Motion Control System Design Of Crawler Coal Mine Rescue Robot

With the development of the robot technology,robots have been applied in all of our life.In this epidemic of new crown pneumonia,various indoor killing robots have played a vital role in disinfecting and sterilizing the indoor environment,avoiding medical care.Risk of infection of personnel.Also in the coal mine environment,robots also have great application prospects.China is a large coal country and the demand for coal resources is huge.However,China's coal mining is generally difficult and accidents occur frequently.Therefore,an emergency rescue assistance is urgently needed.This type of robot usually needs to work in a mine disaster environment with unknown environment,so the robot environment mapping and path planning are its core technologies.This paper uses crawler robot as the experimental platform,and its software and hardware are designed to meet its navigation requirements.This article will use the ROS operating system as the development framework,which is mainly divided into two major blocks,namely,mapping and navigation.Its software system is mainly divided into closed-loop control of the robot chassis,global and local path planning,and mapping of unknown environments.Through ROS distributed communication,each function is modularized,and the control function is loosely coupled to facilitate the expansion and development of robot functions.(1)After necessary modeling and analysis of the robot,the kinematics model,attitude solution model,and map model are mainly analyzed.The purpose is to enable the closed-loop control of robot motion and the modeling of the mine environment.According to the SLAM part,the theoretical analysis methods of particle mapping,Gauss-Newton method,and graph optimization are respectively analyzed.Based on the above algorithms,Gmapping,Hector-slam,and cartographer are transplanted.Based on these three algorithms,GAZEBO physical simulation platform is transplanted.After verification,the results show that the Gmapping algorithm relies too much on the odometer;Hector-slam has high requirements on hardware;the cartographer has the best mapping effect in complex large environments and is suitable for complex environments in coal mine environments.Therefore,this algorithm meets the requirements of rescue robots.(2)The robot's path planning algorithm is analyzed.The algorithm is divided into global path planning and DWA-based local path planning algorithms.For global path planning,two methods based on Dijkstra and A * based are analyzed.After matlab simulation,A* is determined.The algorithm has more advantages in computing efficiency;the dynamic path method is adopted for the local path planning algorithm,and it is verified by matlab simulation that this algorithm can solve the local obstacle avoidance problem of the rescue robot,and finally verify the overall performance of the obstacle avoidance algorithm,The robot model is simulated in the open source robot simulator Autolabor Simulator.The actual results prove that the algorithm can complete coal mine rescue tasks in complex environments.(3)The circuit construction and program writing logic analysis of the rescue robot are completed,and its effectiveness is tested under the simulation of the gazebo physical engine.In order to facilitate the performance experiment of the robot and get the torque characteristics of its movement,a human-computer interaction interface based on LabVIEW is designed,which can dynamically detect the torque characteristics of the robot movement and facilitate various experiments of the robot The external interactive interface based on rviz is for the convenience of robot operation and state detection of its operation.During the operation of the robot,the path path,attitude change and drawing model can be obtained,and the underground personnel can be identified by combining with the latest open source YOLOv4,so that the operator can control the robot.