Reinforcement-based black-winged kite algorithm for UAV path planning based on chaotic mapping and water wave dynamic evolution mechanism

Abstract To address the issues of slow convergence and susceptibility to local optima prevalent in mainstream bio-inspired algorithms, this paper proposes a enhanced black-winged kite algorithm (WCTBKA) for unmanned aerial vehicle (UAV) path planning and obstacle avoidance. Specifically targeting the original BKA’s slow convergence and susceptibility to local optima in UAV path planning and obstacle avoidance, WCTBKA incorporates three key enhancements: tent chaotic mapping, a chaotic dynamic weight factor, and a water wave dynamic evolution (WWDE) factor. Tent chaotic mapping replaces the standard random initialization to augment initial population diversity. The chaotic dynamic weight factor processes individuals during both the attacking and migrating phases, thereby accelerating convergence. Concurrently, WWDE factor adaptively balances global exploration and local exploitation. Simulation results demonstrate that WCTBKA achieves superior comprehensive performance in convergence speed and local optima avoidance compared to current state-of-the-art methods. Quantitatively, the proposed algorithm averagely improves the optimal fitness and convergence performance by 30% and 31%, respectively.