Path planning automatic driving control method for presetting carrying route of intelligent robot
By integrating multiple sensors and algorithms into AGV robots, real-time environmental perception and dynamic path planning are achieved, solving the problem of insufficient flexibility of traditional AGV systems in complex environments, improving transportation efficiency and accuracy, and reducing labor costs.
Patent Information
- Application Number
- CN202511139348.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-14
- Publication Date
- 2025-11-21
AI Technical Summary
Traditional AGV systems lack flexibility and adaptability when facing complex production or warehousing environments. They cannot perceive dynamic changes in real time, resulting in low transportation efficiency, easy collisions and congestion, and increased labor costs and management difficulty.
By employing sensors such as LiDAR, binocular depth camera, inertial navigation system (IMU), and ultrasonic radar, combined with ORB-SLAM algorithm and particle filter localization method, the AGV robot achieves real-time environmental perception and dynamic path planning. Through obstacle detection module and path planning module, it ensures that the AGV robot can avoid obstacles and travel along the predetermined route.
It improves the transportation efficiency and accuracy of AGV robots, reduces human intervention, lowers labor costs, and enhances the efficiency and safety of production line and warehouse management.
Smart Images
Figure CN120993909A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of path planning, and in particular to a path planning automatic driving control method. BACKGROUND
[0002] In the current industrial automation transportation system, AGV (Automated Guided Vehicle) has become a widely used core automated handling tool. AGV shuttles in many key fields such as warehousing, logistics and production lines, and plays an irreplaceable important role. In the field of warehousing, AGV can accurately and efficiently complete the in-out warehouse handling work of goods, improve the efficiency and accuracy of warehousing operation, and reduce the mistakes and labor intensity of manual operation. In the logistics link, AGV is a reliable helper for goods sorting and transportation. In the production line, AGV can ensure the continuity and stability of the production process.
[0003] However, the traditional AGV system has some drawbacks in the running process. Specifically, the traditional AGV system generally completes the transportation task through preset routes or manual intervention. Although the preset route ensures the standardization of transportation to a certain extent, it lacks flexibility. Once the production or warehousing environment changes, such as adjustment of shelf position, addition of new equipment, etc., the preset route needs to be re-planned and set, which often consumes a lot of time and manpower, and also affects the normal production. Manual intervention can solve the problem of route change to a certain extent, but manual operation has subjectivity and uncertainty, which can easily lead to untimely and inaccurate transportation tasks, and increase labor costs and management difficulty.
[0004] The existing path planning technology also has some shortcomings. The existing path planning technology usually relies on static route design. When planning the path, various dynamic change factors that may occur in the actual environment are not fully considered, and the ability of real-time perception and dynamic adjustment is lacking. It cannot obtain the information of the surrounding environment in time, and cannot optimize and adjust the path in real time according to the obtained information. When AGV faces complex production or warehousing environment, it cannot run efficiently and stably, and is prone to collision, congestion and other problems, which affects the overall performance and efficiency of the industrial automation transportation system. SUMMARY
[0005] In view of the poor flexibility and insufficient adaptability of the existing AGV system in path planning and automatic driving control, the application provides a path planning and automatic driving control method for preset carrying route of intelligent robot, and the designed AGV robot can accurately position and navigate through various sensors, laser radar, visual recognition and other technologies. The obstacle detection module is used to ensure that the AGV robot can avoid obstacles in a complex environment and travel along the predetermined route, thereby improving the transportation efficiency and accuracy. By setting multiple navigation points, the efficiency of production line and warehouse management is greatly improved.
[0006] In order to achieve the above purpose, the technical scheme of the application is as follows: A path planning and automatic driving control method for preset carrying route of intelligent robot, comprising an AGV robot body, a vehicle-mounted controller, a navigation module, an obstacle detection module, an alarm module, a communication module and a path planning module are arranged on the AGV robot body, the obstacle detection module is connected with the communication module, the communication module is connected with the vehicle-mounted controller, and the vehicle-mounted controller is also connected with the alarm module, the navigation module and the path planning module. The user sends an instruction to the communication module, the communication module processes the user instruction and sends it to the vehicle-mounted controller, according to the user instruction, the vehicle-mounted controller controls the obstacle detection module to obtain an environmental map, the path planning module plans a path, and the navigation module controls the AGV robot to navigate, in the process of moving, the obstacle detection module detects and scans the obstacles in the moving direction of the AGV robot and sends the data to the communication module, the communication module integrates the data and transmits the integrated result to the vehicle-mounted controller, the vehicle-mounted controller decides how to avoid the AGV robot, the alarm module monitors the state of the AGV robot at all times, and if a problem is found, an alarm information is immediately sent to the vehicle-mounted controller, and the vehicle-mounted controller controls the alarm module to send an alarm signal according to the alarm information. The specific steps are as follows: S1, start the power supply, the vehicle-mounted controller performs a series of self-checking operations, after the self-checking is completed, if the self-checking result is correct, step S2 is entered, if the self-checking has an error, an alarm information is sent; S2, according to the instruction information of the user, the obstacle detection module is started, the environmental information is obtained, the ORB-SLAM algorithm is used to construct an environmental map, and the real-time position of the AGV robot in the environmental map is determined; S3, according to the task target, the environmental map and the real-time position of the AGV robot in the environmental map, if there is a preset navigation route, the AGV robot is controlled to move towards the terminal along the preset navigation route, otherwise, the navigation module is used to plan a path, and a global path is generated; S4, according to the global path, the vehicle-mounted controller controls the AGV robot to navigate and control, controls the AGV robot to move towards the terminal, and completes the task.
