CoPRE: Improving Sensitivity in Proprioceptive Contact Detection for Low-Cost Robot Arms

Yuxiao Zhu; Jinzhou Li; Yifei Dong; Muhammad Suhail; Chunyuan Yang; Xinyuan Luo; Haoyu Li; Boyuan Chen

cs.ROSep 23, 2026

CoPRE: Improving Sensitivity in Proprioceptive Contact Detection for Low-Cost Robot Arms

Authors: Yuxiao Zhu, Jinzhou Li, Yifei Dong, Muhammad Suhail, Chunyuan Yang, Xinyuan Luo, Haoyu Li, Boyuan Chen,

Organizations: Duke University · KTH Royal Institute of Technology · Carnegie Mellon University

Abstract

Contact detection during robotic manipulation allows robots to recognize unexpected contact and adapt their motion accordingly. However, in low-cost robot arms without dedicated force or tactile sensors, detecting weak contacts from proprioception is challenging because the resulting changes in joint-level proprioceptive signals can be small compared to normal variation and noise caused by robot motion itself. We introduce Contact-free Proprioceptive Response Estimation (CoPRE), improving proprioceptive contact detection sensitivity using only contact-free motion, without additional force sensors, contact labels, or analytical dynamics models. CoPRE estimate the expected joint torques under contact-free motion from proprioceptive state history and commanded motion, while removing recent observations that may already reflect contact. It then computes the residual between the expected and observed joint torque estimates, and maps this residual to a contact score using a noise-weighted Jacobian. Real-robot experiments on ARX Arm and Unitree G1 show that CoPRE achieves 74.1% and 82.2% recall on the tested contact trials, compared with 0%/0% on ARX and 16.3%/42.2% on G1 for the learned torque-prediction and inverse-dynamics baselines. CoPRE also reaches 90% detection rate for pushing force at 3.5 N on ARX and 5.5 N on G1. To demonstrate the downstream utility of our method, we implement belief-space manipulation planning for obstacle-aware object placement and book insertion where detected contacts update the spatial belief and enable the robot to retreat from blocked motions, adjust its pose, and retry. Project website at https://copre-arm.github.io

Figures & tables

Fig. 1: Weak-contact detection and reactive manipulation with CoPRE. Built-in joint states and torque estimates provide contact feedback on ARX and G1, using contact-free training data without additional force sensors or force labels. Bottom: on ARX, detected contact triggers protective retreat during placement (left) and pose adjustment during book insertion (right).

Fig. 2: CoPRE: from proprioception to contact-guided action. (A) Contact-free recordings train the nominal-response predictor and calibrate the detection threshold. (B) A history of $L$ state–command pairs and known commands predicts $H$ states, excluding observations within that interval. Torque residuals yield a contact score through bias correction and noise-weighted Jacobian aggregation. The schematic omits bias correction. Three consecutive samples at or above threshold confirm contact. (C) Contact feedback guides retreat or pose adjustment and retry.

Fig. 3: Weak-contact detection on ARX. Columns show separately measured reference sliding resistances of the book stacks. Curves show threshold-normalized scores at 0.04 m/s: mean and 10–90% range over trials and model seeds. Dotted lines mark threshold 1; vertical lines mark geometric contact references. Percentages report trial-level detection across all three pushing speeds. Orange (T) uses task-matched contact-free motion calibration; blue (B) adds broad contact-free motion. Thresholds permit zero confirmed calibration alarms. Only CoPRE retains detections under combined calibration.

Fig. 4: Weak-contact detection on the G1 right arm. Top: book stacks with separately measured reference sliding resistances. (a) Trial-level detection within 0.5 s of the geometric contact reference, pooled across three speeds and averaged over model seeds. (b) $F_{90}$ is the lowest tested resistance with at least 90% detection at that and all higher levels; $>5.5$ means it was not reached. Recall pools all trials. Task/Broad FA t is held-out task-matched/broader contact-free time in confirmed alarm (%). All methods share a 1% calibration FA t budget. CoPRE reaches $F_{90}=5.5$ N.

