PC-sliding for vehicles path planning and control : Design and evaluation of robustness to parameters change and measurement uncertainty

A novel technique called PC-Sliding for path planning and control of non-holonomic vehicles is presented and its performances analysed in terms of robustness. The path following is based upon a polynomial curvature planning and a control strategy that replans iteratively to force the vehicle to correct for deviations while sliding over the desired path. Advantages of the proposed method are its logical simplicity, compatibility with respect to kinematics and partially to dynamics. Chained form transformations are not involved. Resulting trajectories are convenient to manipulate and execute in vehicle controllers while computed with a straightforward numerical procedure in real-time. The performances of the method that embody a planner, a controller and a sensor fusion strategy is verified by Monte Carlo method to assess its robustness to parameters changes and measurement uncertainties.