Tunnel survey robot obstacle avoidance method and system based on context steering control

Abstract

The invention discloses a tunnel survey robot obstacle avoidance method and system based on context steering control, and the method comprises the steps: obtaining the terrain and obstacle position and shape information in a tunnel environment, carrying out the processing and analysis of the obtained data, recognizing the features of the terrain, detecting the position and shape of an obstacle, and extracting the context information of the tunnel environment. Forming a rasterized map; determining a global planning path for the robot by using an A * algorithm according to the context information of the tunnel environment; based on the global path information and the dynamic environment change, a context steering control algorithm is used to complete local path planning of the dynamic obstacle avoidance area, so that the robot can avoid dynamic obstacles in the tunnel environment; contextual information of the tunnel environment is continuously sensed and analyzed, and path planning and steering control measures are updated according to new information.

Description

Technical Field

[0001] This invention relates to the field of intelligent obstacle avoidance technology for tunnel surveying robots, and in particular to a method and system for obstacle avoidance of tunnel surveying robots based on context-sensitive steering control. Background Technology

[0002] A tunnel surveying robot is an intelligent robot used to perform surveying tasks in tunnel environments. During tunnel surveying, the robot needs to face various complex terrains, obstacles, and environmental changes. To ensure that the robot can complete its tasks safely and effectively, intelligent obstacle avoidance methods have become a key technology that urgently needs to be solved.

[0003] Invention patent: A method for robot target recognition and motion decision-making based on multi-feature spatiotemporal context, patent publication number: CN106127776B. Although it can achieve target recognition and tracking and use the tracking results as the basis for robot motion decision-making, this patent does not solve the problem of how to avoid dynamic obstacles in a tunnel environment during the tracking and recognition process.

[0004] Invention Patent: Improved Method for Smoothing Path Based on A-star Algorithm and Cubic B-spline Curve Fusion, Patent Publication No.: CN117740020A. Although it solves the problem of path smoothing, it does not solve how to achieve obstacle avoidance of dynamic obstacles in tunnel environment, nor does it achieve obstacle avoidance of robot by combining global path planning and local path planning.

[0005] The inventors discovered that traditional obstacle avoidance methods are mainly based on sensor data and preset rules, which only mechanically avoid static obstacles. In addition, the obstacle avoidance process will produce shaking, which cannot adapt to the complex and ever-changing tunnel environment and greatly increases the survey time. Moreover, existing methods usually lack full utilization of contextual information, and cannot achieve fast and accurate obstacle avoidance for tunnel robots. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for obstacle avoidance in tunnel surveying robots based on context-sensitive steering control. The aim is to overcome the limitations of traditional robot obstacle avoidance methods, which are susceptible to obstacle interference, and to improve the robot's obstacle avoidance capabilities and surveying efficiency in complex environments. By analyzing information such as the robot's surrounding environment and historical trajectories, this method can adaptively adjust the robot's trajectory to avoid collisions with obstacles and quickly complete surveying tasks. The goal of this invention is to improve the robot's intelligence level, enabling more efficient, accurate, and safer tunnel surveying work.

[0007] On the one hand, a method for obstacle avoidance in tunnel surveying robots based on context-based steering control is provided, including:

[0008] The system acquires information on the terrain, location, and shape of obstacles in the tunnel environment. It processes and analyzes the acquired data to identify terrain features, detect the location and shape of obstacles, extract contextual information about the tunnel environment, and form a raster map.

[0009] Based on the contextual information of the tunnel environment, the A* algorithm is used to determine a globally planned path for the robot;

[0010] Based on global path information and dynamic environmental changes, a context steering control algorithm is used to complete local path planning in the dynamic obstacle avoidance area, so as to enable the robot to avoid dynamic obstacles in the tunnel environment.

[0011] It continuously senses and analyzes the contextual information of the tunnel environment, and updates path planning and steering control strategies based on new information.

