A multi-phase DMOC-based optimal trajectory generation methodology for a quadrotor

Abstract

This paper proposes a new multi-phase DMOC based trajectory optimization methodology to solve optimal control problems for mechanical systems. DMOC (Discrete Mechanics and Optimal Control) approach directly derives from discrete Lagrange-D'Alembert principle. The constraints for the optimization of a given cost functional are modeled as Euler-Lagrange equations. In addition to the basic requirements of DMOC, a Multi-phase Trajectory Optimization Strategy is proposed to satisfy some specific requirements and help to improve trajectory generation performance when the system should operate in a relatively complex or special environment; To show its advantages, the numerical simulations illustrate the proposed approach by generating the multi-phase optimal trajectory for a quadrotor, and comparison with another state-of-art direct Gauss Pseudo-spectrum Method (GPM) is presented. The experiment results show that our approach is more efficient to generate optimal trajectory for complex nonlinear problems than GPM, and with prospect of wide application in trajectory optimization.