MAE Seminar: TEAM-CoSTAR at DARPA Subterranean Challenge - Collaborative Autonomous Robots to Explore Extreme and High-risk Terrains
Robotics Research Technologist at NASA's Jet Propulsion Laboratory
California Institute of Technology
Abstract: The DARPA Subterranean or “SubT” Challenge is a robotic/software competition that seeks novel and revolutionary approaches to rapidly map, navigate and search extreme and underground environments. The challenge spans a period of three years with several competitions and a final event. TEAM-CoSTAR (JPL, Caltech, MIT collaboration) is one of the DARPA-funded teams participating in the systems track, which revolves around the development and implementation of physical systems that will be tasked with the traverse, mapping and search of different types of subterranean environments. In this presentation, Agha will discuss some of the research challenges and opportunities introduced by this competition and will present CoSTAR’s approaches and plans to address these challenges. CoSTAR (Collaborative SubTerranean Autonomous Resilient Robots) relies on a team of heterogeneous robots to solve this challenge. Specifically, he will discuss CoSTAR’s current developments and future plans on four fronts to better facilitate the possible grounds of future research collaboration. These include: 1) mobility, with a focus on next-generation multimodal (flying/rolling/bouncing) platforms; 2) localization, focused on developing resilient algorithms for multisensor fusion in perceptually degraded environments (e.g, dark, dust, smoke, etc.) with lightweight sensors; 3) multirobot situational awareness, focused on distributed algorithms for dense mapping and large-scale SLAM; and 4) distributed autonomy, focused on disruption-tolerant communication, active perception and risk-aware coordination of robot swarms.
Team website: https://subt.jpl.nasa.gov/
Bio: Ali Agha is a robotics research technologist with NASA’s Jet Propulsion Laboratory (JPL), Caltech. Previously, he was an autonomy research engineer with Qualcomm Research and a postdoctoral researcher with the Laboratory for Information and Decision Systems at Massachusetts Institute of Technology. He received his doctorate from Texas A&M University, and his research interests include robotic autonomy, mobility and perception; stochastic control systems and filtering theory. Agha manages several projects at JPL on autonomy, control and perception for robotic systems (rovers and aerial vehicles). He was selected as a NASA NIAC Fellow in 2018.