[0012] On the other hand, a tunnel exploration robot obstacle avoidance system based on context-based steering control is provided, including:

[0013] The acquisition module is configured to: acquire terrain, obstacle location and shape information in the tunnel environment, process and analyze the acquired data, identify terrain features, detect the location and shape of obstacles, extract contextual information of the tunnel environment, and form a raster map;

[0014] The global path planning module is configured to determine a globally planned path for the robot using the A* algorithm based on the context information of the tunnel environment.

[0015] The local path planning module is configured to: use a context steering control algorithm to complete local path planning in the dynamic obstacle avoidance area based on global path information and dynamic environmental changes, so as to enable the robot to avoid dynamic obstacles in the tunnel environment;

[0016] The update module is configured to continuously sense and analyze the contextual information of the tunnel environment and update the path planning and steering control strategies based on the new information.

[0017] Furthermore, an electronic device is also provided, including:

[0018] Memory, used for non-transitory storage of computer-readable instructions; and

[0019] Processor, for executing the computer-readable instructions,

[0020] When the computer-readable instructions are executed by the processor, they perform the method described in the first aspect above.

[0021] In another aspect, a storage medium is also provided for non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method described in the first aspect are executed.

[0022] In another aspect, a computer program product is also provided, including a computer program that, when run on one or more processors, is used to implement the method described in the first aspect above.

[0023] The above technical solution has the following advantages or beneficial effects:

[0024] This invention improves the autonomous obstacle avoidance capability of tunnel survey robots through efficient path planning using the A* algorithm and context-directed control algorithm. The robot can adjust its movement direction based on real-time sensor data in complex environments, avoid static and dynamic obstacles, find the shortest path, and improve survey efficiency. Global path planning is achieved through the A* algorithm, and local path planning is achieved through the context-directed control algorithm. By combining the A* algorithm and the context-directed control algorithm, the tunnel robot can achieve rapid and precise obstacle avoidance. Attached Figure Description

[0025] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0026] Figure 1 This is a flowchart outlining the intelligent obstacle avoidance method for tunnel surveying robots based on the A* algorithm and context steering control of the present invention.

[0027] Figure 2 This is the rasterized map of the present invention;

[0028] Figure 3 The following is a detailed flowchart of the intelligent obstacle avoidance method for tunnel surveying robots based on the A* algorithm and context steering control of the present invention.

[0029] Figure 4 This is a diagram showing the weight calculation of each vector direction in this invention;

[0030] Figure 5 This is a linear prediction graph of the vector direction where the maximum weight difference of the present invention is located;

[0031] Figure 6 This is a comparison diagram of the planned path and the actual driving path of the obstacle avoidance robot of the present invention. Detailed Implementation

[0032] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] Example 1

[0034] This embodiment provides an obstacle avoidance method for a tunnel survey robot based on context steering control;

[0035] like Figure 1 As shown, the obstacle avoidance method for tunnel surveying robots based on context-based steering control includes:

[0036] S101: Acquire terrain, obstacle location and shape information in the tunnel environment, process and analyze the acquired data, identify terrain features, detect the location and shape of obstacles, extract contextual information of the tunnel environment, and form a raster map;

[0037] S102: Based on the contextual information of the tunnel environment, use the A* algorithm to determine a global planning path for the robot;

[0038] S103: Based on global path information and dynamic environmental changes, the context steering control algorithm is used to complete the local path planning of the dynamic obstacle avoidance area, so as to enable the robot to avoid dynamic obstacles in the tunnel environment.

[0039] S104: Repeat S101 to S103, continuously perceive and analyze the contextual information of the tunnel environment, and update the path planning and implement steering control based on the new information.

[0040] Context-based intelligent obstacle avoidance methods utilize machine learning and perception technologies, combined with contextual information, to achieve more flexible and intelligent obstacle avoidance decisions. Specifically, this method dynamically adjusts the robot's navigation and movement strategies by perceiving and analyzing the terrain, obstacles, and other robot states in the tunnel environment. It can select appropriate paths and adjust directions based on current contextual information to avoid obstacles and safely complete the surveying task.

