Aiming at the problems of the low trajectory optimization efficiency and unsatisfactory optimization effect of glide guided projectile under complex and multi-constraint conditions such as high dynamic environment and strong state coupling, a trajectory optimization method of glide guided projectile based on particle swarm optimization-hp adaptive Radau pseudospectral method(PSO-hpRPM) algorithm is proposed. This method combines the strong global search ability of the PSO algorithm and the excellent local optimization ability of the hpRPM algorithm to solve the trajectory optimization problem of glide-guided projectiles under complex and multi-constraint conditions. Firstly, the constraints and objective functions of glide-guided projectile dynamics model and trajectory optimization problem are established. Then, hpRPM, PSO and PSO-hpRPM methods are used to optimize glide trajectory, and the three methods are compared. Finaly, to further validate the superiority of the proposed algorithm, we integrate the grey wolf optimization(GWO) algorithm, the whale optimization algorithm(WOA), with the hpRPM algorithm, respectively. We then conduct simulations to compare the performance of these three hybrid optimization algorithms: PSO-hpRPM, GWO-hpRPM, and WOA-hpRPM. The results show that the optimized PSO-hpRPM algorithm has a maximum glide range of 70170.75 m and an average iterative convergence of 6.6 steps under complex and multi-constraint conditions. The comprehensive performance is better than other algorithms and the optimization effect is good.