A novel evasion guidance for hypersonic morphing vehicle via intelligent maneuver strategy

• DOI: 10.1016/j.cja.2024.02.024
• Publication Date: 2024-03-06T03:37:11Z
• Authors: Xun LI; Xiaogang WANG; Hongyu ZHOU; Yu LI
• Keywords: Evasion (ethics); Pursuit-evasion; Morphing

This paper presents a novel evasion guidance law for hypersonic morphing vehicles, focusing on determining the optimized wing's unfolded angle to promote maneuverability based an intelligent algorithm. First, pursuit-evasion problem is modeled as Markov decision process. And agent's action consists of maneuver overload and wings, which different from conventional designed fixed-shape vehicles. The reward function formulated ensure that miss distances satisfy prescribed bounds while minimizing energy consumption. Then, maximize expected cumulative reward, residual learning method proposed proximal policy optimization, integrates optimal linear cases baseline trains optimize performance nonlinear engagement with multiple pursuers. Therefore, offline training guarantees improvement constructed over ones. Ultimately, online implementation includes only analytical calculations. It maps confrontation state attack retaining high computational efficiency. Simulations show can utilize change extend maximum capability. it surpasses strategies by ensuring better efficacy higher