |

[0026] according to | Specify n The sample is And the Monte Carlo sampling theorem yields: |

[0027] So we get the simplified formula: ; S5. Filter paths containing obstacles: Assume the following formula to determine whether there is an obstacle on the current path: ; Introduce the shown In, we get: ; If one of the paths is occupied by an obstacle, the probability of the path is 0 and it will not be considered during the search. The path with the highest value is the robot's execution path. The environmental information obtained by the sensor and the relationship between the starting point and the target point are used to traverse the pre-generated offline path library in turn to find the path with the highest probability as the feasible path. A path containing obstacles is proposed, and the above steps are iterated in the planning until the target point. That is, the optimal path for the current position is not generated in real time by the mine, but the optimal path that best fits the current position in the offline trajectory library is retrieved.

[0028] In the above embodiment, the method effectively solves the problem of slow real-time computing speed of industrial computers in the lane environment of the traditional method, and can also improve the speed and quality of path planning to ensure work safety.

[0029] In Example 1-Example 2, it should be noted that: In S3, the acquired information and data are used to generate an offline trajectory library with reference to the state grid algorithm. The specific steps are as follows: according to the prior map, the continuous two-dimensional space is divided into square grids of equal size, each grid represents a discrete state, and is uniquely identified by the coordinates (x, y); according to the environment in which the grid is located, a state value is assigned to each grid to represent the accessibility or cost of the grid; the A* algorithm is used to find the optimal path from the starting point to the end point on the grid map; the search algorithm will select the next direction of advance according to the state value of the current grid and the heuristic function; the search space is continuously expanded until the optimal path from the starting point to the end point is found, wherein: The A* algorithm determines the current search direction by evaluating the distance from each node to the target node and finds the shortest path.

[0030] It uses the heuristic function f(n)=g(n)+h(n) to evaluate the priority of each node, where: g(n) represents the actual cost (the length of the path traveled) from the starting point to the current node n.

[0031] h(n) represents the estimated cost (heuristic function) from the current node n to the target node.

[0032] The A* algorithm selects the node with the smallest f(n) value for expansion each time, and eventually finds the shortest path from the starting point to the target node.

[0033] Algorithm steps: Add the starting point to the open list (PriorityQueue).

[0034] Repeat the following steps until the target node is found or cannot be expanded further: Select the node n with the smallest f(n) value from the open list.

[0035] Move node n from the open list to the closed list.

[0036] For each neighboring node m of n: If m is in the closed list, skip it.

[0037] If m is not in the open list, add it to the open list and calculate g(m) and f(m).

[0038] If m is already in the open list, check whether the path to m via the current node n is better (i.e. g(m) is smaller). If so, update m's parent node and g(m) and f(m) values.

[0039] If the target node is found, backtrack along the parent node pointer to obtain the shortest path.

[0040] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A path planning optimization method for a support handling robot in a coal mine, characterized in that: The steps include: S1. Obtain information and data first; S2. Process the acquired information and data; S3, using the acquired information and data, and referring to the state grid algorithm to generate an offline trajectory library; S4. Model the planning environment; S5. Filter paths containing obstacles.

2. A method for optimizing path planning of a support handling robot in an underground coal mine according to claim 1, characterized in that: In S1, information and data are first acquired, and the acquired information and data include map information of the coal mine, specification parameters of the transport robot, and operation parameters of the transport robot.

3. The path planning optimization method for a support handling robot in an underground coal mine according to claim 1, characterized in that: In S2, the acquired information and data are processed, and the specific steps of the processing are: first, the data and information are flattened, the data and information are screened one by one, abnormal information and data are screened out and their positions are marked, the screened abnormal information and data are deleted, the deleted abnormal information and data are backtracked, and the deleted abnormal information is filled in, and finally the processed information and data are classified and grouped and numbered, and the group numbering adopts one or more of digital numbering and alphabetical numbering. The software required for data and information processing is one or more of FoneLabDataRetriever software, WiseDataRecovery software, and WindowsFileRecovery software.

4. The path planning optimization method for a support handling robot in an underground coal mine according to claim 1, characterized in that: In the S3, the acquired information and data are used to generate an offline trajectory library with reference to the state grid algorithm, and the specific steps are as follows: according to the prior map, the continuous two-dimensional space is divided into square grids of equal size, each grid represents a discrete state, and is uniquely identified by coordinates (x, y); according to the environment in which the grid is located, a state value is assigned to each grid to represent the accessibility or cost of the grid; the A* algorithm is used to find the optimal path from the starting point to the end point on the grid map; the search algorithm will select the next direction of advance according to the state value of the current grid and the heuristic function; and the search space is continuously expanded until the optimal path from the starting point to the end point is found.

5. A path planning optimization method for a support handling robot in a coal mine according to claim 4, characterized in that: In S3, for the grid where the obstacle is located, its state value is set to positive infinity, indicating that it is unreachable; for the idle passing grid, its state value is set to a positive number, reflecting the cost of passing through the grid.

6. A method for optimizing path planning of a support handling robot in an underground coal mine according to claim 1, characterized in that: In S4, the planning environment is modeled as follows: R is the global environment, Q R is the environment occupied by the robot, M Q is the robot's starting position, N Q is the end point position, is the possible direction that the robot can drive to at the starting position M, is the possible direction that the robot can drive at the edge position L, S Q is the sensing range of the sensor, the obstacle is O, and L is the edge of the sensor's sensing range. is a set of trajectories with the largest probability value, is the probability of reaching the end position N from the starting position, | is the probability of the edge position reaching the end position N, is the probability of the edge position reaching the end position N, is the probability of the sample edge position reaching the end position N, where the formula used in modeling the planning environment is as follows: | ； according to | Specify n The sample is And the Monte Carlo sampling theorem yields: | ； So we get the simplified formula: 。 7. The path planning optimization method for a support handling robot in an underground coal mine according to claim 1, characterized in that: In S5, the paths containing obstacles are filtered: Assume the following formula to determine whether there is an obstacle on the current path: ; Introduce the shown In, we get: .

8. A method for optimizing path planning of a support handling robot in an underground coal mine according to claim 7, characterized in that: In S5, if a certain path is occupied by an obstacle, the probability of the path is 0 and it is not considered during the search. The path with the highest value is the robot execution path.

9. A method for optimizing path planning of a support handling robot in an underground coal mine according to claim 8, characterized in that: In S5, the environmental information obtained by the sensor and the relationship between the starting point and the target point are used to traverse the pre-generated offline path library in turn, find the path with the highest probability as a feasible path, propose a path containing obstacles, and iterate the above steps in the planning until the target point. That is, the optimal path for the current position is not generated in real time under the mine, but the optimal path that best matches the current position in the offline trajectory library is retrieved.

Images