State exclusion	Aggregation	ARX: Task-matched ( $\gamma_{T}$ )			ARX: Combined ( $\gamma_{B}$ )				G1: Task-matched ( $\gamma_{T}$ )			G1: Combined ( $\gamma_{B}$ )
State exclusion	Aggregation	$F_{90}$	R	Broad FA t	$\gamma_{\rm br}/\gamma_{T}$	$F_{90}$	R	Broad FA t	$F_{90}$	R	Broad FA t	$\gamma_{\rm br}/\gamma_{T}$	$F_{90}$	R	Broad FA t
Off	W. Jac.	$>5.5$	$40.0$	$4.67$	$1.52$	$>5.5$	$3.7$	$0.42$	$>5.5$	$65.9$	$0.88$	$0.96$	$>5.5$	$64.4$	$0.84$
Off	Std. norm	$>5.5$	$56.3$	$74.61$	$3.53$	$>5.5$	$0.0$	$0.63$	$>5.5$	$12.6$	$1.82$	$1.06$	$>5.5$	$9.6$	$1.11$
On	Std. norm	$3.5$	$88.9$	$85.48$	$2.60$	$>5.5$	$0.0$	$11.79$	$>5.5$	$49.6$	$5.40$	$1.18$	$>5.5$	$31.9$	$1.02$
On	W. Jac.	$3.5$	$74.8$	$3.68$	$1.02$	$3.5$	$74.1$	$3.32$	$5.5$	$83.0$	$0.92$	$0.94$	$5.5$	$82.2$	$0.86$

Fig. 5: Ablations of prediction, scoring, and temporal design. Panels compare (a) state exclusion and residual aggregation, (b) noise weighting, (c) state exclusion across horizons

H

, (d) G1 detection windows with fixed thresholds, (e) prediction horizons, (f) predictor families, and (g) prediction and temporal filtering. Off/On disables/enables state exclusion within the prediction interval. Std. norm denotes the standardized torque residual norm; W./U. Jac. denotes weighted/unweighted Jacobian scoring.

\gamma_{T}

and

\gamma_{\rm br}

are thresholds for task-matched and broad contact-free motion, with

\gamma_{B}=\max(\gamma_{T},\gamma_{\rm br})

for combined calibration, used in (b)–(g). For G1, this maximum is computed per fold and seed; reported ratios divide the mean broad threshold by the mean task threshold.

F_{90}

(N) is the lowest tested reference sliding resistance with at least 90% detection at that and all higher levels;

>5.5

indicates that the criterion is unmet. R is overall trial recall (%); FA t is the percentage of held-out broad contact-free motion spent in confirmed alarm. Blue marks defaults; Direct and Instantaneous share the same configuration. State exclusion and weighted aggregation jointly retain sensitivity under broader calibration.

Fig. 6: Contact-guided manipulation on ARX. (A) A cup obstructs placement of a carried block. CoPRE triggers retreat and alternate placement, leaving the cup upright (3a–5a); the neural baseline continues forward and tips it over (3b–5b). Insets: threshold-normalized detector scores; dotted lines: threshold 1. (B) Contact with a neighboring book prompts pose adjustment and a successful insertion retry. Both sequences are qualitative demonstrations.

Robotic Perception

Robotic perception enables robots to interpret scenes, objects, motion, and contact from cameras and other sensors to guide action.

See all 320 papers

Jun 23, 2026

NoContactNoWorries: Estimating Contact through Vision and Proprioception for In-Hand Dexterous Manipulation

Perceiving physical contact is fundamental to dexterous manipulation. While robots often rely on dedicated hardware tactile sensors, humans exhibit a remarkable ability to infer contact by integrating visual information with an innate sense of their body's pose and movement. Inspired by this embodied perceptual skill, we investigate whether a robot can learn to infer contact from vision, an approach that also offers a scalable alternative to tactile hardware specifically for binary contact estimation, which faces practical challenges in cost, fragility, and integration. We present NoContactNoWorries, a transformer-based multimodal framework that fuses RGB-D vision with the robot's proprioception to infer binary contact states as a pseudo-tactile signal for hand-object interactions. We validate by training a single contact prediction model on multiple objects and show that the inferred contact signal supports downstream reinforcement learning agents for in-hand object reorientation, generalizing to novel objects. Experiments in both simulation and on a real-world robot validate our approach, highlighting the feasibility of inferring contact from vision and proprioception. Project Page: https://soham2560.github.io/no-contact-no-worries/

