Intelligent navigation of humanoid robots using modified African vulture-inspired fuzzy control

Path planning and navigation control of humanoid robots in uneven and unstructured environments are challenging due to terrain uncertainty, stability requirements, and real-time decision-making constraints. This study presents an intelligent path-planning and navigation technique using a modified African vulture-assisted fuzzy logic controller (MAV-FLC). The proposed method combines the global search capability of a modified African vulture (MAV) optimization with the adaptive reasoning of fuzzy logic control (FLC) to generate smooth, collision-free, and energy-efficient paths. The vulture-based strategy improves exploration-exploitation balance, while the fuzzy controller effectively manages local terrain variations and motion uncertainties. The MATLAB and WEBOTs software are used to simulate path generation and route navigation for the humanoid robot within the designed environment. The performance of MAV-FLC is evaluated through extensive simulation and real-time experimentation on a humanoid robot navigating uneven terrain. Later, the outputs are compared with particle swarm optimization (PSO), ant colony optimization (ACO) and African vulture optimization algorithm (AVOA) using path distance and navigation time as performance metrics. Finally, experimental outputs show that MAV-FLC achieves over 10% improvement in both path optimality and time efficiency compared to PSO, ACO and AVOA. Moreover, the deviation between simulated and experimental outputs remains within 5%, confirming the robustness and practical applicability of the proposed approach. The outputs demonstrate the effectiveness of MAV-FLC for reliable humanoid robot navigation in real-world environments.