aiqbench analysis
The paper proposes an analytic robot self-model that fits Gaussian appearance from images and identifies metric screw-axis kinematics from sparse end-effector poses, without neural deformation networks or URDF/CAD templates. Its most useful point for AI readers is that a good image fit does not guarantee accurate kinematics: two models differing only in joint axes had nearly identical PSNR but 13.7-fold different end-effector errors. Results are reported on simulated arms and a quadruped plus one real six-axis arm, so real-world generalization remains to be shown.
Analytic Robot Self-Model: Learning Screw-Axis Kinematics for Rendering and Model-Based Control
arXiv:2610.15558v1 Announce Type: new Abstract: Visual self-modelling lets robots learn body models from observations, but such models usually encode kinematics implicitly rather than exposing metric joint parameters for computation and control. Moreover, a good fit to images does not by itself establish accurate kinematics. Here we present an explicit analytic robot self-model that requires neither deformation networks nor URDF/CAD templates. Multi-view images and joint readings establish Gaussian appearance and provide motion evidence for rigid link assignments, whereas sparse end-effector poses independently identify metric screw axes. The resulting closed-form forward kinematics and analytic Jacobian can be evaluated beyond the fitting ranges. We validate the model on a simulated seven-axis arm, a simulated twelve-joint quadruped and a real six-axis arm. In simulation, it achieves millimetre-level reaching and pick-and-place, as well as closed-loop quadruped walking. On the real arm, it renders held-out configurations, predicts end-effector positions with a median error of 6.4 mm, compared with 11.7 mm for a calibrated nominal model, and supports reaching with external marker feedback. A controlled comparison shows that image fit alone is not enough: two Panda models differing only in their joint axes are 0.54 dB apart in PSNR, yet their end-effector prediction errors differ 13.7-fold.
Abstract:Visual self-modelling lets robots learn body models from observations, but such models usually encode kinematics implicitly rather than exposing metric joint parameters for computation and control. Moreover, a good fit to images does not by itself establish accurate kinematics. Here we present an explicit analytic robot self-model that requires neither deformation networks nor URDF/CAD templates. Multi-view images and joint readings establish Gaussian appearance and provide motion evidence for rigid link assignments, whereas sparse end-effector poses independently identify metric screw axes. The resulting closed-form forward kinematics and analytic Jacobian can be evaluated beyond the fitting ranges. We validate the model on a simulated seven-axis arm, a simulated twelve-joint quadruped and a real six-axis arm. In simulation, it achieves millimetre-level reaching and pick-and-place, as well as closed-loop quadruped walking. On the real arm, it renders held-out configurations, predicts end-effector positions with a median error of 6.4 mm, compared with 11.7 mm for a calibrated nominal model, and supports reaching with external marker feedback. A controlled comparison shows that image fit alone is not enough: two Panda models differing only in their joint axes are 0.54 dB apart in PSNR, yet their end-effector prediction errors differ 13.7-fold.
Robotics (cs.RO)
arXiv:2610.15558 [cs.RO]
(or arXiv:2610.15558v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.15558 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 287 words
Related coverage
Exploring the Intrinsic Geometry of Diffusion Models with Constrained Inverse Kinematics
arXiv · Jul 22, 2026
Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills
arXiv · Aug 5, 2026
ChainSplat: A Physics-Inspired Screw-Theoretic Model for Learning Deformable Linear Object Dynamics from Multi-View RGB Videos
arXiv · Aug 31, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026