The tunnel exit is easily blocked with a large traffic. One reason is that when the driver out of the tunnel, the added ramp resistance leads to the speed drop, but the driver is imperceptible. The present solutions are more concentrated in the transport sector, regulations are applied to all vehicles through the tunnel, easily distract the driver attention and disturb the operator accelerating away from the tunnel. In this paper, when the system judges the speed drop is due to driver's unconscious behavior through the road and external illumination changes ,then the system gives the driver accelerated reminder. Firstly, establish the vehicle driving dynamic model. With the road tunnel design standards for a reference, build the tunnel road model. Secondly, based on the dynamic model and the tunnel model, study the relationship between the throttle opening and the duration with the speed change and study the relationship between the engine braking with the speed and the ramp change. Then combining the tunnel lighting and the exit lighting effects on drivers, find the driver unconscious deceleration reasons when the vehicle out of the tunnel，with these reasons as the criterion to judge whether the slowdown is due to the driver's unconscious behavior. Finally, with the Wuhan Yangtze River Tunnel road design data as a reference to establish the road model, with its traffic flow data as a reference to establish the traffic flow model. Using the emulators for the system simulation, confirming whether the system is effective to improve the channel capacity. The study shows the application of the system can effectively reduce the traffic jams: between the start of morning rush and the end of evening peak, the application of the system allows the average traffic congestion time dropped 5% to 10%.To ease the traffic pressure and improve the capacity of the tunnel, the system is some positive effect.