Selected as one of four 2025–2026 MHI ECE Ph.D. Scholars at USC
Paper 'Adaptive Locomotion on Mud through Proprioceptive Sensing of Substrate Properties' accepted to Robotics: Science and Systems (RSS) 2025, with live robot demo
Paper 'Characterizing Robot-Ground Interactions for Autonomous Lunar Construction' accepted to ISER'25
Paper 'Modelling Experts' Sampling Strategy to Balance Multiple Objectives During Scientific Explorations' accepted to HRI '24 with Honorable Mention
Paper on flipper-mud interactions accepted to IEEE Robotics and Automation Letters (RA-L)
Demonstrated robot’s capability to measure regolith geotechnical properties in real time at LPSC 2025
Presented research at ICRA 2024 (Yokohama), AAAI 2025, APS March Meeting, and others
Papers accepted to WMAC @ AAAI 2025 and LM4Plan @ AAAI 2025
MeetScript accepted to CSCW 2023 (26th ACM Conference on Computer-Supported Cooperative Work and Social Computing)
Research Experience
Worked on NASA’s LASSIE project: developing legged autonomous robots for surface science in analogue environments
Contributed to the TRUSSES project on multi-heterogeneous robot coordination and planning
Researched Scout-Rover cooperation for online terrain strength mapping and traversal risk estimation
Conducted field tests at White Sands National Park (Oct 2025) and Mt. Hood, Oregon (Aug 2023)
Collaborated with Dr. Gaurav S. Sukhatme on embodied proactive agents for human-robot teamwork
Background
Ph.D. Candidate in RoboLAND at the University of Southern California (USC)
Develops direct-drive legged robots that use legs and bodies as active force sensors for real-time terrain estimation and adaptive navigation without vision or LiDAR
Aims to advance legged robots as environment-aware co-scientists for planetary and terrestrial scientific exploration
Contributes to NASA projects LASSIE (Legged geoscience robotic assistant) and TRUSSES (multi-heterogeneous robot planning)
Collaborated with Dr. Gaurav S. Sukhatme on embodied proactive agents for human-robot teaming, developing foundation model frameworks that integrate fast–slow reasoning for real-time proactive assistance