Answers are grounded in this paper and cite their sources.
Complex aerospace missions place increasing demands on guidance accuracy, resource efficiency, and autonomous decision-making. These requirements have extended the role of optimal control from offline trajectory design to onboard guidance under uncertainty and computational constraints. This paper reviews the principles, applications, and development of major optimal control methods for aerospace guidance. A common mathematical formulation is first introduced to connect vehicle dynamics, mission constraints, and performance objectives. Classical trajectory optimization methods are then reviewed, including indirect and direct methods, pseudospectral methods, and global search techniques. The discussion subsequently examines receding-horizon guidance, fast local solvers, and convexification methods for online trajectory optimization. Robust, stochastic, adaptive, and fault-aware approaches are considered for guidance under uncertain and changing conditions. Particular attention is given to learning-based control mappings, learning-enhanced numerical optimization, and reinforcement learning. Integrated guidance and control, distributed coordination, and multistage planning are also discussed in relation to system-level autonomy. Representative aerospace applications are used to compare the main advantages, limitations, and application conditions of these methods. Finally, current challenges and future research directions are outlined for coupled-dynamics planning, computation-constrained guidance, uncertainty and fault adaptation, and cooperative multistage missions.
No takes yet. Share an insight, caveat, or question.
Fang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: