With the widespread application of intelligent driving technologies, more and more practical problems need to be solved urgently. Focusing on the integration of key intelligent driving technologies: V2X(Vehicle to Everything) and vision perception, this paper proposes a system-level solution to coping with undulating road conditions for intelligent vehicle control. By introducing taillight recognition, vehicle detection performance for puddles on rainy nights is enhanced. At the same time, this solution improves the accuracy, environmental adaptability, and robustness for road roughness detection. On extremely undulating road surfaces, it can avoid or mitigate damages caused by large pits or bumps to the vehicle, improving driving comfortability and vehicle control safety. The fusion algorithm used in this paper, based on the application of V2X equipment, fully excavates the ability of visual perception, can deal with extreme working conditions, greatly improving the driving safety. This will make contribution to the practical application of intelligent driving technologies.