Feasibility-aware Imitation Learning from Observation with Multimodal Feedback

TL;DR

This paper introduces FABCO, a feasibility-aware imitation learning method that uses multimodal feedback and robot dynamics models to learn robust robot control policies from observation, overcoming demonstrator-robot differences.

Contribution

FABCO is the first approach to integrate feasibility estimation with multimodal feedback for imitation learning from observation, improving policy robustness and feasibility.

Findings

FABCO improves imitation learning performance by over 3.2 times.

Feasibility feedback reduces infeasible demonstrations and enhances policy stability.

Multimodal feedback from visual and haptic senses promotes feasible demonstrations.

Abstract

Imitation learning frameworks that learn robot control policies from demonstrators' motions via hand-mounted demonstration interfaces have attracted increasing attention. However, due to differences in physical characteristics between demonstrators and robots, this approach faces two limitations: i) the demonstration data do not include robot actions, and ii) the demonstrated motions may be infeasible for robots. These limitations make policy learning difficult. To address them, we propose Feasibility-Aware Behavior Cloning from Observation (FABCO). FABCO integrates behavior cloning from observation, which complements robot actions using robot dynamics models, with feasibility estimation. In feasibility estimation, the demonstrated motions are evaluated using a robot-dynamics model, learned from the robot's execution data, to assess reproducibility under the robot's dynamics. The estimated feasibility is used for multimodal feedback and feasibility-aware policy learning to improve the demonstrator's motions and learn robust policies. Multimodal feedback provides feasibility through the demonstrator's visual and haptic senses to promote feasible demonstrated motions. Feasibility-aware policy learning reduces the influence of demonstrated motions that are infeasible for robots, enabling the learning of policies that robots can execute stably. We conducted experiments with 15 participants on two tasks and confirmed that FABCO improves imitation learning performance by more than 3.2 times compared to the case without feasibility feedback.