Improved Approach to Area Exploration in an Unknown Environment by Mobile Robot using Genetic Algorithm, Real time Reinforcement Learning and Co-operation among the Controllers

This paper explains the methodology applied to make a mobile robot explore an unknown environment accurately, with minimum energy dissipations and more speedily. Essentially it focuses on optimization capability of Genetic Algorithms and their convergence property, and how it can be applied in the domain of path planning. Optimization of path planning by mobile robots in environments known and unknown is a hot area of research. This paper is essentially an improvement over a previous paper on target tracking using Direct Competition in terms of lesser energy utilization, better approach of conducting simulations and interpretation of results. Rigorous generation wise experiments actually make the controllers improve a lot from their sub-minimal competent nature thereby overcoming the Bootstrap Problem. Another key point of the research is the observation of behavior in second set of experiment using the evolved weights after the first experiment, how it affects the fitness and how far proves to be successful in achieving the objective.