Development and validation of the source code for defining the optimal path of a mobile robot using genetic algorithm based TSP

Abstract

Room temperature cyclotron and, Superconducting cyclotron of VEC Centre Kolkata are the experimental facilities being offered for the acceleration of multiply charged heavy positive ions as well as light ions at different energy levels. There exists high intensity of mixed radiation and neutron flux in the cyclotron vault, pit area during the operation of cyclotron. In order to acquire various field information including near real-time video from those inaccessible areas with fixed obstacles navigation of a mobile robot through several intermediate nodes following a fixed path can be a probable solution. In the path planning strategy of a mobile robot to reach the target location may or may not be an optimal one. This paper has concentrated on the development and verification of the source code written in C to find out the optimal path for a mobile robot traversing through several nodes using the concept of Travelling Salesman Problem applying Genetic Algorithm. Superconducting Cyclotron Vault with fixed obstacles has been considered as the target space for defining the optimal path of the mobile robot. The validation of the code has been done by changing the number of initial population and increasing the number of generations. The concept of elitism has been incorporated to find out the optimal path (the most probable solution) among all the generations.