Robots learn to hold and press with adaptable softness
Impedance Cloning: Learning Equilibrium Point Parameters for Contact-Rich Manipulation
Robotics
Summary
Robots often struggle to gently press or hold objects when the shapes they touch change unexpectedly. The authors propose a method where robots learn not just exact movements but the underlying softness and resting position that humans implicitly use when handling objects. This approach helps robots automatically adjust their grip and force to stay stable despite changes in the surface shape. Tests show the method works better than traditional ones in various tasks like wiping surfaces and picking up different cups.
What this means in practice
- •For industrial robot programmers: Program robots handling delicate or variable-shaped objects to maintain consistent contact force without custom sensors.
- •For prosthetics designers: Incorporate learned equilibrium and stiffness parameters to improve the adaptability of robotic prosthetic hands during object interaction.
Authors
Hayato Takahashi, Ryoga Oishi, Yuki Kasuga, Toshiaki Tsuji
Abstract
Contact-rich manipulation requires robots to regulate force against surfaces whose geometry deviates unpredictably from training conditions. Trajectory-based imitation learning, which reproduces observable outputs, breaks down under such shifts. We propose Impedance Cloning, which instead imitates the biomechanical priors that generate motion -- the stiffness and equilibrium point -- and thereby passively absorbs contact uncertainty. Because these parameters encode intent rather than outcome, they generalize across surface geometries where trajectory reproduction does not. We extract them from bilateral teleoperation demonstrations via a particle filter without force/torque sensors and evaluate the framework on two CRANE-X7 manipulators. In a wiping task with joint-space actions, the trajectory-based baseline loses contact below -6 cm, whereas the proposed method maintains a consistent 4-5 N contact force above -6 cm, with a gradual decrease below; with Cartesian-space actions, its force-height slope over 0 to +8 cm is 0.13 +/- 0.03 N/cm, versus 0.34-0.83 N/cm for fixed-impedance baselines. In a pick-and-place task with 10 diverse cups (100 trials), the proposed method succeeds in 84 trials, outperforming the fixed-impedance baseline (74/100) and performing comparably to a variable impedance control baseline (82/100) with one demonstration instead of ten. In a grasping task, the representation reduces torque tracking error with both ILBiT and Mamba backbones, confirming its generality across architectures.