Inspection robot inspection track self-correction method based on SLAM (Simultaneous Localization and Mapping) navigation

Abstract

The invention relates to the technical field of robots, and discloses an SLAM navigation-based inspection robot inspection track self-correction method, which comprises the following steps: S1, acquiring environmental data through a 2D laser radar, an attitude sensor and an odometer sensor, and generating an initial environmental map and a navigation path by using an SLAM mapping algorithm; s2, fusing the laser radar pose data, the odometer pose data and the pose sensor pose data to obtain real-time pose information, comparing the current pose with the target path pose, and calculating a path error; s3, a robot motion instruction is adjusted through a compensation control algorithm based on the path error, and the path deviation is corrected; s4, continuously monitoring path errors and correcting the path errors in real time; and S5, storing the corrected track as a navigation path file. According to the invention, by introducing the robot remote controller, complex functions such as mapping, navigation, navigation point marking, path storage, motion control and the like are modularized, and the operation process is greatly simplified.

Description

Technical Field

[0001] The invention relates to the technical field of robots, and in particular to a self-correction method for inspection tracks of an inspection robot based on SLAM navigation. Background Art

[0002] With the development of intelligent robot technology, inspection robots have been widely used in industrial production, intelligent security, energy and electricity, etc., to replace manual inspection tasks that are highly repetitive and dangerous. These robots can automatically inspect equipment and facilities through autonomous navigation and environmental perception. However, traditional inspection robots still have some technical bottlenecks and limitations in the navigation and path control process.

[0003] Traditional inspection robots usually need to be operated through computers, tablets and other devices, which involves multiple steps such as map building, navigation path planning, and path marking. The operation process is relatively cumbersome. The system interface is complex and requires a high level of technical skills from the users, which increases the difficulty of robot deployment and maintenance and limits its promotion and use in non-professional fields. Summary of the invention

[0004] In order to make up for the above shortcomings, the present invention provides a self-correction method for the inspection trajectory of an inspection robot based on SLAM navigation, aiming to improve the problem that traditional inspection robots usually need to be operated through computers, tablets and other devices, and the operation process is relatively cumbersome.

[0005] In a first aspect, the present invention provides the following technical solution: a method for self-correcting inspection trajectories of an inspection robot based on SLAM navigation, comprising the following steps: S1, Mapping and Navigation Initialization: Collect environmental data through 2D LiDAR, attitude sensor and odometer sensor, and use SLAM mapping algorithm to generate initial environmental map and navigation path; S2, path tracking and posture error detection: Fuse the LiDAR posture data, odometer posture data and posture sensor posture data to obtain real-time posture information, compare the current posture with the target path posture, and calculate the path error; S3, trajectory self-correction and compensation control: adjust the robot motion instructions based on the path error through the compensation control algorithm to correct the path deviation; S4, path self-correction closed loop iteration: continuously monitor path errors and make real-time corrections; S5. Path saving and updating: Save the corrected trajectory as a navigation path file for subsequent inspection tasks.

[0006] Preferably, the SLAM mapping algorithm collects environmental point cloud data through 2D laser radar, uses a de-distortion algorithm to remove measurement errors caused by the color and material of the object, and outputs accurate laser radar data.

[0007] Preferably, the path tracking and posture error detection are completed by the following steps: S201, using a Kalman filter algorithm to fuse the laser radar posture data, the odometer posture data, and the posture data of the attitude sensor; S202, calculate path error and attitude error .

[0008] Preferably, the trajectory self-correction and compensation control is based on the PD control algorithm, and the compensation control formula is: Among them, U t Control output, K p is the proportional gain, K d is the differential gain, δP t is the current path error.

[0009] Preferably, the path self-correcting closed loop iteration comprises the following steps: S401, continuously adjusting motion control instructions based on error feedback; S402: Optimize and reduce path cumulative error.

[0010] Preferably, the path saving and updating includes saving the final corrected path trajectory as a navigation path point file for subsequent inspection.

