Ergodic Imitation for Adaptive Exploration around Demonstrations
Addresses the challenge of environmental or observational mismatch in robotic imitation learning by providing a principled exploration strategy grounded in demonstrations.
Proposes an adaptive ergodic imitation method that constructs a target distribution from demonstration geometry to generate trajectories interpolating between tracking and exploration, enabling robots to adapt to mismatches between training and deployment conditions.
In robotics, a common challenge in imitation learning is the mismatch between training and deployment conditions, caused, for example, by environmental changes or imperfect observation and control. When a robot follows a nominal trajectory under such mismatch, it may become stuck and fail to complete the task. This calls for adaptive online exploration strategies that remain grounded in demonstrations. To this end, we propose an adaptive ergodic imitation approach that constructs a target distribution from the geometry of the retrieved demonstrations and uses it to generate trajectories that adaptively interpolate between tracking and exploration. Our method extends ergodic control beyond its traditional role in area-coverage and search by incorporating demonstrations into a retrieval-based receding-horizon framework for adaptive imitation.