Research On Outdoor Positioning And Navigation Technology Of Wheeled Mobile Robot Based On Multi-sensor Information Fusion

With the development of social productivity,industrial fields such as automated container terminal require transport robots with a higher level of intelligence.Therefore,the problem of outdoor positioning and navigation of wheeled mobile robots has attracted widespread attention.At present,the outdoor positioning and navigation of robots mainly rely on the multi-sensor information fusion technology based on the Global Positioning System(GPS).However,trees,tall buildings,dust and sand often affect the navigation process,causing problems like GPS signal loss and changing road conditions,which seriously reduces the accuracy and stability of navigation.Research on the outdoor positioning and navigation technology of wheeled mobile robots is of great significance for the early realization of unmanned logistics transportation and the improvement of the level of intelligence in the industrial field.In this paper,we study the multi-sensor information fusion method to solve the outdoor positioning problem of wheeled mobile robots.An event-triggered Kalman Filtering(KF)algorithm is proposed based on sensors such as GPS,odometer and inertial measurement unit(IMU).And a GPS estimator is designed,which can provide the KF estimator with continuously available position information,making the filtering process unaffected.The state change of the robot is used as the trigger event of KF,the KF estimator responds and gives a new state estimation only when the robot’s state deviates more than a threshold.Experimental results showed that,compared with traditional KF,the event-triggered KF could deal with the GPS signals interruption to ensure the accuracy of outdoor navigation and reduced the time complexity of the algorithm.Based on the research on the navigation execution of wheeled mobile robots,an adaptive robot control system based on slip rate control is proposed.In which we propose a method for identifying the road adhesion coefficient and calculating the optimal driving wheel slip rate.The designed PID controller takes the slip rate of the driving wheel as the controlled object,and adjusts it to the optimal slip rate by changing the output torque of the driving wheel motor,to provide the optimal traction force for the robot.The experimental results showed that when the robot travelled on different roads,this control system could always control the slip rate of the driving wheel stably,reduce the slip of the driving wheel,and improve the accuracy and stability of navigation execution.Finally,the filtering algorithm and the robot control system are combined to conduct physical experiments.The Komodo robot is used to perform point-to-point navigation and outdoor tracking tests in real scenes.The experimental results showed that the outdoor positioning and navigation scheme proposed in this paper could greatly improve the positioning and navigation accuracy of the robot in both pose estimation and navigation execution.At the same time,it could solve the problems of GPS signal loss and changing road conditions,increase the stability of the robot’s outdoor positioning and navigation.