Control and trajectory generation of an Ackerman vehicle by dynamic linearization

This paper proposes a trajectory controller and a trajectory generation algorithm for an Ackerman vehicle. The definition of the system is based on a dynamic model which makes it possible to describe it in terms of the input signals normally reacheable for the driver: the torque of the engine related with the level of the gas pedal, and the movement of the steering wheel. This way, this control method can be implemented on any existing vehicle with minimum changes. For handling the nonlinearities present in the system, the mathematical description of the Ackerman vehicle is arranged in such a way that it fulfills the flatness conditions, making it possible to apply dynamic linearization. Another benefit of the proposed flat description is the close relationship between the flat outputs and the geometrical description of the path, so that the development of the control law for reaching any point in the surroundings of the vehicle is straightforward. In this connection, the trajectory generation algorithm automates a way of defining a trajectory out of a set of intermediate points. The robustness of the system against disturbances is assured by a closed loop control algorithm designed on the flat system.