ISSN 3083-4471 API Registry JSON Sitemap
TIRAS Transactions on Intelligent Robotics and Autonomous Systems An open access journal of Kaggleyes Academy
TIRAS > Vol. 1, No. 3 (September 2026) > KAGG:2609.00006
Vol. 1, No. 3 (September 2026) — Article KAGG:2609.00006

System Identification versus Learned Models for Commodity Quadruped Velocity Interfaces

Authors: Lin Zhao, Maximilian Weber, Julia Hoffmann, Omar Farouk

Perception and Control Group, Technical University of Munich, Munich, Germany

[Received 2026-08-31; accepted 2026-09-11; published online 2026-09-14]

Abstract:

Navigation stacks for commodity quadrupeds command the robot through a vendor velocity interface whose response to commands, including delay, lag, cross-axis coupling, and saturation, is undocumented and differs between platforms. Planners therefore need a calibrated model of this interface, but it is unclear whether classical system identification or a learned model should provide it. This paper compares, under matched probe budgets and on the same planner-conditioned probe data, structured identification of a state-delay-plus-saturation model, a Gaussian-process surrogate, and a multilayer perceptron with one million parameters. Scoring uses navigation-relevant criteria: worst-direction tracking error, extrapolation error on unseen command regimes, calibration time, and sensitivity to command noise. On a Unitree Go2 and a Unitree B2, with 60 probe sessions per platform, 12 m of probe motion per session, and 100 held-out command regimes per platform, the structured model attains 9.2% worst-direction tracking error and 14.6% extrapolation error, against 9.7% and 16.3% for the Gaussian process and 9.5% and 15.6% for the perceptron, with the shortest calibration time (44 s) and the lowest noise sensitivity (2.4 points at 3% noise). The margins are modest, but the failure cases are informative: at 3% command noise the perceptron's extrapolation error on the B2 rises to 21%, whereas the structured model stays at 16.1%; conversely, the perceptron captures the B2's saturation knee 9% better in-regime. The resulting decision rule is to identify the structured model by default, and adopt a learned model only when the saturation shape is the binding constraint.

Journal: Transactions on Intelligent Robotics and Autonomous Systems (TIRAS)
Volume / Issue: Vol. 1, No. 3 (September 2026)
Article ID: KAGG:2609.00006
DOI: 10.5803/kagg.2609.00006
Received: 2026-08-31
Accepted: 2026-09-11
Published online: 2026-09-14
License: CC BY 4.0

References (42)