[0011] Preferably, the functions are called by remote control trigger control, including mapping function, navigation function, navigation point marking function, path saving function and motion control function.

[0012] In a second aspect, the present invention provides the following technical solution, a patrol robot inspection trajectory self-correction system based on SLAM navigation, comprising: The sensor module is used to collect the robot's motion and environmental data in real time, including 2D laser radar, attitude sensor and odometer sensor. The laser radar is responsible for collecting environmental point cloud data, the attitude sensor provides the robot's current attitude angle information, and the odometer sensor obtains displacement data. Data processing module, used to process and fuse sensor data, including laser radar point cloud data dedistortion, laser pose calculation and Kalman filter data fusion; The trajectory error detection and compensation module is responsible for detecting the current path deviation of the robot and generating correction instructions. It calculates the path error and posture error by comparing the current posture with the target posture in real time, and generates motion compensation instructions using the PD control algorithm to adjust the robot's motion path and correct the trajectory deviation. The path self-correction and storage module is used to continuously detect and correct path errors in real time based on a closed-loop feedback system. The corrected path information is saved as a navigation path point file, which can be used for subsequent automatic navigation tasks. It also supports the updating and optimization of path data. The remote control trigger and feedback module is used to call system functions through the industrial remote control, including map building, navigation, marking navigation points, path saving and motion control functions. After the button is triggered, it provides corresponding voice prompt function to feedback the current operation status.

[0013] In the third aspect, the invention provides the following technical solution: a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned SLAM navigation-based inspection robot inspection trajectory self-correction method when executing the computer program.

[0014] In a fourth aspect, the present invention provides the following technical solution: a readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned SLAM navigation-based inspection robot inspection trajectory self-correction method.

[0015] The present invention has the following beneficial effects: 1. In the present invention, by introducing the robot remote control, complex functions such as mapping, navigation, marking navigation points, path saving and motion control are modularized, and each function can be triggered with one button, which greatly simplifies the operation process. Users do not need to rely on computers or tablets for visual operations, nor do they need to have high-level technical skills. The remote control provides intuitive operation and voice feedback, which reduces the technical threshold and training requirements of users, allowing non-professional users to quickly get started, improving the convenience of operation and the efficiency of task execution.

[0016] 2. In the present invention, through SLAM navigation technology, combined with lidar, attitude sensor and odometer data, the robot can accurately obtain environmental mapping and real-time location information; use Kalman filtering algorithm for multi-sensor data fusion, eliminate noise and interference, provide high-precision posture estimation, and significantly reduce path drift and cumulative errors; trajectory error detection and closed-loop self-correction control mechanism ensure that the robot can dynamically correct the path and improve the accuracy and stability of inspection tasks.

[0017] 3. In the present invention, by real-time monitoring of the error between the current position of the robot and the target path, the system can quickly identify the position deviation and heading angle deviation; the PD controller is used for error compensation, and the proportional control is used to quickly respond to the path deviation, and the differential control is used to suppress the violent oscillation of the error change, thereby effectively correcting the robot's motion trajectory; the error feedback closed-loop mechanism can continuously optimize the path so that the robot always stays on the planned path. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A method flow chart of the inspection trajectory self-correction method of the inspection robot based on SLAM navigation proposed by the present invention; Figure 2 This is a system architecture diagram of the inspection robot inspection trajectory self-correction system based on SLAM navigation proposed in the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0020] Embodiment 1 Reference Figure 1 In a first embodiment of the present invention, the present invention provides a self-correction method for inspection trajectory of an inspection robot based on SLAM navigation, comprising the following steps: S1, Mapping and Navigation Initialization: Collect environmental data through 2D LiDAR, attitude sensor and odometer sensor, and use SLAM mapping algorithm to generate initial environmental map and navigation path; S2, path tracking and posture error detection: Fuse the LiDAR posture data, odometer posture data and posture sensor posture data to obtain real-time posture information, compare the current posture with the target path posture, and calculate the path error; S3, trajectory self-correction and compensation control: adjust the robot motion instructions based on the path error through the compensation control algorithm to correct the path deviation; S4, path self-correction closed loop iteration: continuously monitor path errors and make real-time corrections; S5. Path saving and updating: Save the corrected trajectory as a navigation path file for subsequent inspection tasks.

