Improved Astar Algorithm for Path Planning of Marine Robot

Marine robot plays an important role in the marine research due to its good application prospects. Path planning provides the necessary information for marine robot to accomplish missions, and classic methods for path planning can be roughly divided into pre-planning and realtime planning, in wich Astar is an algorithm widely applied in path pre-planning of mobile robot. Classic Astar only generates a series of way-points for robots in formation of Descartes coordinate point which is based on a two-value grid map. The result has approximately optimal distance, however, the path does not conform with the motion constraint of robot. This paper proposes an improved algorithm in consideration of the orientation constraint of marine robot for Astar algorithm, and introduces related work about environment modeling. Path generated via this proposed method is more appropriate than classic Astar's in practical application.