Navigation method for robot to quickly arrive at specified target point

Abstract

The invention belongs to the technical field of mobile robot navigation, and particularly discloses a navigation method for a robot to quickly arrive at a specified target point, which comprises the following steps of: creating a point location, drawing a preset path, selecting the shortest path in the preset path as a current routing inspection path, and performing strict routing inspection in a navigation task execution process. When long-time parking caused by obstacles is found in the inspection process, a rapid arrival algorithm is executed to rapidly plan a new path, key point location information is obtained, a new global path is generated according to the key point location information, and then the robot is controlled by a planner to move so as to continue the inspection task. According to the navigation method for the robot to rapidly arrive at the designated target point, the possibility that the robot stops midway is greatly reduced while it is guaranteed that the robot strictly advances according to the preset path, the inspection efficiency is improved while the safety is guaranteed, and the navigation method for the robot to rapidly arrive at the designated target point can be applied to various mobile platforms such as a wheel type mobile platform, a crawler type mobile platform and a leg-foot type mobile platform.

Description

Technical Field

[0001] This invention belongs to the field of mobile robot navigation technology, and specifically discloses a navigation method for a robot to quickly reach a designated target point. Background Technology

[0002] In special operating environments such as petroleum and chemical industries, which are flammable, explosive, and highly corrosive, the core principle for ensuring the safety of inspection robots is to strictly follow the preset global path. In such environments, there is dense equipment, crisscrossing pipelines, and high-risk areas such as leakage risk areas and high-temperature and high-pressure reaction devices. If the robot autonomously generates a navigation path through a real-time replanning algorithm, it may expose it to unknown or high-risk areas, thereby greatly increasing the probability of failure or loss of control.

[0003] Therefore, during inspections, robots often need to follow a preset path. However, this strongly constrained path mode has certain limitations in actual operation: First, strictly following a fixed route during inspections can easily cause the robot to stop midway due to temporary obstacles, leading to inspection interruptions; second, when performing urgent or special inspection tasks, the existing system cannot generate the optimal path in real time to move from the current position to the preset target point along the shortest path; in addition, if the robot encounters an insurmountable obstacle along the way, it cannot quickly return to the charging station along the optimal route.

[0004] Therefore, in order to meet the above requirements, it is necessary to design a navigation method that strictly follows the path, that is, strictly follows the route, and can quickly reach the designated point, so as to ensure that the optimal path can be continuously selected from multiple inspection routes until the target point is reached. Summary of the Invention

[0005] To address the problems in the background art, this invention discloses a navigation method for a robot to quickly reach a designated target point. First, points are created and a preset path is drawn. During the navigation task, the robot strictly follows the path. If an obstacle is found during the inspection process, causing the robot to stop for a long time, a fast arrival algorithm is executed to quickly plan a new path. This ensures that the robot strictly follows the preset path while greatly reducing the possibility of the robot stopping midway, thus improving inspection efficiency while ensuring safety.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A navigation method for a robot to quickly reach a designated target point includes the following steps: Step 1: Create a location. The operator drives the robot to the designated location and marks key locations on the map using a mobile app. The key locations include waypoints, navigation points, and charging points. Step 2: Draw a preset path. After setting the points, connect the key points in sequence to generate multiple preset paths. The preset path includes several key points and path segments between the key points. The key points and preset paths together constitute complete path topology information, i.e., path map. Step 3: Perform navigation task. The robot first leaves the charging pile and starts to perform inspection task after leaving the pile. It selects the shortest path in the preset path as the current inspection path. During the inspection, the robot strictly follows the current inspection path. Step 4: During the inspection process, the robot monitors the status of obstacles in the path in real time. When an obstacle is detected that causes the robot to stop and the dwell time exceeds a preset threshold, the fast arrival algorithm is executed. Step 5: Load the path map data and calculate the path length between any two adjacent key points. The path map contains information on all key points in the map and preset path information. Step 6: Construct a fast arrival request, starting from the current point and ending at the target point in the task queue. Add a path segment from the current point back to the previous transit point in the path graph, and delete the currently impassable inspection path segments to obtain a new path graph. Step 7: Execute Dijkstra's algorithm in the new path graph to find the shortest path from the starting point to the destination; Step 8: Obtain key point information in the new path, including point name, whether reversing is required, robot pose, and path curvature; Step 9: Generate a new global path based on the key point information, and then control the robot to switch paths through the planner to continue the inspection task; Step 10: If the robot continues to encounter obstacles on the new path and is unable to move for a long time, repeat steps 5-9 until the robot successfully moves to the target point. Step 11: If a new path cannot be found in step 7, then find the shortest path between the current point and the upper pile point, return to the upper pile point, and perform upper pile charging.

[0007] Furthermore, in the navigation method for the robot to quickly reach the designated target point, in step 1, the waypoints are used to constrain the preset path and ensure that the robot strictly follows the preset path; the navigation points are used to instruct the robot to identify specific targets in order to complete the inspection task; and the charging point is the location where the robot docks with the charging pile for automatic charging. At this point, the robot will temporarily turn off the obstacle avoidance function to ensure that the charging is completed smoothly.

