CAAD: Computer Architecture for Autonomous Driving

TL;DR

This paper analyzes the computing requirements of autonomous driving, reviews existing platforms, and discusses design approaches to achieve high performance with low power, thermal, and cost constraints.

Contribution

It provides a comprehensive overview of autonomous driving computing tasks and explores design strategies for efficient platforms.

Findings

Existing platforms vary in performance and efficiency

Design approaches can balance high performance with low power and thermal dissipation

Guidelines for future autonomous driving computing platform development

Abstract

We describe the computing tasks involved in autonomous driving, examine existing autonomous driving computing platform implementations. To enable autonomous driving, the computing stack needs to simultaneously provide high performance, low power consumption, and low thermal dissipation, at low cost. We discuss possible approaches to design computing platforms that will meet these needs.