Abstract
<title>Abstract</title> <p>Robust navigation in real-time dynamic environments remains a critical challenge for autonomous mobile robots, as traditional planners often exhibit slow convergence and limited adaptability to environmental changes. This paper proposes a hybrid intelligent navigation framework that combines an enhanced global planner with an adaptive local motion controller. The global planner, AP-RRT, extends the RRT algorithm through potential-guided sampling, adaptive stepsize adjustment, and optimised path refinement, improving exploration efficiency and path quality. For local navigation, two adaptive weighted Dynamic Window Approach variants, AW-DWA and DA-DWA, are introduced to ensure safe and responsive motion in dynamic environments. The integration of global and local planning enables continuous path tracking while adapting to environmental uncertainties. The proposed framework is evaluated through simulation and real-world experiments against state-of-the-art methods. Results show that AP-RRT reduces path length by 19% and improves path smoothness by 67% comparedto standard RRT. Additionally, AW-DWA achieves a 20% higher success rate,35% lower path deviation, and 74% improvement in path smoothness over the Improved DWA. These results demonstrate that the proposed framework provides an effective and robust solution for real-time autonomous navigation in complex dynamic environments.</p>