[0041] Further, in step S101: acquiring terrain, obstacle location and shape information in the tunnel environment, processing and analyzing the acquired data, identifying terrain features, detecting the location and shape of obstacles, extracting contextual information of the tunnel environment, and forming a raster map; wherein, the contextual information of the tunnel environment includes: terrain, obstacle location and shape information in the tunnel; obstacles include: passages with a width less than a set size, roadblocks, garbage, cables, mud, rocks, and moving vehicles.

[0042] Furthermore, the method includes: using lidar and cameras to acquire the position and shape information of static obstacles as the main body of static obstacle avoidance; the static obstacles include: terrain in the tunnel construction environment, passages with a width less than a set size, roadblocks, garbage, cables, and mud;

[0043] Dynamic obstacles are considered as the main body of dynamic obstacle avoidance during robot surveying; the dynamic obstacles include: fallen rocks or moving vehicles;

[0044] The static obstacle avoidance parameters and the dynamic obstacle avoidance parameters are input into the robot's camera to form a grid map. In the grid map, black grids represent static obstacles that the robot cannot cross or collide with; red grids represent slow-down zones where dynamically moving obstacles appear. The robot can pass through the slow-down zone but needs to slow down or prioritize alternative paths; white grids represent areas where the robot can move freely. The grid map is as follows: Figure 2 As shown.

[0045] To address the global path planning and dynamic obstacle avoidance problems of mobile robots in dynamic environments, this invention proposes a hybrid path planning method integrating the A* algorithm and a context-directed control algorithm. First, the A* algorithm is used to plan the global path. Then, based on the global path information and dynamic environmental changes, the context-directed control algorithm is used to complete the local path planning for the dynamic obstacle avoidance area. The flowchart is shown below. Figure 3 As shown.

[0046] The A* algorithm is an efficient algorithm for finding the shortest path between two points in a graph. It optimizes the search process by combining actual cost and estimated cost, and obtains the global path for the robot's movement through the A* algorithm.

[0047] Further, step S102: Based on the contextual information of the tunnel environment, using the A* algorithm to determine a globally planned path for the robot, including:

[0048] S102-1: Define the robot's starting position as the starting node, and define the robot's desired destination position as the target node;

[0049] S102-2: Initialize two lists: an Open list and a Close list. The Open list is used to record nodes to be explored, and the Close list is used to record nodes that have been explored.

[0050] S102-3: Calculate the total cost of the current node; the total cost of the current node is the sum of the actual movement cost from the starting node to the current node and the estimated movement cost from the current node to the target node; the actual movement cost from the starting node to the current node is the Chebyshev distance between the starting node and the current node; the estimated movement cost from the current node to the target node is the Chebyshev distance between the current node and the target node.

[0051] S102-4: Before each step the robot takes, select the node with the lowest total cost from the Open list for exploration. If the currently explored node is the target node, construct the path by backtracking its parent node and end the search.

[0052] S102-5: For the currently explored node, consider new nodes in all possible directions of robot movement. New nodes include: the next hop node for the robot to move in the straight direction from the current node and the next hop node for the robot to move in the diagonal direction. Evaluate the new nodes.

[0053] S102-6: Repeat S102-4 and S102-5.

[0054] Further, S102-5: Evaluating the new node, including:

[0055] 1) If a new node is an obstacle area, mark it as explored and exclude it from consideration;

[0056] 2) If the new node is a passable area but is not yet in the Open list, calculate the total cost of the new node and add it to the Open list;

[0057] 3) If the new node is already in the Open list, check if the cost of the current path is the lowest. If so, use the new node as the next hop node in the robot's movement.

[0058] For the yellow grid dynamic obstacle avoidance area, the context steering control algorithm is used to calculate the robot's turning direction. For each frame of the robot's movement, the next hop node calculated by the A* algorithm is used as the target node for the robot's movement, and the node where the dynamic obstacle is located is used as the danger node, indicating that the node is impassable.