[0007] Specifically, the obstacle detection module comprises a laser radar, a binocular depth camera, an inertial navigation system (IMU) and at least eight ultrasonic radars, and the laser radar, the binocular depth camera, the ultrasonic radars and the inertial navigation system (IMU) are connected with the communication module; the laser radar and the binocular depth camera are arranged at the front end of the AGV robot body, the inertial navigation system (IMU) is arranged at the center line point of the AGV robot body, and the ultrasonic radars are arranged around the AGV robot body.
[0008] Specifically, the self-checking operation comprises: hardware self-checking: checking whether each hardware component is working normally; software initialization: starting the control software, loading necessary programs and configurations including task scheduling, navigation and communication; sensor calibration: initializing and calibrating the laser radar, the binocular depth camera and the inertial navigation system (IMU).
[0009] Specifically, the alarm information is sent in the form of multi-level output, and the alarm classification and priority are as follows: Level 1, the type is divided into emergency alarm, the triggering scene includes battery overheating and emergency stop button triggering, and the corresponding strategy is to stop immediately, sound and light alarm and remote notification; Level 2, the type is divided into fault alarm, the triggering scene includes laser radar failure and motor stall, and the corresponding strategy is to suspend the task and try to restart the sensor; Level 3, the type is divided into warning information, the triggering scene includes map boundary approach and power less than 30%, and the corresponding strategy is to prompt the user and automatically plan the charging.
[0010] Specifically, the map type of the environment map is a grid map.
[0011] Specifically, when determining the real-time position of the robot in the map, a Monte Carlo localization method based on particle filtering is adopted, particle distribution is generated by random sampling, sensor data is matched with map features, and finally particle weights are updated according to the probability positioning mechanism of particle filtering to realize real-time positioning.
[0012] Specifically, the initial position of the AGV robot in the environment map is determined again each time the AGV robot starts.
[0013] Specifically, the path planning comprises global path planning and local path planning. The target of the global path planning is to search for the optimal path from the starting point to the ending point in the static map; when performing global path planning, the A* algorithm is adopted, and a heuristic function is introduced into the A* algorithm to speed up the path search speed. The local path planning aims at: real-time avoidance of dynamic obstacles, adjustment of the global path; in the local path planning, a dynamic window method is used, sampling in the feasible speed range is carried out, avoidance and target approaching abilities corresponding to each speed are evaluated, and the optimal speed is selected.
[0014] Specifically, when the AGV robot is navigated and controlled, a kinematics model of the differential wheel type robot is used to construct a theoretical basis; on the speed control, the rotating speeds of the left and right wheels are respectively vl , vr , the central speed v =2 vl + vr , the angular speed ω = dvr − vl , d is the wheel base; On the control algorithm, a PID control algorithm and a model predictive control algorithm are included.
[0015] Specifically, the model predictive control algorithm is based on the constructed kinematics model of the robot, and adopts a rolling optimization strategy to optimize and calculate the control amount of multiple future periods.
[0016] Compared with the prior art, the beneficial effects of the present application are: AGV robots can automatically carry and transport goods, greatly improving the efficiency of production line and warehouse management. Compared with traditional manual carrying, AGV robots can work 24 hours a day without interruption, and can automatically complete the transportation of goods according to the predetermined route and task, saving a lot of manpower and time. Since AGV robots can replace traditional manual transportation tasks, the demand for manual labor can be significantly reduced. In some large production and warehouse environments, the use of AGV robots helps to reduce labor costs and frees employees from heavy carrying work, allowing them to engage in more valuable work and improve safety. BRIEF DESCRIPTION OF DRAWINGS
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.
