Hybrid GA and Monte Carlo Simulation for Safe Navigation in Dynamic Environments

This research presents a hybrid path planning framework that integrates Genetic Algorithms (GA) with Monte Carlo simulation to enable safe and efficient navigation of intelligent wheelchairs in environments populated with both static and dynamic obstacles. The proposed method employs GA to generate and optimize candidate trajectories, while Kalman filtering is utilized to predict obstacle motion and Monte Carlo sampling to evaluate collision probabilities under uncertainty. A safety-oriented mutation strategy is incorporated to preserve population diversity and accelerate convergence toward collision-free, geometrically efficient paths. Extensive simulations demonstrate that the algorithm achieves statistically negligible collision rates, rapid convergence within the first generations, and robust adaptability to stochastic variations in obstacle dynamics. Comparative analysis further highlights superior performance over conventional approaches in terms of trajectory clarity, computational efficiency, and responsiveness to environmental changes. Owing to its predictive modeling capability and resilience against motion noise, the framework is well-suited for real-world deployment in assistive mobility platforms and mobile robotic systems operating across diverse indoor and outdoor environments.

کلیدواژه ها:

Intelligent wheelchair path simulation ، Monte Carlo method ، Genetic algorithm ، Obstacle avoidance ، Autonomous systems