Reentry guidance with constrained impact for hypersonic weapon by novel particle swarm optimization

The novel reentry guidance law is proposed for hypersonic weapons to strike a stationary ground target at a constrained impact angle. This guidance law is based on proportional navigation guidance (PNG) whose gain is set as the function of range-to-go. To be resistant to disturbances and adaptive to various missions, the gain is refreshed in every guidance cycle. To lessen the control effort, the gain is optimized by a novel particle swarm optimization (NPSO) algorithm. NPSO needs a small number of particles for it can convert infeasible particles into feasible ones. Moreover, a mutation mechanism is introduced to accelerate convergence. In addition, there is only one single terminal constraint, that is, the impact angle in the optimization problem, because the terminal position is automatically identified using PNG. Compared with existing guidance laws, the proposed one needs neither complex derivation nor prior assumption. It also takes into consideration the constraints in lateral acceleration and look angle, which are often neglected in PNG-based laws. The adaptability under different scenarios, the robustness under disturbances and the potential for online application are demonstrated by simulation results. Numerical examples also show the superiority of NPSO when compared with the GPOPS.