[0018] Figure 1 is a structural schematic diagram of the AGV robot of the present application.
[0019] Figure 2 is a control relationship schematic diagram of part of the modules in the AGV robot of the present application. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] Example 1 An automated driving control method for pre-planned transport routes of an intelligent robot includes an AGV robot body, as shown in Table 1. The AGV robot body is equipped with an onboard controller, a navigation module, a battery module, an obstacle detection module, an alarm module, a charging module, a communication module, and a path planning module. The charging module is connected to the battery module, the battery module is connected to the onboard controller, the obstacle detection module is connected to the communication module, the communication module is connected to the onboard controller, and the onboard controller is also connected to the alarm module, navigation module, and path planning module. Figure 1 and Figure 2 As shown.
[0022] Table 1 - Structural Composition of AGV Vehicle Body Name Function and composition Battery module Battery + Coulomb counter Obstacle detection module Single-line laser radar + ultrasonic radar + binocular depth camera + IMU Alarm module Acoustic-optic lamp Charging module Charging and discharging control board + fuse Communication module 8 Gb network ports + RS485 concentrator Path planning module Software Vehicle-mounted controller Motor driver + edge computing box Navigation module Software The onboard controller controls the AGV robot's movements by pre-setting multi-point coordinates. It receives real-time data from the navigation module, obstacle detection module, and battery module; adjusts motion parameters based on the output instructions from the path planning module; and triggers an alarm module when an abnormal state is detected. Control functions include: map management, path navigation, path planning, AGV guidance control, autonomous charging control, and alarm information management, promptly handling and providing feedback on various abnormal situations that occur during robot operation.
[0023] Map management: The parameter server dynamically switches the map path (rosparam set / map_server / map_file path / to / map.yaml), and automatically loads the corresponding map based on the initial positioning results.
[0024] Autonomous charging control: The charging pile emits infrared signals or ultrasonic waves of a specific frequency. The robot detects the signal strength through the corresponding sensor (ultrasonic sensor) and moves along the direction of signal enhancement.
[0025] AGV guidance and control: Multiple target coordinate points (such as [p1(x1,y1), p2(x2,y2),..., pn(xn,yn)]) are set manually or automatically, and the AGV performs operations one by one according to the above coordinates.
[0026] Navigation module, for building an environment map, determining the current spatial coordinates, and transmitting a dynamic path coordinate sequence to the vehicle controller based on the environment map.
[0027] Battery module, for providing power for all modules, and calculating the remaining power in real time by coulomb counting method, and triggering the charging module when the power is low.
[0028] Obstacle detection module, including: Laser radar (LiDAR): for three-dimensional environment scanning, generating point cloud data, realizing obstacle detection and distance measurement (such as the basis of SLAM algorithm).
[0029] Binocular depth camera: through image recognition to perceive the environment (such as object classification, semantic map construction), which can improve the accuracy of obstacle detection combined with deep learning (such as YOLO, SSD).
[0030] Ultrasonic radar: auxiliary near-range obstacle avoidance, low cost but limited accuracy, suitable for dynamic obstacle warning.
[0031] Inertial navigation system (IMU): measures the attitude and acceleration of the robot, used for motion trajectory calculation (dead reckoning method).
[0032] Specifically, the laser radar is arranged at the front end of the AGV robot body, the IMU is arranged at the center line point of the AGV robot body, the binocular depth camera is also arranged at the front end of the AGV robot body, and the eight ultrasonic radars are arranged around the AGV robot body. After installation, the coordinate positions are determined in the program, and the hardware does not need to cooperate, and the filtered and fused data is directly given to the algorithm.
[0033] Alarm module, for monitoring the hardware status, checking the software running situation, detecting the running status of other modules, sending alarm through sound and light signals (buzzer, LED light), or using the communication module to push alarm information to the cloud and monitoring center through HTTP protocol, prompting faults, low power or security threats; adopts multi-level output form to send alarm information, as shown in Table 2.
[0034] Table 2 - Alarm classification and priority: Priority Type Example scenario Response strategy Level 1 Emergency alarm Battery overheating, emergency stop button triggered Immediate shutdown + acoustic-optic alarm + remote notification Level 2 Fault alarm Laser radar failure, motor stall Pause task + try to restart sensor Level 3 Warning information Approaching map boundary, battery level below 30% Prompt user + automatically plan charging Charging module, for navigating to the charging pile and docking the charging interface when the battery module sends a low power signal; monitoring the charging state, preventing overcharging, and resuming task execution after being fully charged.
