Microcontroller-Based Multi-Objective Genetic Algorithm for Mobile Robots Path Planning

Abstract

Path planning is a fundamental task for a mobile robot. Path planning is an attempt to find the most effective route from the starting point of the mobile robot's position to a certain destination. Some of the commonly used path planning methods are the A * algorithm, Djikstras algorithm, D* algorithm, and others. Apart from these methods, there are metaheuristic methods such as multi-objective genetic algorithms. This method adopts the principles of natural selection and Darwin's theory of evolution. However, this method has the disadvantage that it requires a large enough memory and a fast processor. Mobile robots usually use microcontrollers with limited memory and low-end processors. This study proposes a micro controller-based multi-objective genetic algorithm for path planning on mobile robots. The multi-objective genetic algorithm will be applied to the Atmega32Sp microcontroller to find the shortest path length and the least number of turns.