Failure is Not an Option: Our Techniques at the DARPA Subterranean Challenge, Lessons Learned, and Next Steps
March 3, 2023
Chris Heckman of CU Boulder
When we in the robotics research community think of what we'd like autonomous agents to tackle in the future, we often target 'dull, dirty, and dangerous' tasks. However, despite a sustained boom in robotics research over the last decade, the number of places we've seen robotics in use for these tasks has been uninspiring. Successful commercialization of autonomous robots have required significant human scaffolding through teleoperation, and incredible amounts of capital, to achieve, and despite this are still limited by brittle systems and hand-engineered components. The reality seems to be that these tasks are not nearly as dull as they might seem on the surface, and instead require ingenuity for success some small but critical fraction of the time. In this talk, I focus on my recent investigation into where the limits of autonomy are for the highly sought-after application to subterranean emergency response operations. This application was motivated by the DARPA Subterranean Challenge, which just last year concluded with the CU Boulder team 'MARBLE' taking third place and winning a $500,000 prize. In this talk, I will give an overview into the genesis of our solution over three years of effort, especially with respect to mobility, autonomy, perception, and communications. I'll also discuss the implications for present-day robotic autonomy and where we go from here
About the speaker:
Chris Heckman is an Assistant Professor in the Department of Computer Science at the University of Colorado at Boulder and the Jacques I. Pankove Faculty Fellow in the College of Engineering & Applied Science. He earned his BS in Mechanical Engineering from UC Berkeley in 2008 and his PhD in Theoretical and Applied Mechanics from Cornell University in 2012, where he was an NSF Graduate Research Fellow. He had postdoctoral appointments at the Naval Research Laboratory in Washington, DC as an NRC Research Associate, and at CU Boulder as a Research Scientist, before joining the faculty there in 2016. From 2021-2023, Heckman also worked with Amazon as a Visiting Academic with their Scout program, focused on mobile robots for sidewalk delivery.
Heckman's research focuses on autonomy, perception, field robotics, machine learning and artificial intelligence. He directs the Autonomous Robotics and Perception Group, a dynamic and close-knit research team aiming to develop practical and explainable techniques in probabilistic artificial intelligence. His work in computer vision, machine learning and sensor fusion has applications to intelligence, defense, and environmental monitoring. His robotics work is used by both government and industry partners in the development of autonomous vehicles, agricultural platforms and other mobile robots including for medicine, search & rescue, and automation.
0:00 Introduction
3:30 Snapshots from the Finals
4:35 What I'm talking about today
4:53 Multi-agent Autonomy with Radar-Based Localization for Exploration
6:21 Tunnel Circuit Mobility
9:39 Urban Circuit
14:24 Final Event
19:05 Perception Sensor Head
19:57 The Human Supervisor's role
22:57 Anyone founding a startup?
26:43 A "solved problem"?
29:27 Fog and lidar-based mapping
31:03 Visual Challenges
31:57 What about radar?
34:12 Radar Sensors
34:48 Compact Radar
35:00 Radar Measurements
35:19 Doppler Measurements
37:32 Updated System Structure
38:26 Updated Doppler Model
39:29 Radar-Inertial Control
40:00 Sparse Mapping
40:28 Learned Outlier Rejection
41:26 Full Pipeline
42:15 The ColoRadar Dataset
42:48 ColoRadar Demo: Outdoor
42:59 ColoRadar Demo: Subterranean
