Research
A connected research program
My work centers on one question — how do legged robots earn reliability in the physical world? — approached today through modular robot design and control, and built on earlier work in locomotion learning, capability-aware planning, learning from human motion, and field-deployed robot systems.
Modular Legged Robots
When does a reconfigurable legged robot beat a purpose-built one — and what does that flexibility actually cost?
Designing reconfigurable leg modules and the learned controllers that let one robot family walk in many bodies.
Reconfigurable Design Context-Conditioned RL Cost-of-Transport Analysis
Learning-Based Locomotion Control
How can legged robots learn control policies that are fast, stable, and robust enough to run on real hardware?
Reinforcement learning and model-based control combined into layered controllers for multi-legged robots.
Reinforcement Learning Whole-Body Control Adversarial Training
Capability-Aware Planning
What can this robot actually traverse — and how should a planner use that knowledge?
Learning a robot's traverse capability and using it to plan long-range paths the robot can actually execute.
Capability Learning Path Planning Terrain Understanding
Learning from Human Motion
Can robots acquire task-oriented manipulation skills by imitating the style — not just the goal — of human motion?
Adversarial motion priors that transfer the style of human tool use to robot arms.
Adversarial Motion Priors Imitation Learning Sim-to-Real
Field Robot Systems
What does it take to move a robot from a simulation result to a machine that works outdoors, in public, on a schedule?
Complete robot systems — perception, control, communication, and deployment — built and field-tested on real robots, from hexapods to air–ground inspection teams.
System Integration ROS Field Deployment