Genetic Algorithm for Mobile Robot Global Path Planning Application

Path planning is a core technology in mobile robot autonomous navigation systems and the main objective is to find the most efficient path, reducing both time and energy consumption during robot movement between different locations. To achieve this optimal path, various approaches have been studied and implemented, such as environmental modelling, path search algorithms, and optimization criteria. Among the algorithms considered, genetic algorithms (GA) were used to construct a valid, collision-free, and feasible path while considering specific parameters and criteria. The proposed GA was simulated via MATLAB and thoroughly analyzed to identify potential improvements or modifications. By selecting appropriate population sizes, fitness functions, selection, crossover, and mutation operators, the GA’s performance was optimized. Evaluations were made regarding the robot’s trajectory and time taken to reach the goal point in various constructed environments. During the simulation, varying the number of iterations while keeping the initial population constant revealed that the algorithm’s performance, in terms of path length and computation time, was influenced. The simulations consistently demonstrated that the optimized GA generated an optimal path with the lowest cost, highest effectiveness, minimal energy consumption, and ensured safety for the robot. This project led to the development and implementation of a robust path planning system, equipping mobile robots with the best and most feasible paths from their initial points to their goal destinations.