Developed Time-Optimal Model Predictive Static Programming Method with Fish Swarm Optimization for Near-Space Vehicle

To establish the optimal reference trajectory for a near-space vehicle under free terminal time, a timeoptimal model predictive static programming method is proposed with adaptive fish swarm optimization. First, the model predictive static programming method is developed by incorporating neighboring terms and trust region, enabling rapid generation of precise optimal solutions. Next, an adaptive fish swarm optimization technique is employed to identify a sub-optimal solution, while a momentum gradient descent method with learning rate decay ensures the convergence to the global optimal solution. To validate the feasibility and accuracy of the proposed method, a near-space vehicle example is analyzed and simulated during its glide phase. The simulation results demonstrate that the proposed method aligns with theoretical derivations and outperforms existing methods in terms of convergence speed and accuracy. Therefore, the proposed method offers significant practical value for solving the fast trajectory optimization problem in near-space vehicle applications.