Robot Motion Planning

Robot motion planning determines how a robot can move from a start state to a goal while respecting obstacles, its own geometry and dynamics, and task requirements. Classical approaches include graph search, sampling-based planners such as PRM and RRT, potential fields, and trajectory optimization or model predictive control; newer methods use learned policies and generative models to propose or refine motions. Planning is also combined with perception, feedback control, temporal-logic specifications, and task-level reasoning for settings such as manipulation, navigation, and multi-robot coordination. A central challenge is balancing speed and generalization in uncertain, dynamic environments with reliable execution and safety guarantees, especially when learned plans must transfer from simulation to physical robots.