Force-Aware Reinforcement Learning with Hybrid Sensorless Force Estimation for Wheeled-Legged Loco-Manipulation

arXiv 2609.13779 ↗First posted 12 Sep 2026cs.ROcited by 0

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

Abstract (arXiv)

Force-controlled loco-manipulation requires a whole-body policy to coordinate locomotion and arm motion while regulating end-effector interaction forces. This is challenging under floating-base dynamics and changing support contacts, particularly when end-effector force/torque sensing is unavailable for control. This paper presents a force-aware reinforcement learning approach with hybrid sensorless force estimation for wheeled-legged loco-manipulation. The proposed method provides a structured estimate of the end-effector force as an explicit policy observation, enabling force-guided contact behavior without using an end-effector force/torque sensor for control. The force estimate is obtained by combining generalized momentum observation, contact-constrained wrench projection, and temporal residual learning: the model-based components extract the physically structured part of the whole-body disturbance, while the residual network compensates the remaining motion-dependent bias. The estimated force is integrated into a mode-conditioned whole-body policy with an axis-wise force/position selector, allowing free-space motion, pure force regulation, and hybrid force/position control within one controller. Simulation results demonstrate improved sensorless force estimation and force-control performance. Hardware experiments further validate the proposed controller through quantitative valve-rotation and hybrid wiping evaluations, together with force-guided door opening and zero-force human-guided motion on a real wheeled-legged platform.

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 23 of its 34 references that are in the map. First posted on arXiv 12 Sep 2026.

Builds on

23 of 34 references.

Closest ancestors · 2

Same direction, newest first

Other references · 21

2018On the Spectral Bias of Neural Networkscited by 1,341
2021RMA: Rapid Motor Adaptation for Legged Robotscited by 508
2019Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networkscited by 368
2022Deep Whole-Body Control: Learning a Unified Policy for Manipulation and Locomotioncited by 121
2022Concurrent Training of a Control Policy and a State Estimator for Dynamic and Robust Legged Locomotioncited by 88
2023DreamWaQ: Learning Robust Quadrupedal Locomotion With Implicit Terrain Imagination via Deep Reinforcement Learningcited by 66
2024Visual Whole-Body Control for Legged Loco-Manipulationcited by 62
2022Adapting Rapid Motor Adaptation for Bipedal Robotscited by 54
2022Combining Learning-based Locomotion Policy with Model-based Manipulation for Legged Mobile Manipulatorscited by 37
2023Efficient Sim-to-real Transfer of Contact-Rich Manipulation Skills with Online Admittance Residual Learningcited by 21
2024Learning Force Control for Legged Manipulationcited by 21
2022RoLoMa: Robust Loco-Manipulation for Quadruped Robots with Armscited by 16

9 more

2024Whole-body End-Effector Pose Trackingcited by 6
2022Contact Optimization for Non-Prehensile Loco-Manipulation via Hierarchical Model Predictive Controlcited by 5
2025Learning a Unified Policy for Position and Force Control in Legged Loco-Manipulationcited by 5
2023Proprioceptive External Torque Learning for Floating Base Robot and its Applications to Humanoid Locomotioncited by 3
2025Efficient Learning of A Unified Policy For Whole-body Manipulation and Locomotion Skillscited by 3
2023Fine Robotic Manipulation without Force/Torque Sensorcited by 2
2023Sensorless Estimation of Contact Using Deep-Learning for Human-Robot Interactioncited by 2
2022Collision detection and identification for a legged manipulatorcited by 1
2024Multi-Momentum Observer Contact Estimation for Bipedal Robotscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way