[0059] Further, S103: Based on global path information and dynamic environmental changes, a context-based steering control algorithm is used to complete local path planning in the dynamic obstacle avoidance area, so as to enable the robot to avoid dynamic obstacles in the tunnel environment, specifically including:

[0060] S103-1: The number of robot divergence vector directions is 2. 4 Set the maximum distance threshold to D. maxWhen the distance between the robot and the target node exceeds D max The robot calculates the position of the next hop node using the A* algorithm. When the distance between the robot and the dangerous node exceeds D... max The robot will no longer avoid the dangerous node; the dangerous node is the node where the dynamic obstacle is located, and the dynamic obstacle refers to the rock that has fallen out in the tunnel;

[0061] S103-2: Create an obstacle avoidance mapping array and a target mapping array. The obstacle avoidance mapping array is used to store the distance between the current node and the dangerous node, and the target mapping array is used to store the distance between the current node and the target node.

[0062] S103-3: For the target mapping array, the distance weight between the i-th vector direction of the current node and the target node. The formula is as follows:

[0063]

[0064] Among them, I i For the current node 2 4 The distance between each vector direction and the target node;

[0065] S103-4: For the obstacle avoidance mapping array, the distance weight between the i-th vector direction of the current node and the target node. The formula is as follows:

[0066]

[0067] Among them, D i For the current node 2 4 The distance between each vector direction and the dangerous node;

[0068] S103-5: Calculate the weight difference ω for the i-th vector direction of the current node. i The formula is as follows:

[0069]

[0070] Robot weight calculation, such as Figure 4 As shown;

[0071] S103-6: Compare the weight difference ω i The direction of the vector containing the maximum value is denoted as max, and the weight difference along this vector direction is denoted as ω. maxLinear prediction is performed on the two vector directions preceding the direction of the vector containing `max` to obtain the first predicted line; linear prediction is then performed on the two vector directions following the direction of `max` to obtain the second predicted line; the intersection point of the first and second predicted lines is obtained, and the vector direction of the intersection point is denoted as `maxIndex`. `maxIndex` is the vector direction of the robot's next movement. Figure 5 As shown.

[0072] Further, S104: continuously sensing and analyzing the contextual information of the tunnel environment, and updating the path planning and steering control strategy based on the new information, includes:

[0073] S104-1: Determine whether the robot's current node is the target node of the local path planning. If it is the target node of the local path planning, it means that the robot has left the dynamic obstacle avoidance area, and then proceed to S104-2; if it is not the target point of the local path planning, it means that the robot is still in the dynamic obstacle avoidance area, and the context steering control algorithm is still used to complete the local path planning of the dynamic obstacle avoidance area.

[0074] S104-2: Determine whether the robot's current node is the target node of the global path planning. If yes, end the process; otherwise, return to S102.

[0075] The robot's final travel path is as follows Figure 6 As shown, the green line represents the path planned by the A* algorithm, and the red line represents the dynamic obstacle avoidance path during the robot's actual driving process.

[0076] It should be understood that the obstacle avoidance area is divided into global path planning and local path planning. The target point of local path planning is the destination of the local obstacle avoidance area, and the target point of global path planning is the final destination of the entire journey.

[0077] Example 2

[0078] This embodiment provides an obstacle avoidance system for a tunnel exploration robot based on context-sensitive steering control, including:

[0079] The acquisition module is configured to: acquire terrain, obstacle location and shape information in the tunnel environment, process and analyze the acquired data, identify terrain features, detect the location and shape of obstacles, extract contextual information of the tunnel environment, and form a raster map;

[0080] The global path planning module is configured to determine a globally planned path for the robot using the A* algorithm based on the context information of the tunnel environment.

[0081] The local path planning module is configured to: use a context steering control algorithm to complete local path planning in the dynamic obstacle avoidance area based on global path information and dynamic environmental changes, so as to enable the robot to avoid dynamic obstacles in the tunnel environment;

[0082] The update module is configured to continuously sense and analyze the contextual information of the tunnel environment and update the path planning and steering control strategies based on the new information.

