Computing Systems for Autonomous Driving: State of the Art and Challenges

The recent proliferation of computing technologies (e.g., sensors, computer vision, machine learning, and hardware acceleration), the broad deployment communication mechanisms DSRC, C-V2X, 5G) have pushed horizon autonomous driving, which automates decision control vehicles by leveraging perception results based on multiple sensors. key to success these systems is making a reliable in real-time fashion. However, accidents fatalities caused early deployed arise from time time. real traffic environment too complicated for current driving understand handle. In this paper, we present state-of-the-art including seven performance metrics nine technologies, followed twelve challenges realize driving. We hope paper will gain attention both automotive communities inspire more research direction.