[0008] Furthermore, in the navigation method for the robot to quickly reach the designated target point, in step 2, the preset path supports dynamic configuration of the robot's travel direction and trajectory curvature parameters according to actual needs. If there are multiple independent paths, more complex inspection task logic can be generated through the path combination function.

[0009] Furthermore, in the navigation method for the robot to quickly reach the designated target point, in step 2, to prevent the robot from being interrupted due to a single path blockage, redundant connection paths are set between key inspection points to ensure task continuity.

[0010] Furthermore, the navigation method for the robot to quickly reach a designated target point is based on a mobile robot equipped with a LiDAR, inertial measurement unit, encoder, and industrial control computer hardware platform.

[0011] Furthermore, the navigation method for the robot to quickly reach the designated target point allows for rapid inspection of both temporary and special inspection tasks. The robot uses a rapid arrival algorithm to search for the shortest path between the current point and the target point.

[0012] By adopting the above technical solution, the present invention has the following beneficial effects: The navigation method for a robot to quickly reach a designated target point disclosed in this invention first creates points and draws a preset path. The shortest path among the preset paths is selected as the current inspection path. During the navigation task, the robot strictly follows the path. If an obstacle is found during the inspection, causing a long-term stop, a fast arrival algorithm is executed to quickly plan a new path, obtain key point information, and generate a new global path based on the key point information. Then, the planner controls the robot's movement to continue the inspection task. This ensures that the robot strictly follows the preset path while greatly reducing the possibility of the robot stopping midway. It improves inspection efficiency while ensuring safety. The navigation method for a robot to quickly reach a designated target point disclosed in this invention can be applied to various mobile platforms such as wheeled, tracked, and legged robots. Attached Figure Description

[0013] Figure 1 This is a flowchart of the inspection task execution process for the navigation method of the robot to quickly reach the designated target point according to the present invention. Figure 2 This is a flowchart of the fast arrival algorithm in the navigation method for robots to quickly reach a specified target point in this invention; Figure 3 This is a schematic diagram of the path switching in the rapid arrival algorithm of the navigation method for robots to quickly reach a specified target point in this invention. Detailed Implementation

[0014] The following is in conjunction with the appendix Figure 1-3The navigation method for a robot to quickly reach a designated target point disclosed in this invention will be further described.

[0015] This invention discloses a navigation method for robots to quickly reach designated target points. It is implemented using a mobile robot equipped with hardware platforms such as LiDAR, IMU, encoder, and industrial control computer. The method is supported by the ROS (Robot Operating System) framework. The robot utilizes multi-sensor fusion positioning technology to acquire its own pose information on the map in real time and with high accuracy. Users can conveniently mark key points via a mobile app and connect these points sequentially to generate a preset path. During the inspection process, the robot strictly follows the preset path. When the robot receives an inspection command, it first departs from the charging station and, upon reaching the departure point, formally begins the inspection task. Figure 1 As shown, its specific operation process is as follows: Step 1: Create points. During deployment, multi-sensor fusion technology is used to acquire the robot's centimeter-level precise pose on the map in real time. Specifically, after the operator navigates the robot to the target point, they trigger a pose recording command via a mobile app. The system automatically parses the real-time transformation parameters TF from the robot's chassis coordinate system to the map coordinate system, completing the storage of spatial coordinates and attitude data for that point. The point information is recorded and stored in a JSON file for later use. The created inspection points are divided into three categories: waypoints, navigation points, and charging points. Waypoints are used to constrain the robot's trajectory, ensuring it strictly follows the preset path. Navigation points are used to identify specific target objects that the robot needs to recognize. Charging points are dedicated locations for performing automatic charging operations. At these locations, the robot will temporarily disable obstacle avoidance to complete the charging docking. Step 2: Draw the preset path. Connect the points created in Step 1 in sequence through the mobile APP to form a complete preset path. Multiple redundant connection paths can be configured between key inspection points to ensure task continuity. During the drawing process, the robot's travel direction and trajectory curvature parameters can be dynamically set. All path information and point data are stored in structured JSON format, which together constitute a dynamically callable path topology network, i.e., a path map. Step 3: After receiving the inspection instruction, the robot executes the navigation task. The robot first leaves the charging pile and begins to perform the inspection task after leaving the pile. During the inspection, the robot strictly follows the preset path with the shortest path. For situations where multiple path tasks need to be performed simultaneously, multiple paths can be combined to form a combined path. The purpose of this is to make the configuration of inspection tasks more flexible and to adjust the inspection order and scope according to actual needs. Step 4: During the inspection process, the robot will detect obstacles in real time. If an obstacle causes the robot to stop for more than a set threshold, the fast arrival algorithm will be executed. The flowchart of the fast arrival algorithm is as follows: Figure 2 As shown; Step 5: Load the path map data and calculate the path length between any two adjacent key points. The path map contains all preset paths and point information on the map. A schematic diagram of the preset paths is shown below. Figure 3 As shown, it includes straight paths and curved paths; Step 6, construct a fast-delivery request, such as Figure 3 As shown, when the robot encounters an obstacle while moving from point B to point C, the current position H will be used as the starting point for the fast arrival request, and the target point E in the task queue will be used as the ending point for the fast arrival request. The path segment H→B from the current point to the previous passing point is added to the path graph, while the currently impassable path segment B→C is deleted to obtain a new path graph. Step 7: Switch paths and execute Dijkstra's algorithm in the new path graph to find the shortest path from the starting point to the destination. If multiple feasible paths of equal length exist, the robot will execute the candidate path with the highest priority. The robot will comprehensively evaluate path priority based on factors such as obstacle distribution characteristics and path familiarity (historical traversal frequency). Figure 3 As shown, the inspection path will be changed from B→C→D→E to B→F→G→C→D→E; Step 8: Obtain key location information in the new path, including the location name, whether reversing is required, robot pose (i.e., position and orientation), and path curvature information. Step 9: Generate a new global path based on the key point information, and then control the robot to move through the planner to continue the inspection task; Step 10: If the robot continues to encounter obstacles on the new path and is unable to move for a long time, repeat steps 5-9 until the robot successfully moves to the target point. Step 11: If a new path cannot be found in step 7, then find the shortest path between the current point and the upper pile point, return to the upper pile point, and perform upper pile charging.

