Enhancing the carbon reduction of alternative energy vehicles through intelligent road environment information recognition using YOLOv8

Against the backdrop of China's implementation of the dual carbon plan and the rapid global development of new energy vehicles, there has been substantial industry and policy support for the advancement of alternative energy vehicles powered by hydrogen fuel cells, methanol fuel, and other substitutes. The level of intelligence in these vehicles is now on par with that of traditional new energy vehicles. In order to achieve the dual carbon goals, it is essential not only to develop various new alternative energy sources but also to prioritize optimized energy usage. In particular, the application of autonomous driving technology allows for precise decision-making and autonomous operation based on road environment information perception. This significantly enhances driving efficiency while reducing unnecessary energy consumption. However, existing road sign recognition algorithms exhibit poor robustness in complex environments. To address this issue, a YOLOv8-based model was designed to simulate obstacle occlusion using mosaic data and add Gaussian noise to reduce overall image exposure brightness—thus simulating scenarios with impaired visual conditions under real-world driving environments. The results indicate that this model has achieved a relatively effective balance by maintaining high precision while ensuring good real-time performance. Test results demonstrate that the improved YOLOv8JH network exhibits lower loss and an 87.1% mean average precision (mAP) on the dataset—a 25.1% improvement over previous iterations. With a high frame rate of 104.16 FPS and an average detection speed per image at 9.6ms, it meets real-time detection requirements. This method strikes a balance between detection speed and accuracy suitable for meeting most road traffic sign detection requirements in diverse driving conditions; however, further optimization is needed for target detection methods under extreme weather conditions.