An optimization method for urban autonomous vehicle path planning from the perspective of internet of vehicles

With the rapid development of autonomous vehicle technology, optimizing path planning has become crucial for improving urban traffic efficiency and safety. However, traditional path planning methods for urban autonomous vehicles often suffer from high path deviation, elevated collision risk probability, and extended optimization time. To address these challenges, this study proposes an innovative optimization method for urban autonomous vehicle path planning from the perspective of the Internet of Vehicles. Firstly, a kinematic model for autonomous vehicles is constructed, which is then integrated with Internet of Vehicles technology to comprehensively perceive urban road environment information. Subsequently, a grid map is generated based on the automotive kinematic model and the acquired road environment data. A potential field ant colony algorithm is devised by enhancing pheromone update strategies and heuristic functions. Through iterative searches within the grid map, this algorithm effectively identifies the optimal path, thereby realizing the optimization of urban autonomous vehicle path planning. Experimental results demonstrate that the proposed method achieves a remarkable minimum path deviation of 0.04 and a minimal collision risk probability of 0.06. Moreover, the optimization time for urban autonomous vehicle path planning is confined between 0.25s and 0.72s. These outcomes signify that the proposed method not only enhances the accuracy and safety of autonomous vehicle path planning but also substantially reduces the computational burden and time consumption. This makes it highly promising for practical applications in urban autonomous driving scenarios, potentially revolutionizing the future of urban transportation.