Soham Patil, Avirup Das, Sourabh Bhosale +1

Robotics Research Center (RRC), International Institute of Information Technology (IIIT), Hyderabad, India · Department of Computer Science, The University of Manchester

Vision-Based Tactile Sensing Dexterous Manipulation In-Hand Manipulation Robotic Tactile Sensing Multimodal Sensor Fusion

Ask PDF

Jul 3, 2026

Current as Touch: Proprioceptive Contact Feedback for Compliant Dexterous Manipulation

Compliance is essential for dexterous manipulation, yet existing solutions often rely on external tactile or force sensors that are costly, fragile, and difficult to deploy on low-cost robot hands. We propose a proprioception-driven framework that learns contact-aware compliance cues from motor current and joint states. Since motor current is closely related to actuator torque, it provides an intrinsic signal for perceiving contact force, object resistance, and grasp stability without additional sensing hardware. Rather than estimating external wrenches or commanding torque, our method predicts a compliance reference position: an ideal joint-position target for a standard PD controller whose induced position error generates appropriate grasping force. This position-based formulation is compatible with mainstream teleoperation and policy-learning pipelines, while enabling the robot to adapt interaction forces from real-time proprioceptive feedback. Thus, motor current serves not only as a force proxy but also as a learnable proprioceptive contact signal for compliance reference prediction. Experiments on multiple dexterous hands and contact-rich tasks, including fragile object handling, sustained surface contact, thin-object retrieval, and dynamic load adaptation, show stable compliant grasping, safer and more efficient teleoperation, and improved downstream policy learning without external tactile or force sensors.

Chenyang Ma, Yunchao Yao, Zhenyu Wei +3

University of North Carolina at Chapel Hill

Robotic Control Compliant Robot Control Dexterous Manipulation Contact-Rich Robotic Manipulation

Ask PDF

Aug 30, 2026

Blind Dexterity: Whole-Body Humanoid Manipulation via Pure Proprioception

We present blind, whole-body manipulation skills on a Unitree G1 humanoid using only onboard proprioception, without cameras, markers, force-torque, or tactile sensors. Despite this minimal sensing, the trained policies exhibit surprising capability across qualitatively different tasks: push-resilient bipedal walking without IMU feedback, active soccer ball trapping with a foot, seeking and lifting a suitcase by its handle, and mounting a randomly positioned skateboard. We argue that these capabilities arise from a key underappreciated signal: the way the joint encoder readouts evolve under purposeful compliant contact, effectively forming a whole-body tactile channel. By generating contact-rich motions, the trained policies actively probe the environment; as a result, task-relevant object state (e.g., pose) becomes increasingly decodable from short proprioceptive histories. We expose this information using compact task-specific state estimators trained alongside, but fully separately from, the policies; their prediction errors decrease rapidly after informative contact. Our results indicate that joint encoder-based proprioception, combined with compliant actuation (now widely available on commercial robots and low-cost motors) is already a strong, practical substrate for whole-body dexterous manipulation and interactive perception, and therefore a natural foundation on which richer sensing can be layered.

Aditya Bhatt, Oleg Kaidanov, Puze Liu +1

Intelligent Autonomous Systems Lab, TU Darmstadt · German Research Center for AI (DFKI) · Tongji University, Shanghai Research Institute for Autonomous Intelligent Systems +1

Whole-Body Robotic Manipulation Robotic Manipulation Humanoid Robotics Contact-Rich Robotic Manipulation Robotic Perception

Ask PDF

Abstract

Robotic Perception

Explore similar work

NoContactNoWorries: Estimating Contact through Vision and Proprioception for In-Hand Dexterous Manipulation

Current as Touch: Proprioceptive Contact Feedback for Compliant Dexterous Manipulation

Blind Dexterity: Whole-Body Humanoid Manipulation via Pure Proprioception