[0083] It should be noted that the acquisition module, global path planning module, local path planning module, and update module mentioned above correspond to steps S101 to S104 in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should be noted that these modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0084] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0085] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0086] Example 3

[0087] This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory to cause the electronic device to perform the method described in Embodiment 1.

[0088] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0089] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0090] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0091] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0092] Those skilled in the art will recognize that the units and algorithm steps described in connection with the various examples of this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0093] Example 4

[0094] This embodiment also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the method described in Embodiment 1.

[0095] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A tunnel survey robot obstacle avoidance method based on contextual steering control, characterized in that: include: Obtain information about the terrain, obstacle location and shape in the tunnel environment, process and analyze the acquired data, identify terrain features, detect the location and shape of obstacles, extract contextual information about the tunnel environment, and form a rasterized map; Based on the contextual information of the tunnel environment, the A* algorithm is used to determine a global planning path for the robot; Based on the global path information and dynamic environment changes, the contextual steering control algorithm is used to complete the local path planning of the dynamic obstacle avoidance area, so that the robot can avoid dynamic obstacles in the tunnel environment. Continuously perceive and analyze the contextual information of the tunnel environment, and update the path planning and steering control strategies based on the new information.

2. The tunnel survey robot obstacle avoidance method based on contextual steering control as claimed in claim 1, characterized in that: Obtaining information on the terrain, location and shape of obstacles in the tunnel environment, processing and analyzing the acquired data, identifying terrain features, detecting the location and shape of obstacles, extracting context information of the tunnel environment, and forming a rasterized map; wherein the context information of the tunnel environment includes: information on the terrain, location and shape of obstacles in the tunnel; obstacles include: passages with a width less than a set size, roadblocks, garbage, cables, mud, rocks, and moving vehicles; Alternatively, the method includes: using a laser radar and a camera to obtain the position and shape information of static obstacles as the subject of static obstacle avoidance; the static obstacles include: terrain in a tunnel construction environment, passages with a width less than a set size, roadblocks, garbage, cables and mud; dynamic obstacles are regarded as the subject of dynamic obstacle avoidance during robot surveying; the dynamic obstacles include: fallen rocks or moving vehicles.

3. The obstacle avoidance method for a tunnel survey robot based on contextual steering control as claimed in claim 1, characterized in that: Based on the context information of the tunnel environment, the A* algorithm is used to determine a global planning path for the robot, including: (1): Define the robot's starting position as the starting node and the robot's expected position as the target node; (2): Initialize two lists: Open list and Close list. The Open list is used to record the nodes to be explored, and the Close list is used to record the nodes that have been explored. (3): Calculate the total cost of the current node; the total cost of the current node is: the sum of the actual moving cost from the starting node to the current node and the estimated moving cost from the current node to the target node; the actual moving cost from the starting node to the current node is the Chebyshev distance between the starting node and the current node; the estimated moving cost from the current node to the target node is the Chebyshev distance between the current node and the target node; (4): Before each step of the robot's movement, the node with the lowest total cost is selected from the Open list for exploration. If the node currently being explored is the target node, a path is constructed by backtracking its parent node, and the search ends; (5): For the currently explored node, consider new nodes in all possible movement directions of the robot. The new nodes include: the next hop node of the robot moving straight from the current node and the next hop node of the robot moving diagonally, and evaluate the new nodes; (6): Repeat (4) and (5).

4. The tunnel survey robot obstacle avoidance method based on contextual steering control as claimed in claim 3 is characterized in that: The new node is evaluated, including: 51) If the new node is an obstacle area, mark it as explored and exclude it from consideration; 52) If the new node is a traversable area but is not yet in the Open list, calculate the total cost of the new node and add it to the Open list; 53) If the new node is already in the Open list, check whether the cost of the current path is the lowest. If so, use the new node as the next hop node during the robot's walking process.

