Memory Based A-Star Algorithm for Path Planning of a Mobile Robot

"Machine learning is a branch of artificial intelligence which concerns the construction and study of systems that can learn from data. Mobile robots are the one which have locomotion in the environment and are not fixed to any particular physical location. Planning path for these mobile robots is the most critical part, which can be accomplished using any of the well-known algorithms like Dijkstra's algorithm, Genetic algorithm etc., among which -- A-Star algorithm has been chosen currently for path planning in this paper. A-Star algorithm uses a best-first search and finds the least-cost path from a given initial node to one or many goal nodes. AStar can be termed as a memory-less algorithm as it doesn’t remember the path traversed between the same set of nodes initially. In this paper an approach to store and reuse of the pre-calculated paths has been proposed. Using this hybrid memory based A-Star algorithm mobile robots can be easily made to traverse the given environment. Initially the robots explore the environment using normal A-Star and once it learns about its environment, it exploits the environment by recalling the paths traversed and accomplishes the tasks even faster. Hence this algorithm is named Memory Based A-Star Algorithm."