Direction Matters: Learning Force Direction Enables Sim-to-Real Contact-Rich Manipulation

• URL: http://arxiv.org/abs/2602.14174v1
• Date: Sun, 15 Feb 2026 14:57:13 GMT
• Title: Direction Matters: Learning Force Direction Enables Sim-to-Real Contact-Rich Manipulation
• Abstract summary: We propose a framework that leverages expert-designed controller logic for transfer.<n>Inspired by the success of privileged supervision in tasks, we employ a human-designed finite state machine based position/force controller in simulation.<n>Unlike force magnitudes, which are highly sensitive to simulation inaccuracies, force directions encode high-level task geometry and remain robust across the sim-to-real gap.
• Score: 28.707088673008172
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Sim-to-real transfer for contact-rich manipulation remains challenging due to the inherent discrepancy in contact dynamics. While existing methods often rely on costly real-world data or utilize blind compliance through fixed controllers, we propose a framework that leverages expert-designed controller logic for transfer. Inspired by the success of privileged supervision in kinematic tasks, we employ a human-designed finite state machine based position/force controller in simulation to provide privileged guidance. The resulting policy is trained to predict the end-effector pose, contact state, and crucially the desired contact force direction. Unlike force magnitudes, which are highly sensitive to simulation inaccuracies, force directions encode high-level task geometry and remain robust across the sim-to-real gap. At deployment, these predictions configure a force-aware admittance controller. By combining the policy's directional intent with a constant, low-cost manually tuned force magnitude, the system generates adaptive, task-aligned compliance. This tuning is lightweight, typically requiring only a single scalar per contact state. We provide theoretical analysis for stability and robustness to disturbances. Experiments on four real-world tasks, i.e., microwave opening, peg-in-hole, whiteboard wiping, and door opening, demonstrate that our approach significantly outperforms strong baselines in both success rate and robustness. Videos are available at: https://yifei-y.github.io/project-pages/DirectionMatters/.

