Sensor fusion for ultrasonic array in mobile robotics: A comparative study of fuzzy, dempster and bayesian approaches

The key technology in the automatic ability of mobile robot is real-time obstacle avoidance and navigation. The main problem in research is that mobile-robot needs plenty of environmental information on moving and the information should be processed in high speed to satisfy real-time request. Three data fusion methods, based on fuzzy theoty, Dempster-Shafer evidence theory and Bayian deduction theory, are put into application on mobile robot's environmental recognition and navigation. A simplified simulation is accomplished for the comparison of the three algorithm navigation.