Intelligent path planning for autonomous ground vehicles in dynamic environments utilizing adaptive Neuro-Fuzzy control

Autonomous Ground Vehicles (AGVs) are increasingly deployed across diverse industries, where enhancing their operational efficiency is critical, particularly in dynamic environments. This study proposes a hybrid control strategy that integrates an improved A* algorithm for path planning with a Proportional-Integral-Derivative (PID) controller adaptively tuned by an Adaptive Neuro-Fuzzy Inference System (ANFIS) for path correction. The enhanced A* algorithm, combined with the Dynamic Window Approach (DWA), significantly reduces computational overhead while improving pathfinding speed by incorporating the vehicle's kinematic and dynamic constraints. To address non-linearities in AGV movement, the ANFIS framework continuously fine-tunes PID parameters in real-time based on sensor feedback, improving the system's ability to correct path deviations in complex terrains. Experimental results demonstrate the efficacy of the proposed method. The enhanced A* algorithm achieves an average path search time of 2.52 s, significantly faster than the traditional A* algorithm's 5.56 s. It also reduces the average search grid size from 160 to 100, yielding a shorter path length of 27.44 m compared to 32.25 m, reflecting a more efficient path search process. Additionally, the ANFIS-PID control algorithm achieves a convergence time of 0.038 s, ensuring smooth path correction with robust stability under varying load conditions. Comparisons with state-of-the-art techniques, including Rapidly-Exploring Random Trees (RRT) and Probabilistic Roadmap Algorithm, highlight the enhanced A* algorithm's competitive performance, particularly in resource-constrained, real-time applications. The integration of ANFIS with PID control enhances AGV navigation by enabling adaptive, real-time path correction, improving performance in dynamic environments across agricultural, industrial, logistics, and autonomous transportation applications.