Trajectory generation for a class of nonlinear systems with input and state constraints

Autonomous vehicles with nonlinear dynamics need to have a planned reference trajectory and a feedback controller to accomplish the task of traveling from a launch point to a goal point. Traditionally the planning stage has either been done from a purely geometrical point of view without regarding the dynamic constraints of the system, or by time-consuming numerical optimization including the dynamic constraints and input bounds. This paper presents a new way to generate a trajectory quickly when a nonlinear system is input-output linearizable while satisfying the dynamic constraints. An example of a trajectory based on a simplified nonlinear longitudinal helicopter model with minimum time criteria is presented.