[0021] Specifically, this method modularizes the complex functions of the inspection robot by introducing a robot remote control, simplifies the operation process by key triggering, and significantly reduces the technical requirements and training costs for users, which is specifically reflected in the following aspects: Button trigger to simplify the operation process The robot remote control divides the core functions into multiple independent buttons, including mapping, navigation, marking navigation points, path saving and motion control. Users only need to press the corresponding button to start each function with one click, without the need for complicated configuration.

[0022] For example, when the user presses the "Mapping" button, the robot can trigger the SLAM mapping function, scan the environment in real time and generate a map; when the user presses the "Navigate" button, the robot can automatically navigate along the saved path without additional operation.

[0023] Lower technical threshold and reduce operational errors Traditional inspection robots require the use of computers or tablets to perform complex visual operations, and users need to have a high level of technical skills and proficiency. This method simplifies the operation process through a remote control, allowing non-professional users to quickly get started.

[0024] The remote control is intuitive and easy to operate. Combined with the voice feedback function, it can provide real-time prompts of the current operation status, such as "map creation started" and "path saving completed", effectively avoiding operational errors and improving usage efficiency.

[0025] Reduce dependence on external devices and improve convenience Using a remote control instead of the traditional human-computer interaction interface eliminates the reliance on external devices such as computers and tablets. Users can control the robot functions anytime and anywhere, making operation more convenient.

[0026] The remote control transmits data with the robot through wireless communication, and the user does not need to access complicated control programs, further reducing the difficulty of operation.

[0027] Improve task execution efficiency and reduce training costs The simplified operation mode allows users to master the basic functional operations of the robot in a short time without the need for complex training, thus reducing training time and costs.

[0028] At the same time, the convenient operation of the remote control improves the efficiency of task execution, allowing operators to focus more on the inspection task itself without having to spend energy debugging the robot path or configuring system parameters.

[0029] The SLAM mapping algorithm collects environmental point cloud data through 2D lidar, uses a de-distortion algorithm to remove measurement errors caused by object color and material, and outputs accurate lidar data.

[0030] Specifically, this step focuses on mapping the environment data collected by 2D LiDAR to resolve data errors caused by factors such as object color, material, and reflectivity. Specifically, it includes: Point cloud data collection: LiDAR obtains a large amount of point cloud data by scanning the environment. Each point cloud contains distance and direction information.

[0031] Data de-distortion: In order to address data distortion caused by light reflection, object surface characteristics, etc. during the LiDAR scanning process, the system identifies and eliminates abnormal data.

[0032] Data compensation and correction: For the blank areas after removing abnormal data, interpolation compensation is performed based on the surrounding valid data to reconstruct complete point cloud data.

[0033] Posture data calculation: Convert the processed laser point cloud data into the position and posture information of the robot in the environmental coordinate system to ensure the accuracy of the mapping data and provide a high-precision map for subsequent navigation.

[0034] Path tracking and pose error detection are completed through the following steps: S201, using a Kalman filter algorithm to fuse the laser radar posture data, the odometer posture data, and the posture data of the attitude sensor; S202, calculate path error and attitude error .

[0035] Specifically, this step improves the accuracy of the robot's position information by multi-sensor data fusion. The main process is as follows: Multi-source data acquisition: obtain position information from the lidar, obtain speed and position information from the odometer, and obtain the robot's attitude angle and angular velocity information from the attitude sensor.

[0036] Data fusion processing: Use the Kalman filter algorithm to fuse the data of each sensor to remove random noise and errors in the sensor collected data and improve the reliability of the posture data.

[0037] Error detection: The system compares the real-time position information with the target path point information to determine the robot's current path deviation, including position deviation and posture deviation, providing a basis for subsequent trajectory correction.

