Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Experience-Based Feasibility-Aware Generative Adversarial Imitation from Observation (EF-GAIfO) to address the embodiment mismatch problem in learning robot behaviors from state-only human demonstrations. EF-GAIfO dynamically estimates demonstration feasibility from the robot's own experience, progressively expanding the feasible region as the policy improves. This adaptive approach allows for more effective incorporation of demonstrations and has been validated on both simulated and real-world robotic tasks.
Robots can now learn from human demonstrations even when their bodies are different, by dynamically judging what's possible based on their own experience.
With the increasing use of robot-free demonstration interfaces that provide state trajectories without action labels, imitation from observation has become a promising approach for learning robot behaviors from human demonstrations. However, due to differences in embodiment and dynamics between humans and robots, demonstrated human motions may not be feasible for the robot, potentially degrading policy performance. In this study, we propose Experience-Based Feasibility-Aware Generative Adversarial Imitation from Observation (EF-GAIfO), which estimates the feasibility of state-only demonstrations from the robot's own experience rather than relying on explicit dynamics models or large prior exploration datasets. A key feature of EF-GAIfO is that the notion of feasibility evolves with policy learning: as the policy improves and the robot experiences a broader range of state transitions, the feasible region is progressively expanded, allowing additional demonstrations to be incorporated into learning. This enables feasibility-aware imitation that adapts to the current stage of policy learning, rather than relying on a pre-designed feasibility criterion. We validate the effectiveness of EF-GAIfO on a locomotion task in simulation and on a real quadruped robot performing a object-reaching-and-grasping task.