[0035] Communication module, for data interaction, receiving user instructions, uploading work logs or sensor data, supporting Wi-Fi6, Bluetooth 5.2 and 4G / 5G communication standards.
[0036] A path planning module is configured to generate an optimal path based on the map data of the navigation module and the detection results of the obstacle detection module, update the path in real time to avoid temporary obstacles, and optimize efficiency (e.g., avoid congested areas).
[0037] In addition, the AGV robot body is also provided with a touch screen for human-computer interaction, including output of instructions, display of AGV operation, etc.
[0038] The specific steps are as follows: S1, start the power supply, and the vehicle-mounted controller performs a series of self-checking operations, and after the self-checking is completed, if the self-checking result is correct, step S2 is entered; if there is an error in the self-checking, an alarm information is sent.
[0039] The self-checking operation includes: Hardware self-checking: check whether each hardware component is working normally, such as motor, battery, sensor, control board, etc.
[0040] Software initialization: start the control software, load the necessary programs and configurations such as task scheduling, navigation, communication, etc.
[0041] Sensor calibration: initialize and calibrate the laser radar, binocular depth camera, and inertial navigation system IMU.
[0042] S2, according to the instruction information of the user, start the obstacle detection module, obtain the environment information, construct the environment map by using the ORB-SLAM algorithm, and determine the real-time position of the AGV robot in the environment map.
[0043] The ORB-SLAM algorithm combines ORB feature points and Bundle Adjustment optimization, effectively improves the positioning accuracy and the accuracy of map construction, and provides strong support for subsequent navigation.
[0044] The map type of the environment map is a grid map (Grid Map).
[0045] When determining the real-time position of the robot in the map, a Monte Carlo localization method (MCL) based on particle filtering is adopted, which first randomly samples to generate particle distribution, then matches the sensor (such as laser radar) data with the map features (such as laser radar scanning matching), and finally updates the particle weight according to the probability positioning mechanism of particle filtering to realize real-time positioning and improve the accuracy and robustness of positioning.
[0046] Each time the AGV robot is started, the initial position of the AGV robot in the environment map needs to be determined again.
[0047] S3. Based on the task objective, the environment map, and the real-time position of the AGV robot in the environment map, if a preset navigation route exists, control the AGV robot to move towards the destination along the preset navigation route; otherwise, perform path planning and generate a global path.
[0048] The path planning includes global path planning and local path planning.
[0049] The goal of the global path planning is to search for the optimal path (shortest distance, least time, etc.) from the starting point to the destination in a static map. The A* algorithm, widely used in raster maps, is employed for global path planning, and heuristic functions (such as Manhattan distance) are introduced into the A* algorithm to accelerate path search.
[0050] The goal of the local path planning (dynamic obstacle avoidance) is to avoid dynamic obstacles in real time and adjust the global path. During local path planning, the Dynamic Window Method (DWA) is used: sampling is performed within the feasible speed range, the obstacle avoidance and target approach capabilities corresponding to each speed are evaluated, and the optimal speed is selected.
[0051] S4. Based on the global path, the on-board controller performs navigation control on the AGV robot, controlling the AGV robot to move towards the destination and complete the task.
[0052] When navigating and controlling AGV robots, a kinematic model of a differential wheel robot is used to construct the theoretical basis. For speed control: the left and right wheel speeds are respectively... vl , vr Center speed v =2 vl + vr angular velocity ω = dvr - vl ( d (Wheel width). The control algorithm includes PID control and Model Predictive Control (MPC). The PID control algorithm adjusts the motor speed to ensure the AGV robot can track the speed commands generated by the path planning. In terms of control structure, the PID control algorithm is divided into position loop control and speed loop control, which work together to ensure the accuracy and stability of the AGV robot's motion control.
[0053] The described Model Predictive Control (MPC) algorithm is based on the constructed robot kinematic model and employs a rolling optimization strategy to optimize the control quantities for multiple future cycles. The MPC algorithm is suitable for trajectory tracking tasks in AGV robots and can meet the trajectory tracking requirements in complex environments.
[0054] In the moving process, the vehicle-mounted controller decides the motion of the AGV robot according to sensor information, preset algorithm, motion planning and the like, monitors the running state of the AGV robot, checks battery capacity, motion trajectory, speed and the like parameters, and timely feeds back fault information.
