METHOD OF ROUTE PLANNING IN AUTONOMOUS LOGISTICS CYBERPHYSICAL SYSTEMS USING ARTIFICIAL INTELLIGENCE

Abstract

This paper presents an innovative approach based on the A* algorithm, which includes several key modifications to significantly improve its functionality and efficiency in autonomous vehicle navigation. The approach prioritizes safety and traffic compliance by integrating the A* algorithm with a mechanism that monitors the safe distance to obstacles. In addition, a trajectory smoothing component is introduced. This component improves the final path, creating a smoother and more comfortable trajectory. Detection of dangerous sections of the trajectory is another fundamental aspect of the proposed approach. This is achieved by applying k-means clustering, a powerful machine learning method. Thanks to the clustering of the trajectory segments, the system can recognize critical situations such as sharp turns and driving in the oncoming traffic lane. Identifying these segments allows proactive corrective actions to be taken to transform potentially dangerous scenarios into safer alternatives. One of the revolutionary elements of the approach is the introduction of Reinforcement Learning (RL) technology. The special RL model adapts to dynamic obstacles in real time, increasing the system's ability to respond quickly and efficiently to unexpected road situations. This adaptability is a key factor that makes autonomous logistics systems more secure and versatile. Thus, the method offers a comprehensive and intelligent solution for route planning in autonomous cyber-physical logistics systems. By combining the A* algorithm with state-ofthe- art obstacle avoidance, trajectory smoothing, detection of dangerous segments, and RL adaptability, the path to safer, more efficient, and more adaptive autonomous logistics is being paved. This approach has the potential to revolutionize transportation and delivery, offering a compelling vision of a future where autonomous vehicles will navigate the roads with the highest levels of safety, compliance and efficiency.