[0038] The trajectory self-correction and compensation control is based on the PD control algorithm, and the compensation control formula is: Among them, U t Control output, K p is the proportional gain, K d is the differential gain, δP t is the current path error.

[0039] Specifically, this step corrects the robot's deviated path through error compensation control so that it returns to the predetermined planned path, which specifically includes: Error evaluation: The system calculates the deviation between the robot's current position and the target position in real time, and generates a control amount based on the degree of deviation.

[0040] PD control compensation: In the PD control algorithm, the proportional term is used to quickly respond to position errors, and the differential term is used to control the rate of error change to prevent oscillation or large offsets.

[0041] Motion control execution: The system sends the compensated motion instructions to the robot execution layer, adjusts the robot's direction and speed in real time, and ensures that the robot gradually returns to the planned path.

[0042] The path self-correction closed loop iteration includes the following steps: S401, continuously adjusting motion control instructions based on error feedback; S402: Optimize and reduce path cumulative error.

[0043] Specifically, this step continuously optimizes the path error through a closed-loop feedback mechanism to prevent the cumulative error from affecting the task accuracy: Real-time feedback path error: During the robot execution process, the system continuously monitors the difference between the current position and the target path point and generates feedback information.

[0044] Posture optimization: Kalman filtering is used to optimize the feedback posture data, eliminate interference noise, and ensure the continuity and accuracy of posture data.

[0045] Motion command adjustment: Dynamically adjust the robot's motion control commands based on feedback errors to reduce the deviation of the current path.

[0046] Iterative closed-loop correction: The error correction process is ongoing, ensuring that the robot path continues to approach the ideal planned path, ultimately achieving accurate inspection.

[0047] Path saving and updating includes saving the final corrected path trajectory as a navigation path point file for subsequent inspections.

[0048] Specifically, this step records and stores the corrected path information during the inspection process, including: Waypoint data preservation: The system stores the waypoint data passed by the robot, including position information and attitude angle information, in chronological order as waypoint files.

[0049] Path file update: supports incremental update and full path rewriting of path files to meet the needs of different inspection tasks.

[0050] Navigation path reuse: The saved path file can be used for subsequent tasks. The robot can automatically perform inspection tasks according to the path without re-building the map or planning the path.

[0051] Function calls are made through remote trigger control, including mapping function, navigation function, navigation point marking function, path saving function and motion control function.

[0052] Specifically, this step uses the remote control to conveniently control system functions and provide voice feedback: Remote control button functions: Mapping function: trigger the SLAM algorithm to generate a map of the current environment.

[0053] Navigation function: Start the robot to perform navigation tasks according to the saved path point files.

[0054] Mark navigation point function: record the current location of the robot and save it as a navigation path point.

[0055] Path saving function: save the current inspection track and path point file.

[0056] Motion control function: Control the robot's forward, backward, left turn, right turn and other motion instructions through the remote control.

[0057] Voice prompt function: After the button is triggered, the system voice prompts the current function status, such as "start building map" and "path saving completed", etc., providing clear feedback information to facilitate user operation.

[0058] Embodiment 2: Reference Figure 2 In a second embodiment of the present invention, the present invention provides a patrol robot inspection trajectory self-correction system based on SLAM navigation, comprising: The sensor module is used to collect the robot's motion and environmental data in real time, including 2D laser radar, attitude sensor and odometer sensor. The laser radar is responsible for collecting environmental point cloud data, the attitude sensor provides the robot's current attitude angle information, and the odometer sensor obtains displacement data. Data processing module, used to process and fuse sensor data, including laser radar point cloud data dedistortion, laser pose calculation and Kalman filter data fusion; The trajectory error detection and compensation module is responsible for detecting the current path deviation of the robot and generating correction instructions. It calculates the path error and posture error by comparing the current posture with the target posture in real time, and generates motion compensation instructions using the PD control algorithm to adjust the robot's motion path and correct the trajectory deviation. The path self-correction and storage module is used to continuously detect and correct path errors in real time based on a closed-loop feedback system. The corrected path information is saved as a navigation path point file, which can be used for subsequent automatic navigation tasks. It also supports the updating and optimization of path data. The remote control trigger and feedback module is used to call system functions through the industrial remote control, including map building, navigation, marking navigation points, path saving and motion control functions. After the button is triggered, it provides corresponding voice prompt function to feedback the current operation status.

