Matrix-based Genetic Algorithm for Drone Path Planning in Dynamic Collaborative Environments

Purpose The purpose of this research is to propose an efficient path planning algorithm for single and multiple drones in static and dynamic collaborative environments. With the increasing applications of drones in different fields, it is necessary to provide safe and efficient navigation for drones in unknown environments. In the present research, a matrix-based genetic algorithm (MGA) is proposed to provide safe and efficient path planning with reduced travel time. The proposed method is simulated and tested through MATLAB and real-time experiments to validate the reliability of the proposed method. Design/methodology/approach The research method involves the development and implementation of MGA for drone motion planning. The proposed approach applies the principles of genetic algorithms (GA) to determine optimal paths, where operations such as crossover and mutation are performed using matrix operations to improve convergence speed and reduce travel time. The algorithm represents possible routes as matrices between the start and goal points. Parent paths are selected using the tournament selection method, followed by two-point crossover and mutation to generate new paths, while elitism preserves the best solutions. The approach is validated through both simulation and real-time experiments. Findings Comparison of simulation and real-time experiments demonstrates that the proposed MGA technique is efficient, robust and accurate for both static and dynamic environments, with less than 5.8% deviation between results. The method successfully generates collision-free and optimal paths while avoiding static and dynamic obstacles. Furthermore, MGA outperforms Intelligent Ant Colony Optimization (IACO) and conventional Ant Colony Optimization (ACO) in achieving optimized path length. Originality/value This research contributes to drone path planning through the development and implementation of a novel MGA. The proposed approach demonstrates efficient and reliable performance in both simulation and real-world experiments, providing a strong foundation for future advancements in autonomous drone navigation and path planning in complex environments.