Energy optimal real-time trajectory re-planning of an uninhabited surface vehicle in a dynamically changing environment

This chapter proposes a real-time energy optimal trajectory planning and tracking method for an uninhabited surface vehicle (USV) in dynamically changing environment. The problem is first discretized by Legendre pseudospectral method to transform the trajectory planning problem into a nonlinear programming problem with less discrete points and higher accuracy compared with traditional discretization methods. Then a novel strategy is developed to re-plan the trajectory in real-time during the entire maneuvering. The re-planning strategy uses the values obtained in the last re-planning as its initial guess values and upper and lower bound, which can reduce the computational complexity dramatically. Simulation results show that this novel approach could save a considerable amount of equivalent energy compared with an existing traditional method.