mlsanityData to 14 Oct 2026

Data-Driven Physics Embedded Dynamics with Predictive Control and Reinforcement Learning for Quadrupeds

arXiv 2603.14333 ↗First posted 15 Mar 2026cs.RO cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

State of the art quadrupedal locomotion approaches integrate Model Predictive Control (MPC) with Reinforcement Learning (RL), enabling complex motion capabilities with planning and terrain adaptive behaviors. However, they often face compounding errors over long horizons and have limited interpretability due to the absence of physical inductive biases. We address these issues by integrating Lagrangian Neural Networks (LNNs) into an RL MPC framework, enabling physically consistent dynamics learning. At deployment, our inverse dynamics infinite horizon MPC scheme avoids costly matrix inversions, improving computational efficiency by up to 4x with minimal loss of task performance. We validate our framework through multiple ablations of the proposed LNN and its variants. We show improved sample efficiency, reduced long-horizon error, and faster real time planning compared to unstructured neural dynamics. Lastly, we also test our framework on the Unitree Go1 robot to show real world viability.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 25 of its 31 references that are in the map. First posted on arXiv 15 Mar 2026.

Builds on

25 of 31 references.

References · 25

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Neural Ordinary Differential Equationscited by 3,779
  3. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  4. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
  5. 2019Hamiltonian Neural Networkscited by 712
  6. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 512
  7. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 512
  8. 2019Learning agile and dynamic motor skills for legged robotscited by 504
  9. 2020Lagrangian Neural Networkscited by 398
  10. 2017Asymmetric Actor Critic for Image-Based Robot Learningcited by 238
  11. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 212
  12. 2017Imagination-Augmented Agents for Deep Reinforcement Learningcited by 140
13 more
  1. 2015Model Predictive Path Integral Control using Covariance Variable Importance Samplingcited by 124
  2. 2020Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraintscited by 105
  3. 2023DreamWaQ: Learning Robust Quadrupedal Locomotion With Implicit Terrain Imagination via Deep Reinforcement Learningcited by 66
  4. 2020Learning Off-Policy with Online Planningcited by 53
  5. 2021Neural Networks with Physics-Informed Architectures and Constraints for Dynamical Systems Modelingcited by 41
  6. 2023Hybrid Internal Model: Learning Agile Legged Locomotion with Simulated Robot Responsecited by 39
  7. 2019Deep Lagrangian Networks for end-to-end learning of energy-based control for under-actuated systemscited by 33
  8. 2022Physics-Informed Model-Based Reinforcement Learningcited by 18
  9. 2023Inverse Dynamics Trajectory Optimization for Contact-Implicit Model Predictive Controlcited by 8
  10. 2024PIP-Loco: A Proprioceptive Infinite Horizon Planning Framework for Quadrupedal Robot Locomotioncited by 7
  11. 2022Versatile Real-Time Motion Synthesis via Kino-Dynamic MPC with Hybrid-Systems DDPcited by 3
  12. 2025Investigating Lagrangian Neural Networks for Infinite Horizon Planning in Quadrupedal Locomotioncited by 3
  13. 2025Context-Aware Deep Lagrangian Networks for Model Predictive Controlcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Learning to Walk With Less: A Dyna-Style Approach to Quadrupedal Locomotion
  2. 2023A Real-World Quadrupedal Locomotion Benchmark for Offline Reinforcement Learning
  3. 2025Learning More With Less: Sample Efficient Model-Based RL for Loco-Manipulation
  4. 2024Reinforcement Learning For Quadrupedal Locomotion: Current Advancements And Future Perspectives
  5. 2025Residual MPC: Blending Reinforcement Learning with GPU-Parallelized Model Predictive Control
  6. 2025Reference Free Platform Adaptive Locomotion for Quadrupedal Robots using a Dynamics Conditioned Policy
  7. 2026Towards Torque-Driven Reinforcement Learning for Quadruped Locomotion
  8. 2025MULE: Multi-terrain and Unknown Load Adaptation for Effective Quadrupedal Locomotion