With the rapid development of the new energy vehicle industry, the per capita car ownership has significantly increased. In-vehicle devices are gaining increasing attention for their role in enhancing driver safety and comfort. To assist drivers in real-time monitoring of their physical condition, timely adjustment of their status, reducing the risk of accidents, and improving vehicle energy efficiency, this paper employs Vehicle-mounted millimeter-wave radio equipment to measure the driver’s vital signs. The extracted vital signals are decomposed and reconstructed using a variational mode decomposition algorithm based on an improved snake optimization algorithm to obtain the current driver’s respiratory and heartbeat signals. This algorithm adaptively selects the optimal decomposition parameters according to different input signals. Experimental results demonstrate that the proposed algorithm can achieve real-time detection of the driver’s respiratory and heartbeat rates in new energy vehicles, showing significant advantages over other algorithms. This study provides valuable insights for the application of in-vehicle devices in new energy vehicles.