Fusion of Evolutionary Algorithms and Multi-Neuron Heuristic Search for robotic path planning

The problem of path planning deserves a special mention in the field of robotics as it enables the intelligent systems used in autonomous robots to move the robot from one position to the other. Out of the various methods used for solving the problem of robot path planning, two of the common approaches include multi-neuron heuristic search (MNHS) algorithm and evolutionary algorithms (EA). The MNHS algorithm is an algorithm proposed earlier by the authors for solving uncertain search problems. The algorithm is slow but gives better optimal paths. On the other hand the EA gives results in finite time, but the optimality cannot be guaranteed. In this paper we propose to mix these two techniques to get the added benefits of both these algorithms. The MNHS improves the performance of the algorithm while the EA does the task of time optimization especially in case of complex graphs. The EA carries forward the task of selection of points in the robotic map. These points are checked for feasibility and then converted into a traversable graph. The same is used by MNHS to find the most optimal path from source to destination. In this way the algorithm finds out the best path without robotic collision.