Related papers

• DexTeleop-0: Force-Aware Bimanual Dexterous Teleoperation with Ego-Centric Perception towards Shared Autonomy [6.652421441096025]
  Fine-grained, bimanual dexterous manipulation remains a foundational challenge in robotics.<n>Traditional teleoperation systems fail in contact-rich tasks because embodiment gaps hinder accurate kinematic mapping.<n>We propose a tactile-driven adaptation strategy designed to enable fine-grained manipulation on top of teleoperation pipelines.
  arXiv  Detail & Related papers  (2026-06-22T14:48:58Z)
• IMPACT: Learning Internal-Model Predictive Control for Forceful Robotic Manipulation [43.12484757720005]
  Real-world robotic manipulation tasks often involve forceful interactions with the environment.<n>We propose IMPACT, a framework that decouples these forceful tasks into task-planning and internal-model-based predictive control.
  arXiv  Detail & Related papers  (2026-06-09T13:00:56Z)
• Beyond Binary: Sim-to-Real Dexterous Manipulation with Physics-Grounded Contact Representation [51.463992799813816]
  A primary bottleneck in contact-rich manipulation is the difficulty of collecting real-world data.<n>We introduce Center-of-Pressure, an effective tactile representation grounded in physical principles.<n>We evaluate CoP on two blind, challenging contact-rich manipulation tasks.
  arXiv  Detail & Related papers  (2026-05-27T17:59:02Z)
• ForceFlow: Learning to Feel and Act via Contact-Driven Flow Matching [53.30290192030814]
  ForceFlow is a force-aware reactive framework built upon flow matching.<n>We show that ForceFlow achieves a 37% success rate improvement over the strong baseline ForceVLA.<n>It also exhibits accurate force signal prediction and superior performance in contact force self-regulation.
  arXiv  Detail & Related papers  (2026-05-11T13:27:00Z)
• CONTACT: CONtact-aware TACTile Learning for Robotic Disassembly [5.119250053135017]
  We investigate the role of tactile sensing in robotic disassembly through both simulation and real-world experiments.<n>We compare three sensing configurations: Vision Only, Vision + tactile RGB (TacRGB), and Vision + tactile force field (TacFF)<n>Our results show that tactile sensing plays a critical, task-dependent role in robotic disassembly, with structured force-field representations being particularly effective in contact-dominated scenarios.
  arXiv  Detail & Related papers  (2026-03-09T16:20:49Z)
• ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation [55.467742403416175]
  We introduce a physics-driven neural algorithm that translates large-scale motion capture to humanoid embodiments.<n>We learn a unified multimodal controller that supports both dense references and sparse task specifications.<n>Results show that ULTRA generalizes to autonomous, goal-conditioned whole-body loco-manipulation from egocentric perception.
  arXiv  Detail & Related papers  (2026-03-03T18:59:29Z)
• Minimalist Compliance Control [24.451467626022943]
  Compliance control is essential for safe physical interaction, yet its adoption is limited by hardware requirements such as force torque sensors.<n>We propose Minimalist Compliance Control, which enables compliant behavior using only motor current or voltage signals.<n>We validate our approach on a robot arm, a dexterous hand, and two humanoid robots across multiple contact-rich tasks.
  arXiv  Detail & Related papers  (2026-03-01T04:25:54Z)
• Semantic-Contact Fields for Category-Level Generalizable Tactile Tool Manipulation [82.63833405368159]
  Generalizing tool manipulation requires both semantic planning and precise physical control.<n>We propose Semantic-Contact Fields (SCFields), a unified 3D representation fusing visual semantics with dense contact estimates.<n> Experiments on scraping, crayon drawing, and peeling demonstrate robust category-level generalization.
  arXiv  Detail & Related papers  (2026-02-14T16:05:08Z)
• UniForce: A Unified Latent Force Model for Robot Manipulation with Diverse Tactile Sensors [51.88112610411651]
  We propose a novel unified tactile representation learning framework that learns a shared latent force space across diverse tactile sensors.<n>UniForce reduces cross-sensor domain shift by jointly modeling inverse dynamics (image-to-force) and forward dynamics (force-to-image)<n>To avoid reliance on expensive external force/torque (F/T) sensors, we exploit static equilibrium and collect force-paired data via direct sensor--object--sensor interactions.
  arXiv  Detail & Related papers  (2026-02-01T11:03:01Z)
• Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulation [12.509181374985936]
  Human-like dexterous hands with multiple fingers offer human-level manipulation capabilities.<n>But training control policies that can directly deploy on real hardware remains difficult due to contact-rich physics.<n>We present a practical framework that utilizes dense tactile feedback combined with joint torque sensing to regulate physical interactions.
  arXiv  Detail & Related papers  (2026-01-06T07:26:39Z)
• Towards Transferring Tactile-based Continuous Force Control Policies from Simulation to Robot [19.789369416528604]
  grasp force control aims to manipulate objects safely by limiting the amount of force exerted on the object. Prior works have either hand-modeled their force controllers, employed model-based approaches, or have not shown sim-to-real transfer. We propose a model-free deep reinforcement learning approach trained in simulation and then transferred to the robot without further fine-tuning.
  arXiv  Detail & Related papers  (2023-11-13T11:29:06Z)
• Nonprehensile Riemannian Motion Predictive Control [57.295751294224765]
  We introduce a novel Real-to-Sim reward analysis technique to reliably imagine and predict the outcome of taking possible actions for a real robotic platform. We produce a closed-loop controller to reactively push objects in a continuous action space. We observe that RMPC is robust in cluttered as well as occluded environments and outperforms the baselines.
  arXiv  Detail & Related papers  (2021-11-15T18:50:04Z)
• COCOI: Contact-aware Online Context Inference for Generalizable Non-planar Pushing [87.7257446869134]
  General contact-rich manipulation problems are long-standing challenges in robotics. Deep reinforcement learning has shown great potential in solving robot manipulation tasks. We propose COCOI, a deep RL method that encodes a context embedding of dynamics properties online.
  arXiv  Detail & Related papers  (2020-11-23T08:20:21Z)
• Learning Compliance Adaptation in Contact-Rich Manipulation [81.40695846555955]
  We propose a novel approach for learning predictive models of force profiles required for contact-rich tasks. The approach combines an anomaly detection based on Bidirectional Gated Recurrent Units (Bi-GRU) and an adaptive force/impedance controller.
  arXiv  Detail & Related papers  (2020-05-01T05:23:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.