Study On Wheeled Mobile Robot's Position And Navigation Technology Based On Multi-information Fusion

With the development of intelligence and information society, the research of mobile robot has attracted many domestic and foreign experts' attention. The mobile robots obtain their own information and external environmental information through sensors, then rely on that information to achieve autonomous mobile and completed a certain task. As a single sensor is hard to obtain all the information the system needed, multi-information fusion become the key to the robot to achieve their own decision. With the development of sensor technology, artificial intelligence, computer information technology, the research of mobile robots' positioning and navigation base on multi-sensor information fusion, will go further and get more achievement.In this paper, use pilotless automobile robot for the study. Study on wheeled mobile robot's information acquisition, obstacle avoidance behaviors and other issues. The main contents are as follows:(1) Established the kinematic model based on mobile robot. Selected sensors those appropriate wheeled robot positioning and navigation system to get information and establish the sensor system model.(2) Introduced the characteristics of multi-information fusion technology and frequently-used information fusion structure. Design a sensor information fusion model for wheeled mobile robot. Choose a hybrid structure as the multi-sensor information fusion structure of wheeled mobile robot's positioning and navigation system.(3) By comparing characteristics and advantages of frequently-used method, selected Kalman filter and artificial neural network to fusion multi-information for wheeled mobile robot positioning and navigation system.(4) Design an architectural prototype for wheeled mobile robot positioning and navigation based on multi-wheeled information fusion. And apply Kalman filter and artificial neural network information fusion method to pilotless automobile robot. Reflect the advantages of multi-information fusion.This paper presented a complex system of multi-information fusion model applied to mobile robot positioning and navigation system. Separation of information those described the system status and decision-making information of the system. Select different types of multi-information fusion method for different types of data integration. Finally, achieve this multi-information fusion model to pilotless automobile robot.