  1. T. Zang, H. Huang, S. Cheng, and W. Zhang. GAUGE: Planner-Conditioned Active Calibration of Opaque Quadruped Velocity Interfaces. arXiv:2607.02841, 2026. arXiv:2607.02841
  2. J. Hwangbo, J. Lee, A. Dosovitskiy, D. Bellicoso, V. Tsounis, V. Koltun, and M. Hutter. Learning Agile and Dynamic Motor Skills for Legged Robots. Science Robotics, arXiv:1901.08652, 2019. arXiv:1901.08652
  3. J. Lee, J. Hwangbo, L. Wellhausen, V. Koltun, and M. Hutter. Learning Quadrupedal Locomotion over Challenging Terrain. Science Robotics, arXiv:2010.11251, 2020. arXiv:2010.11251
  4. A. Kumar, Z. Fu, D. Pathak, and J. Malik. RMA: Rapid Motor Adaptation for Legged Robots. Robotics: Science and Systems, arXiv:2107.04034, 2021. arXiv:2107.04034
  5. N. Rudin, D. Hoeller, P. Reist, and M. Hutter. Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning. Conference on Robot Learning, arXiv:2109.11978, 2021. arXiv:2109.11978
  6. T. Miki, J. Lee, J. Hwangbo, L. Wellhausen, V. Koltun, and M. Hutter. Learning Robust Perceptive Locomotion for Quadrupedal Robots in the Wild. Science Robotics, arXiv:2201.08117, 2022. arXiv:2201.08117
  7. G. B. Margolis, G. Yang, K. Paigwar, T. Chen, and P. Agrawal. Rapid Locomotion via Reinforcement Learning. Robotics: Science and Systems, arXiv:2205.02824, 2022. arXiv:2205.02824
  8. G. B. Margolis and P. Agrawal. Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior. Conference on Robot Learning, arXiv:2212.03238, 2022. arXiv:2212.03238
  9. J. Tan, T. Zhang, E. Coumans, A. Iscen, Y. Bai, D. Hafner, S. Bohez, and V. Vanhoucke. Sim-to-Real: Learning Agile Locomotion for Quadruped Robots. Robotics: Science and Systems, arXiv:1804.10332, 2018. arXiv:1804.10332
  10. X. B. Peng, M. Andrychowicz, W. Zaremba, and P. Abbeel. Sim-to-Real Transfer of Robotic Control with Dynamics Randomization. IEEE International Conference on Robotics and Automation, arXiv:1710.06537, 2018. arXiv:1710.06537
  11. W. Yu, J. Tan, C. K. Liu, and G. Turk. Preparing for the Unknown: Learning a Universal Policy with Online System Identification. Robotics: Science and Systems, arXiv:1702.02453, 2017. arXiv:1702.02453
  12. J. Tobin, R. Fong, A. Ray, J. Schneider, W. Zaremba, and P. Abbeel. Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World. IEEE/RSJ International Conference on Intelligent Robots and Systems, arXiv:1703.06907, 2017. arXiv:1703.06907
  13. J. Schoukens and L. Ljung. Nonlinear System Identification: A User-Oriented Road Map. IEEE Control Systems Magazine, arXiv:1902.00683, 2019. arXiv:1902.00683
  14. G. Pillonetto, A. Aravkin, D. Gedon, L. Ljung, A. H. Ribeiro, and T. B. Schön. Deep Networks for System Identification: A Survey. arXiv:2301.12832, 2023. arXiv:2301.12832
  15. S. L. Brunton, J. L. Proctor, and J. N. Kutz. Discovering Governing Equations from Data by Sparse Identification of Nonlinear Dynamical Systems. Proceedings of the National Academy of Sciences, arXiv:1509.03580, 2016. arXiv:1509.03580
  16. H. Mania, M. I. Jordan, and B. Recht. Active Learning for Nonlinear System Identification with Guarantees. Journal of Machine Learning Research, arXiv:2006.10277, 2022. arXiv:2006.10277
  17. M. Buisson-Fenet, F. Solowjow, and S. Trimpe. Actively Learning Gaussian Process Dynamics. Learning for Dynamics and Control, arXiv:1911.09946, 2020. arXiv:1911.09946
  18. T. Lew, A. Sharma, J. Harrison, A. Bylard, and M. Pavone. Safe Active Dynamics Learning and Control: A Sequential Exploration-Exploitation Framework. IEEE Transactions on Robotics, arXiv:2008.11700, 2022. arXiv:2008.11700
  19. M. P. Deisenroth, D. Fox, and C. E. Rasmussen. Gaussian Processes for Data-Efficient Learning in Robotics and Control. IEEE Transactions on Pattern Analysis and Machine Intelligence, arXiv:1502.02860, 2015. arXiv:1502.02860
  20. C. E. Rasmussen and C. K. I. Williams. Gaussian Processes for Machine Learning. MIT Press, ISBN 0-262-18253-X, 2006. ISBN:0-262-18253-X
  21. L. Hewing, J. Kabzan, and M. N. Zeilinger. Cautious Model Predictive Control Using Gaussian Process Regression. IEEE Transactions on Control Systems Technology, arXiv:1705.10702, 2017. arXiv:1705.10702
  22. F. Berkenkamp, M. Turchetta, A. P. Schoellig, and A. Krause. Safe Model-based Reinforcement Learning with Stability Guarantees. Advances in Neural Information Processing Systems, arXiv:1705.08551, 2017. arXiv:1705.08551
