Study Of Spatial Localization And Return Algorithm Of Mobile Robot Charging Pile

With the development of mobile robotics,a complete autonomous charging system is required to ensure that the robot can work efficiently.The Mecanum mobile robot is used as a platform for research and development in this paper.To make sure that the robot can accurately estimate the real time state of charge(SOC)of the battery and return to the charging stack for charging using the return algorithm,four major tasks were performed.The four main areas of work are:On the basis of the extended Kalman filter(EKF)algorithm for battery SOC estimation,the secondorder RC equivalent circuit model was first established and the parameters were identified.Estimated battery terminal voltage is compared with real value and estimated deviation is about 0.1V.The estimation of the battery SOC is based on the joint FFRLS and EKF algorithm,and it is verified that the error in the SOC estimation is kept within 2%compared to the real value.Based on graph-optimized SLAM using the Cartographer algorithm,a front-end local mapping,loop detection,and back-end global optimization framework were adopted,which effectively improved the mapping accuracy and back-end optimization efficiency.An improved version of the Cartographer algorithm was proposed based on traditional algorithms.Using ROS visualization tools Rviz and simulation environment Gazebo,simulations were conducted in both simple and complex environments.The simulation results show that the improved Cartographer algorithm can construct maps with clear contours and accurate obstacle proportions in environments with different obstacle densities.The return algorithm includes remote positioning navigation docking and proximity docking algorithms.A hybrid algorithm based on A*and Dynamic Windows Approach(DWA)is used to study the remote docking of the robot,which uses a hybrid path planning algorithm to move to the vicinity of the charging stack,and switches to a proximity docking algorithm to dock with the charging stack when the charger receives an intermittent 38KHz infrared signal from the charging stack to complete autonomous charging.Based on the hardware platform of Mecanum mobile robot and the ROS software platform,the Cartographer algorithm was used to conduct experiments on constructing the environment map while the location of the charging post was calibrated(charger),and experiments on autonomous charging docking were conducted on the environment map in the laboratory room.The average docking time at 3m and 3.75m from the starting point was 31s and 38s(0.5m/s).The robot was able to dock with the charging station and complete the autonomous charging,as shown by the experimental results.The terminal allows to monitor the charging status and to estimate the SOC of the battery in real time.The estimated full time is 2.66h at 59%SOC.This paper designs and builds an autonomous charging system that can accurately estimate the battery SOC of a mobile robot,enabling the mobile robot to return to charging autonomously at low battery levels,ensuring the robot’s endurance and improving its efficiency.