[Paper Review] F1/10: An Open-Source Autonomous Cyber-Physical Platform

Matthew O’Kelly, Varundev Sukhil+9 more|arXiv (Cornell University)|Jan 24, 2019

Autonomous Vehicle Technology and Safety44 references85 citations

TL;DR

The paper introduces F1/10, an open-source, affordable 1/10-scale autonomous vehicle testbed with full sensing, perception, planning, control, and networking stacks, designed for research and education. It demonstrates how the platform enables real-world experimentation and collaboration in autonomous systems.

ABSTRACT

In 2005 DARPA labeled the realization of viable autonomous vehicles (AVs) a grand challenge; a short time later the idea became a moonshot that could change the automotive industry. Today, the question of safety stands between reality and solved. Given the right platform the CPS community is poised to offer unique insights. However, testing the limits of safety and performance on real vehicles is costly and hazardous. The use of such vehicles is also outside the reach of most researchers and students. In this paper, we present F1/10: an open-source, affordable, and high-performance 1/10 scale autonomous vehicle testbed. The F1/10 testbed carries a full suite of sensors, perception, planning, control, and networking software stacks that are similar to full scale solutions. We demonstrate key examples of the research enabled by the F1/10 testbed, and how the platform can be used to augment research and education in autonomous systems, making autonomy more accessible.

Motivation & Objective

• Provide an open, affordable, and high-performance autonomous cyber-physical platform for research and education.
• Offer modular hardware and software stacks that emulate full-scale autonomous vehicle capabilities.
• Showcase representative research uses including perception, planning, control, V2V, and simulation on a real testbed.
• Provide a pathway from 1/10-scale research to full-scale autonomous vehicles and highlight community adoption.

Proposed method

• Open-source mechanical design and kits for assembling a 1/10-scale autonomous racing car.
• A ROS-based software stack covering perception, planning, control, and coordination.
• A configurable sensor suite including LiDAR, monocular/stereo cameras, and IMU with Jetson TX2 computation.
• A Gazebo-based simulation environment and online resources for education and experimentation.
• Model predictive control and cubic-spline trajectory generation as local planners with an auxiliary learned MPC approximator.
• V2V communication nodes and testing scenarios such as roundabouts and high-speed merges.

Experimental results

Research questions

• RQ1 How can an open-source, affordable 1/10-scale platform enable realistic AV research and education in CPS?
• RQ2 What are the core hardware and software stack components needed to emulate full-scale autonomous vehicle pipelines on a small-scale platform?
• RQ3 How can perception, planning, control, and V2V coordination be prototyped and tested on a common research instrument under real-world constraints?
• RQ4 What is the role of simulation and real-time systems research in validating autonomous driving algorithms on F1/10?

Key findings

• F1/10 provides open-source hardware designs, software stacks, a simulator, and educational programs for autonomous systems research.
• The platform uses ROS with a modular chassis and Jetson TX2, enabling realistic dynamics at 1/10 scale and speeds up to 40 mph.
• Over 20 institutions are utilizing F1/10, fostering a growing community and collaborative research.
• The stack supports diverse research areas including obstacle avoidance, end-to-end driving, MPC, and V2V coordination.
• The distribution includes data sets with over 40,000 labeled images for end-to-end driving experiments.
• A bridge from 1/10-scale research to full-scale autonomous vehicles is demonstrated through case studies and international adoption.