Abstract: Mars Ascent Vehicle(MAV)design is closely related to trajectory design. However,the existing method mostly decoupled the staging optimization and trajectory optimization,resulting in computational inefficiency and poor robustness. A coupled staging-trajectory multiphase optimization algorithm is proposed in this paper for a two-stage MAV design. Considering given vehicle,mission,path and control constraints,the objective of the optimization algorithm is to minimize the Gross Lift-off Mass(GLOM)of the two-stage MAV,and the algorithm is based on Gauss pseudospectral method. By using this algorithm,an optimal staging design is obtained,and an optimal trajectory is designed to minimize propellant consumption simultaneously. And the algorithm solves the problem that trajectory optimization algorithm couldn't converge due to the unreasonable MAV staging design. In addition,numerical simulations show that the coupled staging-trajectory multiphase optimization algorithm have good robustness and strong convergence.