[0059] Specifically, 1. Sensor module Detailed description: The sensor module is used to obtain the robot's surrounding environment information and its own motion state data in real time, providing basic data support for SLAM mapping and path correction. It mainly includes: 2D LiDAR: Scans the surrounding environment and obtains point cloud data, which contains information such as the distance and direction of objects, and serves as the core input for SLAM mapping.

[0060] Inertial Measurement Unit (IMU): includes an accelerometer and a gyroscope, used to measure the acceleration, angular velocity, and attitude angle (Roll, Pitch, Yaw) of the robot in three-dimensional space.

[0061] Odometer sensor: provides the robot's moving speed and displacement information, which is used to estimate the robot's current position information.

[0062] Data synchronization function: ensures that the data collection time of all sensors is synchronized to ensure the accuracy of fused data and avoid inaccurate positioning due to timing errors.

[0063] 2. Data processing module Detailed description: The data processing module is responsible for processing, fusing and optimizing the raw data collected by the sensor to ensure the high-precision requirements of SLAM navigation and path correction. The main functions include: LiDAR data dedistortion: Detect abnormal points in laser point cloud data (such as distortion caused by object material, reflectivity, etc.).

[0064] Invalid data is removed, and data compensation is performed on the missing areas to reconstruct complete point cloud data.

[0065] Laser pose calculation: The dedistorted point cloud data is used in combination with the robot motion information to calculate the robot's position information (x, y) and heading angle in the environment coordinate system.

[0066] Data Fusion: The Kalman filter algorithm is used to fuse the lidar posture, attitude sensor data, and odometer data to remove random noise and improve the accuracy of posture information.

[0067] Real-time posture output: The fused robot position and posture data are provided to the path error detection module and trajectory compensation module.

[0068] 3. Trajectory error detection and compensation module Detailed description: This module is used to detect the deviation of the robot's current path and correct the robot's motion trajectory through compensation control, including: Error detection: Compare the robot's current position information with the target path point information, and calculate the position deviation (path error) and attitude deviation (direction error) in real time.

[0069] Error types include: lateral deviation, longitudinal deviation and heading angle deviation.

[0070] Compensation Control: The PD controller is used to compensate the path error. Proportional control is used to quickly correct the position deviation, and differential control is used to suppress the path oscillation caused by the error change.

[0071] The compensation motion instruction is calculated according to the error size and sent to the motion control execution unit to correct the robot's running path.

[0072] Real-time feedback: The error feedback closed loop continuously optimizes the compensation control instructions to ensure that the robot moves smoothly along the planned path.

[0073] 4. Path self-correction and storage module Detailed description: The path self-correction and storage module is responsible for correcting the path in real time based on error feedback and saving the corrected path information. The main functions include: Path closed loop self-correction: Based on the real-time path error, the robot's motion direction and speed are dynamically adjusted through feedback control.

[0074] Use the optimized posture data to correct the robot's trajectory and gradually reduce the path deviation.

[0075] Path information storage: The corrected path points (position coordinates and attitude angles) of the robot during the inspection process are saved as a path file in chronological order.

[0076] Supports path incremental update and full path coverage to meet different inspection needs.

[0077] Path data management: Provides path data storage, loading and update functions to facilitate subsequent inspection task calls and reduce the workload of repeated mapping and path planning.

[0078] 5. Remote control trigger and feedback module Detailed description: The remote control trigger and feedback module is responsible for invoking system functions and implementing execution status feedback through the remote control. Its main functions include: Remote control function assignment: Mapping button: triggers the SLAM mapping function, generates and saves map files in real time.

