Force-Based Offset Estimation for Keyed Peg-in-Hole Assembly Using Local Gaussian Process Regression

Chandra Yuvesh Aubeeluck; Abilash Philip Madavath; Augustin Raju; Nicolas Pyschny; Felix Hackelöer; Florian Zwanzig

cs.ROAug 18, 2026

Force-Based Offset Estimation for Keyed Peg-in-Hole Assembly Using Local Gaussian Process Regression

Authors: Chandra Yuvesh Aubeeluck, Abilash Philip Madavath, Augustin Raju, Nicolas Pyschny, Felix Hackelöer, Florian Zwanzig

Organizations: Institute of Mechanical Engineering, Cologne University of Applied Sciences, Campus Gummersbach, Germany

Abstract

Key-keyway assembly tasks impose strict geometric constraints and are highly sensitive to grasp pose deviations in uncertain environments. This work presents a force-based offset estimation method for keyed peg-in-hole assembly, embedded within a perception-validation-insertion pipeline. Residual misalignment is estimated directly from wrist force/torque measurements using a local KNN-Gaussian Process hybrid regressor. The framework distinguishes between two contact regimes, hard collision and guided chamfer insertion, and routes inference to a dedicated model for each. Regime classification is achieved via a contact-window duration threshold. KNN combined with a deterministic search using the results of a post-grasp monocular visual validation contributes to an increased accuracy of the regressor model. This approach achieves accurate radial offset estimation in chamfered peg insertion, during a keypoint detection-based pick and place application. Experiments using the integrated force/torque sensor of a collaborative robot arm showed an increase in insertion success rate from 67% to 87% after the pipeline was applied.

Figures & tables

Fig. 1 : Insertion process with keyed circular hole.

Panel of Figure 2 — Fig. 2 : Representative grasp-induced error configurations. Blue cross-hair: TCP; red cross-hair: keyed hole axis.

Fig. 3 : Complete pipeline for picking and insertion.

Fig. 4 : Collision-aware grasp planning. Blue disc: target; others: obstacles.

Fig. 5 : IBVS setup for post-grasp planar pose validation.

Fig. 6 : Key–keyway contact wrench distribution during disc insertion under angular misalignment.

Fig. 7 : Contact wrench distribution under radial misalignment during insertion (z-axis).

Fig. 8 : Contact torque signature comparison over a number of trials for insertion scenarios.

Fig. 9 : PCA projection of force-feature space, coloured by $\|\boldsymbol{\Delta r}^{gt}\|$ . Star: query; circles: $k$ nearest neighbours.

Fig. 10 : (a) Gripper and disc setup, with ArUco markers for shaft localization, (b) collision during placement, (c) trajectory correction applied.

	Insertion	Collision	Angular
Trials	125	160	40
Conditions	24	32	8
Radial offset (mm)	0.5–1.5	2.0–3.5	–
Angular offset ( ∘ )	–	–	$\pm{8–11}$
Feature dim.	30 (6 ch.)	10 (2 ch.)	10 (2 ch.)
Contact data rows	$\geq 100$	$<100$	$<100$

TABLE I: Dataset Summary

\mathbf{p}^{\prime}(s)=\mathbf{p}+T_{\text{pred}}+s\,\mathbf{z}

Algorithm 1 Force-based Contact Correction for Insertion

Fig. 11 : Predicted vs. ground-truth planar offsets and cumulative error distribution.

Metric	GP	GP+KNN	GP+KNN+Mean Prior
MAE $d_{x}$	$0.535$ mm	$0.520$ mm	$\mathbf{0.458}$ mm
MAE $d_{z}$	$1.004$ mm	$\mathbf{0.721}$ mm	$0.734$ mm
MAE $\|\mathbf{e}\|$	$1.216$ mm	$0.972$ mm	$\mathbf{0.947}$ mm
RMSE $\|\mathbf{e}\|$	$1.487$ mm	$1.182$ mm	$\mathbf{1.148}$ mm
Median error	$0.972$ mm	$0.729$ mm	$\mathbf{0.719}$ mm
90th-percentile error	$2.441$ mm	$1.889$ mm	$\mathbf{1.804}$ mm

TABLE II : LOCO cross-validation results evaluating generalization to held-out offset conditions.

Robotic Insertion

Robotic insertion studies how robots place objects through openings or into tissue despite tight tolerances, uncertain alignment, and contact forces.

See all 46 papers

Touch2Insert: Zero-Shot Peg Insertion by Touching Intersections of Peg and Hole
Mar 4, 2026Masaru Yajima, Yuma Shin, Rei Kawakami +2Robotic Insertion Tactile Sensing for Robotic Manipulation
Institute of Science Tokyo, Tokyo, Japan. · Mitsubishi Electric, Kanagawa, Japan.
From Reach to Insert: Tactile-Augmented Precision Assembly under Sub-Millimeter Tolerances
May 6, 2026Xinpan Meng, Siyao Huang, JingPu Yang +6RL for Robotics Contact-Rich Robotic Manipulation
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China
InsertAnything: Generalizable Contact-Rich Precision Insertion from Simulation to Reality
Sep 21, 2026Zhenghua Ma, Xinpan Meng, Zeyu Liu +4RL for Robotics Contact-Rich Robotic Manipulation
School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China, and the State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China · PaXini AI Technology (Beijing) Co., Ltd., Beijing 100080, China

Mar 4, 2026J/K move · Enter open · S save

Touch2Insert: Zero-Shot Peg Insertion by Touching Intersections of Peg and Hole

Masaru Yajima, Yuma Shin, Rei Kawakami, Asako Kanezaki+1

Institute of Science Tokyo, Tokyo, Japan. · Mitsubishi Electric, Kanagawa, Japan.

Robotic Insertion Tactile Sensing for Robotic Manipulation Zero-Shot Robotic Manipulation Object Pose Estimation

Open paper AskPDF arXiv

Reliable insertion of industrial connectors remains a central challenge in robotics, requiring sub-millimeter precision under uncertainty and often without full visual access. Vision-based approaches struggle with occlusion and limited generalization, while learning-based policies frequently fail to transfer to unseen geometries. To address these limitations, we leverage tactile sensing, which captures local surface geometry at the point of contact and thus provides reliable information even under occlusion and across novel connector shapes. Building on this capability, we present Touch2Insert, a tactile-based framework for arbitrary peg insertion. Our method reconstructs cross-sectional geometry from high-resolution tactile images and estimates the relative pose of the hole with respect to the peg in a zero-shot manner. By aligning reconstructed shapes through registration, the framework enables insertion from a single contact without task-specific training. To evaluate its performance, we conducted experiments with three diverse connectors in both simulation and real-robot settings. The results indicate that Touch2Insert achieved sub-millimeter pose estimation accuracy for all connectors in simulation, and attained an average success rate of 86.7% on the real robot, thereby confirming the robustness and generalizability of tactile sensing for real-world robotic connector insertion.

Abstract

Robotic Insertion

Explore similar work

Touch2Insert: Zero-Shot Peg Insertion by Touching Intersections of Peg and Hole

From Reach to Insert: Tactile-Augmented Precision Assembly under Sub-Millimeter Tolerances

InsertAnything: Generalizable Contact-Rich Precision Insertion from Simulation to Reality