Navigation And Control System Integration Of Radiation Inspection Robot Based On ROS

With the continuous breakthroughs in artificial intelligence and robotics,increasingly intelligent mobile robots have come into our lives.However,in the fields of energy and medical care,such as nuclear power plants and hospital imaging rooms,there are security risks such as radioactive substances and pollutant residues,and the traditional manual investigation is inefficient and has personal safety hazards.Aiming at this problem,this paper studies and designs a kind of inspection robot which can be used to detect indoor radiation and pollutants in hospitals,and names it as radiation inspection robot.Based on the Robot Operating System(ROS)platform,this paper studies and designs the autonomous navigation,obstacle avoidance and radiation detection functions of the radiation inspection robot.The indoor space of the hospital is narrow and the ground is smooth.In order to reduce the movement error,the driving mechanism based on Mecanum wheel is designed in this paper.According to the detection requirements of indoor ceiling,wall and ground radiation,this paper designs a detection device composed of multi-stage electric cylinder,joint motor,joint connecting rod,radiation detector and laser displacement sensor,which improves the detection efficiency and expands the detection range,and is innovative.Secondly,this paper designs and builds a two-layer lower computer control system,and establishes the forward and inverse kinematics models of the radiation inspection robot.The function module of the upper computer software system is designed and built,and the functions of mechanical arm control and data acquisition in the detection process are realized.Considering that there are few characteristic points in the indoor environment of the hospital,the positioning accuracy is low and the range is small,which can’t meet the inspection requirements.This paper puts forward a positioning scheme that integrates wheel odometer,lidar,inertial measurement unit,encoder and other sensors,and makes research and experimental comparison between realtime positioning and map construction methods in ROS platform.In order to improve the effect of drawing,this paper improves the RBPF-SLAM algorithm.The Dijkstra algorithm and A*algorithm are compared in global path planning experiments,and A* algorithm is improved to make the path planning effect smoother.The function of autonomous obstacle avoidance is realized by using dynamic window method.Finally,in the indoor scene of the hospital,this paper carried out experimental tests on the map construction,path planning,autonomous obstacle avoidance function and radiation detection function of the radiation inspection robot in the process of autonomous navigation.The test results verify the feasibility of the improved navigation algorithm and show that the radiation inspection robot designed in this paper has good autonomous navigation function.The multi-sensor fusion positioning scheme proposed in this paper can realize autonomous navigation and positioning within 3cm and meet the accuracy requirements;The robot can complete a complete inspection process and meet the functional requirements.