Full Stack engineering in Robot Open Autonomous Racing

Michael Wu

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2022-4
April 14, 2022

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-4.pdf

Autonomous Driving is a research topic that daunts many generations of engineers. With the advancement of AI and computing power, academia and industry has made significant progress in recent decade. However, high fragmentation, incompatibility across fields, and high initial cost still poses a tall barrier to entry prevents junior engineers such as student researcher or even grade school learners to get started, not even so to contribute to real world problems.

The purpose of this paper is to introduce my contributions in the development of the novel system ROAR that attempts to address the problem of fragmentation, incompatibility, and high cost in the autonomous driving industry. ROAR, name originally created for Robot Open Autonomous Racing, is a competition that encourages young students to get their first experience with engineering, allow researchers to advance their project without spending years on infrastructure, and a platform where extra components can be added with ease.

This paper first will go over the high level design principles and rationals in the Introduction section. Then it will go into details, touch upon a wide variety of technical topics, including 3D design, electrical engineering, vehicle dynamics, embedded systems, wireless communications, software architectures, advanced algorithms, and much more. This paper will outline the engineering effort in ROAR from design to actual implementation and the eventual verification and testing.


BibTeX citation:

@techreport{Wu:EECS-2022-4,
    Author = {Wu, Michael},
    Editor = {Yang, Allen and Sastry, S. Shankar},
    Title = {Full Stack engineering in Robot Open Autonomous Racing},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2022},
    Month = {Apr},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-4.html},
    Number = {UCB/EECS-2022-4},
    Abstract = {
Autonomous Driving is a research topic that daunts many generations of engineers. With the advancement of AI and computing power, academia and industry has made significant progress in recent decade. However, high fragmentation, incompatibility across fields, and high initial cost still poses a tall barrier to entry prevents junior engineers such as student researcher or even grade school learners to get started, not even so to contribute to real world problems.  


The purpose of this paper is to introduce my contributions in the development of the novel system ROAR that attempts to address the problem of fragmentation, incompatibility, and high cost in the autonomous driving industry. ROAR, name originally created for Robot Open Autonomous Racing, is a competition that encourages young students to get their first experience with engineering, allow researchers to advance their project without spending years on infrastructure, and a platform where extra components can be added with ease. 


This paper first will go over the high level design principles and rationals in the Introduction section. Then it will go into details, touch upon a wide variety of technical topics, including 3D design, electrical engineering, vehicle dynamics, embedded systems, wireless communications, software architectures, advanced algorithms, and much more. This paper will outline the engineering effort in ROAR from design to actual implementation and the eventual verification and testing.}
}

EndNote citation:

%0 Report
%A Wu, Michael
%E Yang, Allen
%E Sastry, S. Shankar
%T Full Stack engineering in Robot Open Autonomous Racing
%I EECS Department, University of California, Berkeley
%D 2022
%8 April 14
%@ UCB/EECS-2022-4
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-4.html
%F Wu:EECS-2022-4