Bio-inspired hybrid path planning for efficient and smooth robotic navigation

Abstract Robotic navigation in complex high-dimensional environments faces great challenges, especially in achieving efficient exploration, collision-free trajectory planning, and robust performance under dynamic conditions. Traditional optimization-based methods often suffer from limited adaptability, premature convergence, or insufficient obstacle handling. To address these challenges, we propose a novel hybrid path planning framework that integrates the Ant Colony Optimizer (ACO), the Whale Optimizer (WOA), the Artificial Potential Field (APF), and a random jump mechanism. This hybrid integration is characterized by the complementary global exploration, local optimization, and smooth trajectory generation, and is supported by a relativistic potential model and quantum-inspired mutation. Experimental results in various unstructured scenarios show that our method achieves significant performance improvements over six commonly used benchmarks including genetic algorithms (GA), genetically engineered word optimization (GWO), and quantum-inspired mutation (IRRT), with a 28.3% reduction in collision frequency and a 19.6% improvement in average path smoothness after preliminary simple quantification. In addition, our algorithm achieves the highest average navigation speed while maintaining the lowest variance, indicating that planning is consistent and efficient. These advantages make this method particularly suitable for practical robotics applications involving human passengers and dynamic obstacle environments.