Hybrid Potential Field of Navigation in Games

Path-finding plays an important role in the computer game industry. Generally, the A* algorithm and its various extensions are used in traditional path-finding methods, but they can only find a path for a single agent once a time. Recently, some potential field path planning methods have been proposed. These methods can be accomplished by multiple agents simultaneously with little additional computation. This paper reviews three different Potential Field path-planning methods for games. The main method is based on Potential fields, and we also introduce two Hybrid Potential Fields of Navigation in games and a method of implementing the algorithm with a hyper-target that people set. The first paper shows the performance of a potential field algorithm and tries to solve some problems in the algorithm. A hybrid navigation algorithm is introduced in the second paper, utilizing A* and potential fields, to handle the problems in RTS games better. The Third paper presents a real-time Voronoi-like path planning method using a combination of flow field and A* algorithm to navigate dynamic environments with obstacles, addressing issues of local minima and providing a flexible, efficient solution for collision-free navigation.