  23. K. Chua, R. Calandra, R. McAllister, and S. Levine. Deep Reinforcement Learning in a Handful of Trials Using Probabilistic Dynamics Models. Advances in Neural Information Processing Systems, arXiv:1805.12114, 2018. arXiv:1805.12114
  24. A. Nagabandi, G. Kahn, R. S. Fearing, and S. Levine. Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning. IEEE International Conference on Robotics and Automation, arXiv:1708.02596, 2018. arXiv:1708.02596
  25. A. Nagabandi, I. Clavera, S. Liu, R. S. Fearing, P. Abbeel, S. Levine, and C. Finn. Learning to Adapt in Dynamic, Real-World Environments through Meta-Reinforcement Learning. International Conference on Learning Representations, arXiv:1803.11347, 2019. arXiv:1803.11347
  26. G. Torrente, E. Kaufmann, P. Föhn, and D. Scaramuzza. Data-Driven MPC for Quadrotors. IEEE Robotics and Automation Letters, arXiv:2102.05773, 2021. arXiv:2102.05773
  27. L. Bauersfeld, E. Kaufmann, P. Foehn, S. Sun, and D. Scaramuzza. NeuroBEM: Hybrid Aerodynamic Quadrotor Model. Robotics: Science and Systems, arXiv:2106.08015, 2021. arXiv:2106.08015
  28. M. O'Connell, G. Shi, X. Shi, K. Azizzadenesheli, A. Anandkumar, Y. Yue, and S.-J. Chung. Neural-Fly Enables Rapid Learning for Agile Flight in Strong Winds. Science Robotics, arXiv:2205.06908, 2022. arXiv:2205.06908
  29. D. Kim, J. Di Carlo, B. Katz, G. Bledt, and S. Kim. Highly Dynamic Quadruped Locomotion via Whole-Body Impulse Control and Model Predictive Control. arXiv:1909.06586, 2019. arXiv:1909.06586
  30. R. Grandia, F. Jenelten, S. Yang, F. Farshidian, and M. Hutter. Perceptive Locomotion through Nonlinear Model-Predictive Control. IEEE Transactions on Robotics, arXiv:2208.08373, 2023. arXiv:2208.08373
  31. X. Xiao, B. Liu, G. Warnell, and P. Stone. Motion Planning and Control for Mobile Robot Navigation Using Machine Learning: A Survey. Autonomous Robots, arXiv:2011.13112, 2022. arXiv:2011.13112
  32. X. Xiao, J. Biswas, and P. Stone. Learning Inverse Kinodynamics for Accurate High-Speed Off-Road Navigation on Unstructured Terrain. IEEE Robotics and Automation Letters, arXiv:2102.12667, 2021. arXiv:2102.12667
  33. H. Karnan, K. S. Sikand, P. Atreya, S. Rabiee, X. Xiao, G. Warnell, P. Stone, and J. Biswas. VI-IKD: High-Speed Accurate Off-Road Navigation Using Learned Visual-Inertial Inverse Kinodynamics. IEEE International Conference on Robotics and Automation, arXiv:2203.15983, 2022. arXiv:2203.15983
  34. L. Wellhausen and M. Hutter. ArtPlanner: Robust Legged Robot Navigation in the Field. Field Robotics, arXiv:2303.01420, 2023. arXiv:2303.01420
  35. N. Rudin, D. Hoeller, M. Bjelonic, and M. Hutter. Advanced Skills by Learning Locomotion and Local Navigation End-to-End. IEEE/RSJ International Conference on Intelligent Robots and Systems, arXiv:2209.12827, 2022. arXiv:2209.12827
  36. J. Truong, D. Yarats, T. Li, F. Meier, S. Chernova, D. Batra, and A. Rai. Learning Navigation Skills for Legged Robots with Learned Robot Embeddings. IEEE/RSJ International Conference on Intelligent Robots and Systems, arXiv:2011.12255, 2021. arXiv:2011.12255
  37. Z. Fu, A. Kumar, A. Agarwal, H. Qi, J. Malik, and D. Pathak. Coupling Vision and Proprioception for Navigation of Legged Robots. IEEE/CVF Conference on Computer Vision and Pattern Recognition, arXiv:2112.02094, 2022. arXiv:2112.02094
  38. D. Shah, A. Sridhar, A. Bhorkar, N. Hirose, and S. Levine. GNM: A General Navigation Model to Drive Any Robot. arXiv:2210.03370, 2022. arXiv:2210.03370
  39. B. Lakshminarayanan, A. Pritzel, and C. Blundell. Simple and Scalable Predictive Uncertainty Estimation Using Deep Ensembles. Advances in Neural Information Processing Systems, arXiv:1612.01474, 2017. arXiv:1612.01474
  40. C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger. On Calibration of Modern Neural Networks. International Conference on Machine Learning, arXiv:1706.04599, 2017. arXiv:1706.04599
  41. J. R. Gardner, G. Pleiss, D. Bindel, K. Q. Weinberger, and A. G. Wilson. GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration. Advances in Neural Information Processing Systems, arXiv:1809.11165, 2018. arXiv:1809.11165
  42. D. P. Kingma and J. Ba. Adam: A Method for Stochastic Optimization. International Conference on Learning Representations, arXiv:1412.6980, 2015. arXiv:1412.6980
Cite this article Lin Zhao, Maximilian Weber, Julia Hoffmann, Omar Farouk, "System Identification versus Learned Models for Commodity Quadruped Velocity Interfaces," Transactions on Intelligent Robotics and Autonomous Systems, 2026, vol. 1, no. 3, KAGG:2609.00006. DOI: 10.5803/kagg.2609.00006.