Outdoor Mobile Robot Localization Based On Information Fusion Theory And Technology Research

The localization of outdoor mobile robot is one of the hotspot of intelligent robot. With the development of multisensor data fusion, there is a trend to improve the precision of robot localization with the application of multisensor data fusion technology. The popurse of this paper is to analyse Kalman Filter and H_∞ Filter and apply them to the localization of outdoor mobile robot. Based on the outdoor environment and the localization system of THMR-V, a special filter is proposed in the paper.In this paper, we discuss the conception, characteristic, sorts of the localization of outdoor mobile robot. Common technology and sensors are introduced in detail. Then we discuss the conception, basic principle, technologies and application fields of multisensor data fusion. Based on the analysis of localization technology and multisensor data fusion, we apply multisensor data fusion to localization system. Kalman Filter Theory, H_∞ Filter Theory and the sensors of THMR-V localization system are analysed. Based on this, a special Kalman Filter is designed, implemented using Matlab and simulated using simulated data and practical data. The results under different motion models are compared and the localization Estimated Standard Deviation is computed. The results show that the Kalman Filter improves the precision of the robot localization. But it is necessary for the Kalman Filter to know the characteristic of noise. In the outdoor environment, it is difficult to get the characteristic of noise. So a H_∞ Filter is proposed to solve this problem. The Kalman Filter and the H_∞ Filter are compared. The results show that the application of the H_∞ Filter improves the adaptability of the localization system to the outdoor environment. This way offers an effective method for the development of localization system.