Search papers, labs, and topics across Lattice.
This study investigates the reliance of Action Chunking with Transformers (ACT) on implicit cues from teleoperation for contact-rich manipulation tasks. By introducing an observation-centric ACT variant that predicts follower joint states instead of leader commands, the authors reveal that removing the teleoperation-induced discrepancy significantly impairs force-critical performance. However, incorporating simple joint-torque proxies derived from onboard motor current successfully recovers and enhances contact-aware behavior across multiple real-world manipulation tasks.
Removing implicit cues from teleoperation cripples force-awareness in manipulation tasks, but torque proxies can restore and even enhance performance.
Contact-rich manipulation requires policies to infer interaction state from signals that are often weakly observable through vision and kinematics alone. Action Chunking with Transformers (ACT) has shown strong performance in fine-grained manipulation, but many deployments collect demonstrations through leader-follower teleoperation, where tracking error between commanded leader motion and executed follower motion implicitly encodes contact, resistance, and constraint violation. This paper examines whether ACT's apparent force-awareness depends on this hidden interaction cue. We introduce an observation-centric ACT variant that predicts future follower joint states instead of leader commands, thereby removing the teleoperation-induced discrepancy signal while preserving the rest of the learning pipeline. We then evaluate whether simple joint-torque proxies, derived from onboard motor current or joint effort, can recover contact-aware behavior without external force/torque sensors. Across four real-world tasks spanning surface following, insertion, stiffness discrimination, and force-based stopping, removing the implicit cue leads to severe failures in force-critical phases. In contrast, torque-augmented policies recover robust contact behavior and improve the base ACT policy. These results demonstrate that, on real hardware, the implicit teleoperation cue is a recoverable source of force-awareness, where torque signals are available, a simple proxy matches, surpasses, or further enhances it.