[0016] For ad-hoc inspection tasks, the robot can use a rapid arrival algorithm to find the shortest path between the current point and the target point, thus achieving rapid inspection.

[0017] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A navigation method for a robot to quickly reach a designated target point, characterized in that, Includes the following steps: Step 1: Create a location. The operator drives the robot to the designated location and marks key locations on the map using a mobile app. The key locations include waypoints, navigation points, and charging points. Step 2: Draw a preset path. After setting the points, connect the key points in sequence to generate multiple preset paths. The preset path includes several key points and path segments between the key points. The key points and preset paths together constitute complete path topology information, i.e., path map. Step 3: Perform navigation task. The robot first leaves the charging pile and starts to perform inspection task after leaving the pile. It selects the shortest path in the preset path as the current inspection path. During the inspection, the robot strictly follows the current inspection path. Step 4: During the inspection process, the robot monitors the status of obstacles in the path in real time. When an obstacle is detected that causes the robot to stop and the dwell time exceeds a preset threshold, the fast arrival algorithm is executed. Step 5: Load the path map data and calculate the path length between any two adjacent key points. The path map contains information on all key points in the map and preset path information. Step 6: Construct a fast arrival request, starting from the current point and ending at the target point in the task queue. Add a path segment from the current point back to the previous transit point in the path graph, and delete the currently impassable inspection path segments to obtain a new path graph. Step 7: Execute Dijkstra's algorithm in the new path graph to find the shortest path from the starting point to the destination; Step 8: Obtain key location information in the new path, including the location name, whether reversing is required, robot pose, and path curvature. Step 9: Generate a new global path based on the key point information, and then control the robot to switch paths through the planner to continue the inspection task; Step 10: If the robot continues to encounter obstacles on the new path and is unable to move for a long time, repeat steps 5-9 until the robot successfully moves to the target point. Step 11: If a new path cannot be found in step 7, then find the shortest path between the current point and the upper pile point, return to the upper pile point, and perform upper pile charging.

2. The navigation method for a robot to quickly reach a designated target point according to claim 1, characterized in that, In step 1, waypoints are used to constrain the preset path and ensure that the robot strictly follows the preset path; navigation points are used to instruct the robot to identify specific targets in order to complete the inspection task; and charging points are the locations where the robot docks with charging piles for automatic charging. At these locations, the robot will temporarily disable obstacle avoidance to ensure successful charging.

3. The navigation method for a robot to quickly reach a designated target point according to claim 2, characterized in that, In step 2, the preset path supports dynamic configuration of the robot's travel direction and trajectory curvature parameters according to actual needs. If there are multiple independent paths, the path combination function can generate more complex inspection task logic.

4. The navigation method for a robot to quickly reach a designated target point according to claim 3, characterized in that, In step 2, to prevent the robot from being interrupted due to a single path blockage, redundant connection paths are set between key inspection points to ensure task continuity.

5. The navigation method for a robot to quickly reach a designated target point according to claim 3, characterized in that, The navigation method is based on a mobile robot equipped with a LiDAR, inertial measurement unit, encoder and industrial control computer hardware platform.

6. The navigation method for a robot to quickly reach a designated target point according to claim 3, characterized in that, For temporary and special inspection tasks, the robot uses a rapid arrival algorithm to find the shortest path between the current point and the target point, enabling rapid inspection.

Images