SSVisual: Intelligent Start-Stop System

Abstract

In order to reduce fuel consumption, many vehicles are equipped with idle start-stop systems. However, due to complex environment in the real world, vehicles with start-stop systems often experience short-term idling and frequent start-stops. It may greatly accelerate equipment deterioration hence reduce driving comfort. To resolve this problem, we propose SSVisual, an intelligent start-stop system by utilizing collected traffic information. Novel approaches are developed to effectively detect the states of traffic lights and traffic conditions based on image recognition techniques. Given the collected information, SSVisual can determine if it is necessary to shut down the engine at the current idle speed. Moreover, SSVisual can be used for both online and offline environments to visualize the performance of different strategies for research purpose.