Collision-free intelligent path planning in uncertain terrains for wheeled mobile robots based on dynamic learning Cuckoo Search Algorithm

Abstract

Intelligent path planning is crucial for mobile robots to reach their destination safely and efficiently. This work presents three variants of a Dynamic Learning Cuckoo Search (DLCS) algorithm to automatically generate obstacle-free optimal paths for mobile robots operating in uncertain terrains. Unlike conventional Cuckoo Search–based approaches, the proposed DLCS integrates a dynamic learning strategy and a multi-objective fitness function that improves obstacle avoidance, path continuity and overall path optimality. The proposed DLCS framework enhances path planning efficiency by reducing path length and travel time while maintaining collision-free navigation. Simulation results indicate that the DLCS-II variant achieves faster convergence and consistently shorter paths compared to existing state-of-the-art methods. To assess the reliability of the proposed approach, real-world experiments were conducted using a Khepera-IV mobile robot. The experimental results demonstrate good consistency with simulation outcomes, with path length deviations below 6% and travel time deviations below 2.5 s during robot motion.