Realization of reactive control for multi purpose mobile agents

Official URL: http://risc01.sabanciuniv.edu/record=b1064369 (Table of Contents)

The development of control method for an autonomous mobile robot that can be part of a multiagent system is the subject of this thesis. Mobile robots are built different purposes; for that reason, physical size, shape and mechanics, such as driving mechanism, of the robots would be very different. They are most of the time required to realize more than one goal at a time. Hence, as a minimum, required control must work in real-time and be able to respond to sudden changes that may occur in the environment. There is extensive research carried out, about autonomous mobile and stationaru robots, ranging from path planning and obstacle avoidance to multiple autonomous mobile robots. Proposed approach, for mobile robot control, is a layered structure and supports multi sensor perception. Each control layer uses only necessary sensor functions for its own process. Moreover, layers of control do not evaluate the signals to the necessary form asked by the layers. In this work, potential field method is implemented for obstacle avoidance. Unlike other solutions in the same framework, in this work assumed repulsive forces of the obstacles and attractive force of the goal point are treated separately. Then obstacles avoidance and goal tracking are fused in such a way that major drawbacks of the potential field method are overcome. Proposed control is tested on simulations where different scenarios are studied. The simulation results confirmed the high performance of the method. The proposed control is a potential alternative for mobile robots control operating in dynamic environments and as an agent in a multiagent system