[0079] Navigation button: Start the navigation function and control the robot to move along the preset path points.

[0080] Mark navigation point button: record the current position of the robot as a navigation path point and save it to the path file.

[0081] Path save button: save the current mapping results and inspection path information.

[0082] Motion control buttons: control the robot's forward, backward, left turn, right turn and other movements.

[0083] Wireless communication function: The remote control sends the user key signal to the robot controller through wireless communication to trigger the corresponding system function module.

[0084] Voice feedback prompt: After receiving the remote control button signal, the system provides voice feedback to prompt the status of the currently executed function, such as "start map building", "path saving completed", "navigation started", etc.

[0085] The voice feedback function allows users to understand the system operation status in real time and avoid operational errors.

[0086] Summary of system module workflow The sensor module collects point cloud data of the robot's surrounding environment and its own motion status data to provide real-time input for the system.

[0087] The data processing module performs de-distortion processing and Kalman filter fusion on the sensor data, and outputs high-precision robot position and posture information.

[0088] The trajectory error detection and compensation module monitors the path deviation in real time and generates compensation motion instructions to correct the trajectory through the PD controller.

[0089] The path self-correction and storage module continuously optimizes the path based on the error feedback closed loop and saves the corrected trajectory data as a path file.

[0090] The remote control trigger and feedback module triggers system functions through the remote control and confirms the current operation status through voice feedback.

[0091] Embodiment 3 The third embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the inspection trajectory self-correction method of the inspection robot based on SLAM navigation in the above-mentioned embodiment.

[0092] Embodiment 4 The fourth embodiment of the present invention is based on the same inventive concept. A computer device proposed by the present invention comprises: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory, and execute the inspection trajectory self-correction method of the inspection robot based on SLAM navigation of the above embodiment.

[0093] Embodiment 5 1. System initialization and status definition When the system is initialized, the robot's state vector is defined, including position, velocity, attitude, acceleration bias, angular velocity bias and gravity vector, specifically: in: P t : Robot position status.

[0094] V t : Robot speed status.

[0095] R t : Robot attitude angle status.

[0096] b a : Accelerometer bias.

[0097] b g : Angular velocity sensor bias.

[0098] g t : gravity vector.

[0099] 2. State update equation The update of the robot's posture depends on the IMU data (acceleration a and angular velocity ω), combined with the state equation as follows: 1. Location status update: 2. Speed ​​status update: in: a: acceleration obtained by IMU.

[0100] b a : Acceleration bias.

[0101] M a : Acceleration noise.

[0102] g: gravity vector.

[0103] 3. Posture status update: in: ω: angular velocity obtained by IMU.

[0104] b g : Angular velocity bias.

[0105] M g : Angular velocity noise.

[0106] (⋅) ∧ : Represents the transformation from a vector to an antisymmetric matrix.

[0107] 4. Bias update: The update relationship of the bias term is: 5. Gravitational constant: 3. Error state equation and simplification The error state equation of the system is defined according to the handwritten content: 1. Position error equation: 2. Speed ​​error equation: 3. Attitude error equation: 4. Bias error update: Acceleration bias error: Angular velocity bias error: 5. Gravity error update: 4. Error compensation control Combined with the error state, the PD controller is used to calculate the compensation control amount and correct the robot path: 1. Path error calculation: Position error: Attitude error: 2. Calculation of compensation control quantity: Using PD control algorithm: in: Kp: Proportional gain, adjusts the real-time response speed of the error.

[0108] Kd: Differential gain, suppresses rapid changes in error.

[0109] δPt: Position error.

[0110] 5. Noise Model The noise model is defined as follows: 1. Acceleration noise: 2. Angular velocity noise: 3. Bias noise: 6. Status and Observation Updates 1. Status update: Through the state transfer matrix F: in: F: state transfer matrix.

[0111] W: process noise.

[0112] 2. Observation update: Combined with observation model: in: Z: observation vector.

[0113] h(x): observation model.

