Bi-HIL: Bilateral Control-Based Multimodal Hierarchical Imitation Learning via Subtask-Level Progress Rate and Keyframe Memory for Long-Horizon Contact-Rich Robotic Manipulation

TL;DR

Bi-HIL introduces a hierarchical imitation learning framework that combines subtask progress modeling and keyframe memory to enhance long-horizon, contact-rich robotic manipulation, showing improved robustness and coordination.

Contribution

The paper presents a novel bilateral control-based hierarchical imitation learning approach that explicitly models subtask progression and integrates keyframe memory for better long-horizon manipulation.

Findings

Bi-HIL outperforms flat policies in real-robot tasks.

Explicit subtask progression modeling improves robustness.

Keyframe memory enhances hierarchical coordination.

Abstract

Long-horizon contact-rich robotic manipulation remains challenging due to partial observability and unstable subtask transitions under contact uncertainty. While hierarchical architectures improve temporal reasoning and bilateral imitation learning enables force-aware control, existing approaches often rely on flat policies that struggle with long-horizon coordination. We propose Bi-HIL, a bilateral control-based multimodal hierarchical imitation learning framework for long-horizon manipulation. Bi-HIL stabilizes hierarchical coordination by integrating keyframe memory with subtask-level progress rate that models phase progression within the active subtask and conditions both high- and low-level policies. We evaluate Bi-HIL on unimanual and bimanual real-robot tasks, demonstrating consistent improvements over flat and ablated variants. The results highlight the importance of explicitly modeling subtask progression together with force-aware control for robust long-horizon manipulation. For additional material, please check: https://mertcookimg.github.io/bi-hil