Distributed Vehicle Back Propagation Neural Network Cooperative Positioning Method With Fireworks Algorithm

This In the context of autonomous driving within vehicular networks, the accuracy of vehicle positioning is crucial for smooth operation. However, single navigation systems, such as satellite navigation and inertial navigation, cannot fully guarantee continuous high-precision positioning of vehicles. Therefore, achieving high-precision positioning through information collaboration between vehicles has become a primary approach. This article proposes a large-scale vehicle cooperative positioning method based on neural networks. This method addresses the characteristics of vehicles freely clustering and dispersing during travel by introducing principal component analysis (PCA) to process navigation information, reducing computational complexity. Additionally, it employs the Fireworks Neural Network method to rapidly integrate navigation information within the vehicular network, ensuring positioning accuracy and stability during vehicle operation. Compared with existing cooperative positioning methods, experimental results show that the proposed method has faster convergence speed and greater positioning stability.