Urbanization has been extensively increased in the last decade. In proportion, the number of vehicles throughout the world is increasing broadly. The detailed survey of available optimal path algorithms is done in this article, and to ease the overall traveling process, a dynamic algorithm is proposed. The proposed algorithm takes into consideration multiple objectives like dynamic traffic density, distance, history data, etc. and provides an optimal route solution. It is hinged on reinforcement learning and capable of deciding the optimal route on its own. A comparative analysis of the proposed algorithm is done with a genetic algorithm, particle swarm optimization algorithm, and the artificial neural networks algorithm. Through simulation results, it is proved that the proposed algorithm has better efficiency, decision making, and stability. It will ease the driver's headache and make the journey more comfortable with traffic less short distance routes that will minimize overall travel time making a positive impact on traffic jams, accidents, fuel consumption, and pollution.
Dynamic Algorithm Selection Using Reinforcement Learning
