H Hajizadeh - Elec. Eng. Dept., Shahed University, Tehran, Iran
S Seyedtabaii - Elec. Eng. Dept., Shahed University, Tehran, Iran

To track a lane as fast as possible while the boundaries are not crossed an infinite horizon model predictive approach is devised to supply appropriate direction and speed commands. The robot is equipped with steering and driving motors which are PID controlled to partially attenuate the disturbances and unmodelled dynamics. The design core is a nonlinear optimization problem that is solved offline using intelligent methods. To reduce the computation load, the search space is narrowed by a careful type of coding. The design is implemented in MATLAB for a vehicle to follow an arbitrary route with right and left turns under various angles. The simulation results show that while the robot remains exactly within the lane the time passed remains remarkably low, very close to running at its maximum speed.

Path following, Intelligent optimization, Fast trajectory tracking, Time-optimal control