[0114] 7. Error Covariance Propagation The formula for error covariance propagation is: in: J k : Jacobian matrix, expressed as: Jacobian matrix of attitude error: 8. Remote control trigger and path storage Function trigger: The following functions are triggered by the remote control: Mapping function Navigation Mark navigation point function Path saving function Motion control function Path storage: Save the corrected path points as a navigation path file: Voice prompts: After the remote control triggers a button, the system provides voice feedback to prompt the current function execution status.

[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0116] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-correction method for inspection robot inspection trajectory based on SLAM navigation, characterized in that: The following steps are involved: S1, Mapping and Navigation Initialization: Collect environmental data through 2D LiDAR, attitude sensor and odometer sensor, and use SLAM mapping algorithm to generate initial environmental map and navigation path; S2, path tracking and posture error detection: Fuse the LiDAR posture data, odometer posture data and posture sensor posture data to obtain real-time posture information, compare the current posture with the target path posture, and calculate the path error; S3, trajectory self-correction and compensation control: adjust the robot motion instructions based on the path error through the compensation control algorithm to correct the path deviation; S4, path self-correction closed loop iteration: continuously monitor path errors and make real-time corrections; S5. Path saving and updating: Save the corrected trajectory as a navigation path file for subsequent inspection tasks.

2. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: The SLAM mapping algorithm collects environmental point cloud data through 2D laser radar, uses a de-distortion algorithm to remove measurement errors caused by object color and material, and outputs accurate laser radar data.

3. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: The path tracking and posture error detection are completed by the following steps: S201, using a Kalman filter algorithm to fuse the laser radar posture data, the odometer posture data, and the posture data of the attitude sensor; S202, calculate path error and attitude error .

4. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: The trajectory self-correction and compensation control is based on the PD control algorithm, and the compensation control formula is: Among them, U t Control output, K p is the proportional gain, K d is the differential gain, δP t is the current path error.

5. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: The path self-correcting closed loop iteration comprises the following steps: S401, continuously adjusting motion control instructions based on error feedback; S402: Optimize and reduce path cumulative error.

6. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: The path saving and updating includes saving the final corrected path trajectory as a navigation path point file for subsequent inspections.

7. The inspection robot inspection trajectory self-correction method based on SLAM navigation according to claim 1 is characterized in that: Function calls are made through remote trigger control, including mapping function, navigation function, navigation point marking function, path saving function and motion control function.

8. The inspection robot inspection trajectory self-correction system based on SLAM navigation is characterized by: The inspection robot inspection trajectory self-correction method based on SLAM navigation as described in any one of claims 1 to 7 comprises: The sensor module is used to collect the robot's motion and environmental data in real time, including 2D laser radar, attitude sensor and odometer sensor. The laser radar is responsible for collecting environmental point cloud data, the attitude sensor provides the robot's current attitude angle information, and the odometer sensor obtains displacement data. Data processing module, used to process and fuse sensor data, including laser radar point cloud data dedistortion, laser pose calculation and Kalman filter data fusion; The trajectory error detection and compensation module is responsible for detecting the current path deviation of the robot and generating correction instructions. It calculates the path error and posture error by comparing the current posture with the target posture in real time, and generates motion compensation instructions using the PD control algorithm to adjust the robot's motion path and correct the trajectory deviation. The path self-correction and storage module is used to continuously detect and correct path errors in real time based on a closed-loop feedback system. The corrected path information is saved as a navigation path point file, which can be used for subsequent automatic navigation tasks. It also supports the updating and optimization of path data. The remote control trigger and feedback module is used to call system functions through the industrial remote control, including map building, navigation, marking navigation points, path saving and motion control functions. After the button is triggered, it provides corresponding voice prompt function to feedback the current operation status.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the inspection trajectory self-correction method of the inspection robot based on SLAM navigation as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by the processor, the inspection trajectory self-correction method of the inspection robot based on SLAM navigation as described in any one of claims 1 to 7 is implemented.

Images