[0055] The above merely describes preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A path planning and automatic driving control method for a smart robot with a preset transport route, characterized in that, The system includes an AGV robot body, which is equipped with an on-board controller, a navigation module, an obstacle detection module, an alarm module, a communication module, and a path planning module. The obstacle detection module is connected to the communication module, the communication module is connected to the on-board controller, and the on-board controller is also connected to the alarm module, the navigation module, and the path planning module. The user sends instructions to the communication module, which processes the instructions and sends them to the vehicle controller. Based on the user instructions, the vehicle controller controls the obstacle detection module to acquire an environmental map, the path planning module to plan a path, and the navigation module to navigate the AGV robot. During travel, the obstacle detection module scans for obstacles in the AGV robot's direction of travel and sends the data to the communication module. The communication module integrates the data and transmits the integrated result to the vehicle controller. The vehicle controller decides how the AGV robot should avoid obstacles. The alarm module constantly monitors the AGV robot's status and immediately sends an alarm message to the vehicle controller if a problem is detected. The vehicle controller then controls the alarm module to issue an alarm signal based on the alarm message. The specific steps are as follows: S1. Power on. The vehicle controller performs a series of self-tests. Once the self-test is complete, if the self-test results are correct, proceed to step S2. If an error is found during the self-test, an alarm message will be issued. S2. Based on the user's instructions, start the obstacle detection module, acquire environmental information, construct an environmental map using the ORB-SLAM algorithm, and determine the real-time position of the AGV robot in the environmental map. S3. Based on the task objective, the environment map, and the real-time position of the AGV robot in the environment map, if a preset navigation route exists, control the AGV robot to move towards the destination along the preset navigation route; otherwise, use the navigation module to perform path planning and generate a global path. S4. Based on the global path, the on-board controller performs navigation control on the AGV robot, controlling the AGV robot to move towards the destination and complete the task.
2. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 1, characterized in that, The obstacle detection module includes a lidar, a binocular depth camera, an inertial navigation system (IMU), and at least eight ultrasonic radars. The lidar, binocular depth camera, ultrasonic radars, and IMU are all connected to the communication module. The lidar and binocular depth camera are located at the front end of the AGV robot body, the IMU is located at the center point of the AGV robot body, and the ultrasonic radars are located around the AGV robot body.
3. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 2, characterized in that, The self-test operation includes: Hardware self-test: Checks whether each hardware component is working properly; Software initialization: Start the control software and load the necessary programs and configurations, including task scheduling, navigation, and communication. Sensor calibration: Initialize and calibrate the lidar, binocular depth camera, and inertial navigation system (IMU).
4. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 3, characterized in that, The alarm information is issued in a multi-level output format, and the alarm categories and priorities are as follows: Level 1 is classified as an emergency alarm, and the triggering scenarios include battery overheating and emergency stop button triggering. The corresponding strategies are immediate shutdown, audible and visual alarm plus remote notification. Level 2 is classified as a fault alarm, and the triggering scenarios include lidar failure and motor stall. The corresponding strategy is to pause the task and attempt to restart the sensor. Level 3, categorized as warning messages, is triggered by scenarios including approaching map boundaries and battery levels below 30%. The corresponding strategy is to prompt the user to add automatic charging planning.
5. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 4, characterized in that, The environmental map is a raster map.
6. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 5, characterized in that, To determine the robot's real-time location on the map, a particle filtering-based Monte Carlo localization method is used. First, random sampling is used to generate a particle distribution. Then, sensor data is matched with map features. Finally, the particle weights are updated based on the probabilistic localization mechanism of particle filtering to achieve real-time localization.
7. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 6, characterized in that, Each time the AGV robot starts, its initial position on the environmental map is redefined.
8. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 7, characterized in that, The path planning includes global path planning and local path planning; The goal of the global path planning is to search for the optimal path from the starting point to the destination in a static map. When performing global path planning, the A* algorithm is used, and a heuristic function is introduced into the A* algorithm to speed up the path search. The goal of the local path planning is to avoid dynamic obstacles in real time and adjust the global path. When performing local path planning, a dynamic window method is used. By sampling within the feasible speed range, the obstacle avoidance and target approach capabilities corresponding to each speed are evaluated, and the optimal speed is selected.
9. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 8, characterized in that, When navigating and controlling AGV robots, a kinematic model of a differential wheel robot is used to build the theoretical foundation; for speed control: the rotational speeds of the left and right wheels are respectively... vl , vr Center speed v =2 vl + vr angular velocity ω = DVR - vl , d Wheelbase; In terms of control algorithms, it includes PID control algorithms and model predictive control algorithms.
10. The path planning and automatic driving control method for a pre-set transport route of an intelligent robot according to claim 9, characterized in that, The model predictive control algorithm is based on the constructed robot kinematic model and adopts a rolling optimization strategy to optimize the control quantities for multiple future cycles.