Minimum-time Control of a Two-wheeled Differentially Driven Vehicle in the Presence of Slip

In this paper we propose a solution for the minimum-time control of a two-wheeled differentially driven mobile robot in the presence of slip between the wheels and the ground. Starting from the Lagrangian equations of the system a Newton-Eulerian model of the robot is obtained by adding longitudinal and lateral forces between the tyres and the ground, expressed by means of the Pacejka equation. The travelling time of the robot from an initial point to an end point has to be minimised subject to actuator constraints. A Chebyshev series approach is used for the optimisation of the trajectory. Because this optimisation procedure requires a significant computational effort, several trajectories are calculated off line and then they are sampled and used for the training of a neural network controller. This controller is then employed on line in order to make the mobile robot follow the desired path.