5. The tunnel survey robot obstacle avoidance method based on contextual steering control as claimed in claim 1, characterized in that: Based on the global path information and dynamic environment changes, the contextual steering control algorithm is used to complete the local path planning of the dynamic obstacle avoidance area, so that the robot can avoid dynamic obstacles in the tunnel environment, including: The number of robot divergence vector directions is 2 4 , set the maximum distance threshold to D max , when the distance between the robot and the target node exceeds D max The robot calculates the position of the next hop node according to the A* algorithm. When the distance between the robot and the dangerous node exceeds D max , the robot will no longer avoid the dangerous node; the dangerous node is the node where the dynamic obstacle is located, and the dynamic obstacle refers to the rock that falls off in the tunnel; Create an obstacle avoidance mapping array and a target mapping array. The obstacle avoidance mapping array is used to store the distance between the current node and the dangerous node, and the target mapping array is used to store the distance between the current node and the target node. For the target mapping array, the distance weight between the i-th vector direction of the current node and the target node The formula is as follows: Among them, I i 2 for the current node 4 The distance between the vector direction and the target node; For the obstacle avoidance mapping array, the distance weight between the i-th vector direction of the current node and the target node The formula is as follows: Among them, D i 2 for the current node 4 The distance between the vector direction and the dangerous node; Calculate the weight difference ω for the i-th vector direction of the current node i , the formula is as follows: Compare weight difference ω i The vector direction where the maximum value of is located is recorded as max, and the weight difference of this vector direction is recorded as ω max , perform linear prediction on the first two vector directions of the vector direction where max is located to obtain the first predicted straight line; perform linear prediction on the last two vector directions of the vector direction where max is located to obtain the second predicted straight line; obtain the intersection point between the first predicted straight line and the second predicted straight line, and the vector direction of the intersection point is recorded as maxIndex, and maxIndex is the vector direction of the robot's next move.

6. The obstacle avoidance method for a tunnel survey robot based on contextual steering control as claimed in claim 1, characterized in that: Continuously perceive and analyze the contextual information of the tunnel environment and update the path planning and steering control strategies based on the new information, including: Determine whether the node where the robot is currently located is the target node of the local path planning. If it is the target node of the local path planning, it means that the robot has left the dynamic obstacle avoidance area. Then determine whether the node where the robot is currently located is the target node of the global path planning. If it is not the target point of the local path planning, it means that the robot is still in the dynamic obstacle avoidance area, and the contextual steering control algorithm is still used to complete the local path planning of the dynamic obstacle avoidance area. Determine whether the node where the robot is currently located is the target node of the global path planning. If so, end the process. If not, use the A* algorithm to determine a global planning path for the robot based on the context information of the tunnel environment.

7. A tunnel survey robot obstacle avoidance system based on contextual steering control, characterized in that: include: An acquisition module is configured to: acquire terrain, obstacle location and shape information in a tunnel environment, process and analyze the acquired data, identify terrain features, detect obstacle location and shape, extract context information of the tunnel environment, and form a rasterized map; A global path planning module is configured to: determine a global planning path for the robot using an A* algorithm based on context information of the tunnel environment; A local path planning module is configured to: complete local path planning in a dynamic obstacle avoidance area using a contextual steering control algorithm based on global path information and dynamic environmental changes, so as to enable the robot to avoid dynamic obstacles in a tunnel environment; The update module is configured to continuously perceive and analyze the context information of the tunnel environment and update the path planning and steering control strategies according to the new information.

8. An electronic device, comprising: a memory for non-transitory storage of computer readable instructions; as well as a processor for executing the computer readable instructions, Wherein, when the computer-readable instructions are executed by the processor, the method described in any one of claims 1 to 6 is executed.

9. A storage medium characterized by being non-temporary The computer-readable instructions are stored, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method according to any one of claims 1 to 6 are executed.

10. A computer program product, characterized in that The invention comprises a computer program, which is used to implement the method according to any one of claims 1 to 6 when running on one or more processors.

Images