Robotic path planning using multi neuron heuristic search

Robotics is a highly multi-disciplinary field which attracts the attention of many researchers from diverse fields. One of the major studied problems in Robotics is the problem of Robotic Path Planning. The problem deals with finding of the path that can be used by the robot for navigation purpose without any collision. The output of this algorithm is then implemented for physically moving the real robot on the desired path. In this paper we have used Multi-Neuron Heuristic Search (MNHS) which is an advanced form of A* algorithm. The MNHS was earlier proposed by the authors for special cases where the heuristics changes sharply and it was shown to be a powerful algorithm in the same context. In this paper we apply the MNHS for the Robot Path Planning. The motivation is to make the problem robust against the uncertainties that might arise like the sudden discovery that the path being followed does not lead to the goal. The MNHS has better capabilities to solve maze-like maps where the uncertainty is extremely high. Another such area is when the robot enters into a highly chaotic area. Here it might be better to go with a path that is less chaotic or has lesser number of obstacles. We tested the algorithm for numerous test cases. In all the cases, the MNHS was able to solve the problem of path planning well.