Fuzzy Reinforcement Rectilinear Trajectory Learning