Map building and pose tracking are central tasks ter an autttaewotls mobile robot's navigation. The intention of this paper is to introduce our research in mobile robot indoor map building and pose tracking using laser range finder. The line feature extraction methods, which include improved angle histograms algorithm and weighted least square fitting, are proposed firstly. We then discuss the criterion of lines merging in map building. The pose tracking approach based on Extended Kalman Filter is also presented in this paper. Conclusions and experimental results based on a real mobile platform are also given in the final part of this paper. Index Terms -Map building, pose tracking, laser scanner, extend Kalman fdter, uncertainty representation.

No abstract

This paper proposes the motion control system in a hybrid architecture for middle-size soccer robot. A hybrid architecture combining hierarchical architecture and subsumption architecture is composed of four levels, such as the central control unit, the sensor information processing unit, the motion control system and the execution level. A local path pfanning algorithm fitting for autonomous soccer robot in dynamic environment is applied, and several actions and behaviors are properly designed for soccer robot motion control. The effectiveness of the proposed motion control system is validated by simulating and implementing it in a real soccer robot platform SmartROB2.