Online Inertia Parameter Estimation for Unknown Objects Grasped by a Manipulator Towards Space Applications

TL;DR

This paper tackles online estimation of inertia parameters for unknown grasped objects in space, addressing the challenge posed by floating bases. It extends recursive least squares with log-determinant divergence to enforce physical consistency and incorporates momentum conservation for orbital scenarios. The results show that momentum-based regression yields robust, faster convergence than force-based methods, applicable to both ground and on-orbit servicing contexts. The approach advances capability for accurate dynamics-aware manipulation of unknown debris in space, with potential for real-world on-orbit implementations and adaptive control integration.

Abstract

Knowing the inertia parameters of a grasped object is crucial for dynamics-aware manipulation, especially in space robotics with free-floating bases. This work addresses the problem of estimating the inertia parameters of an unknown target object during manipulation. We apply and extend an existing online identification method by incorporating momentum conservation, enabling its use for the floating-base robots. The proposed method is validated through numerical simulations, and the estimated parameters are compared with ground-truth values. Results demonstrate accurate identification in the scenarios, highlighting the method's applicability to on-orbit servicing and other space missions.

Figures (6)

• Figure 1: A conceptual illustration of a base-floating orbital robot equipped with a robot arm capturing space debris.
• Figure 2: MuJoCo dynamics simulation to estimate unknown inertial parameters of the captured target by a base-floating orbital servicer.
• Figure 3: Parameter and force prediction errors (Fixed base, Target 10kg).
• Figure 4: Comparison of predicted and actual inertia elements (Fixed base, Target 10 kg and $\alpha_\eta=2.5\%$).
• Figure 5: Parameter and force-momentum prediction errors (Floating base, Target 50kg and $\alpha_\eta=2.5\%$).