Abstract:
Ground robots are classified by locomotion into three major categories i.e. wheeled, tracked and Legged robots. Typically, wheeled robots are energy efficient and simple to control. Autonomous navigation requires integration of position and orientation sensing, path planning, obstacle avoidance and vehicle control. Autonomous navigation can be divided in to four sub tasks i.e. perceiving and modeling the environment, localizing the vehicle, plan and decide desired motion and execute the desired motion. This thesis mainly focuses on decide desired motion and its execution using Vector pursuit method. Vector pursuit is a geometric path tracking technique based on Screw theory developed by Sir Robert Ball. It is robust with large tracking error. This technique determines desired turning radius from vehicle current position and orientation to position and orientation at look ahead point. By using turning radius wheel velocities of vehicle were determined. The simulation results show that the vehicle satisfy its constraint and follow the waypoints accurately.
