Research on Trajectory Optimization Method of Rocket Vertical Recovery Landing Phase

Aiming at the trajectory optimization problem of the rocket vertical recovery landing phase, this paper proposes a sequential convex optimization trajectory optimization method based on the pseudospectral method for planning the nominal trajectory. Firstly, considering the multi-constraint conditions of the rocket landing phase, a dynamic model for the vertical landing of the rocket is established. The optimization problem takes less fuel consumption as the performance index. The nominal trajectory planning is accomplished using the pseudospectral method. Then, considering the non-convexity of the state equation and multi-constraint conditions, the problem is transformed into a second-order cone programming (SOCP) problem by using lossless convexification, linearization and discretization methods. Finally, the optimization problem is solved by sequential iteration of the interior point method. The simulation experiments show that this sequential convex optimization method can obtain the optimal solution after a few iterations. This method for solving the optimal trajectory is efficient and accurate.