Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Data-Driven Contact-Aware Control Method for Real-Time Deformable Tool Manipulation: A Case Study in the Environmental Swabbing

Siavash Mahmoudi, Amirreza Davar, Dongyi Wang

TL;DR

This paper tackles the challenge of real-time deformable tool manipulation by introducing State-Adaptive Koopman LQR (SA-KLQR), a data-driven control framework that linearizes nonlinear tool-surface interactions via lifted observables and adapts through region-based operator switching. The approach is augmented with a tactile sensing regime, including an embedded FSR within the sponge, and a centroid-based fuzzy regulation to preserve even force distribution and suppress tool distortion during swabbing. Empirical results show SA-KLQR achieving superior force tracking and trajectory stability compared with PID and SMC across multiple trajectories and frequencies, along with improved surface coverage due to the centroid regulation. The work bridges data-driven Koopman modeling with classical optimal control, delivering a practical, real-time solution for deformable tool manipulation in industrial hygiene applications with potential for broader automation in food safety.

Abstract

Deformable Object Manipulation (DOM) remains a critical challenge in robotics due to the complexities of developing suitable model-based control strategies. Deformable Tool Manipulation (DTM) further complicates this task by introducing additional uncertainties between the robot and its environment. While humans effortlessly manipulate deformable tools using touch and experience, robotic systems struggle to maintain stability and precision. To address these challenges, we present a novel State-Adaptive Koopman LQR (SA-KLQR) control framework for real-time deformable tool manipulation, demonstrated through a case study in environmental swab sampling for food safety. This method leverages Koopman operator-based control to linearize nonlinear dynamics while adapting to state-dependent variations in tool deformation and contact forces. A tactile-based feedback system dynamically estimates and regulates the swab tool's angle, contact pressure, and surface coverage, ensuring compliance with food safety standards. Additionally, a sensor-embedded contact pad monitors force distribution to mitigate tool pivoting and deformation, improving stability during dynamic interactions. Experimental results validate the SA-KLQR approach, demonstrating accurate contact angle estimation, robust trajectory tracking, and reliable force regulation. The proposed framework enhances precision, adaptability, and real-time control in deformable tool manipulation, bridging the gap between data-driven learning and optimal control in robotic interaction tasks.

Data-Driven Contact-Aware Control Method for Real-Time Deformable Tool Manipulation: A Case Study in the Environmental Swabbing

TL;DR

This paper tackles the challenge of real-time deformable tool manipulation by introducing State-Adaptive Koopman LQR (SA-KLQR), a data-driven control framework that linearizes nonlinear tool-surface interactions via lifted observables and adapts through region-based operator switching. The approach is augmented with a tactile sensing regime, including an embedded FSR within the sponge, and a centroid-based fuzzy regulation to preserve even force distribution and suppress tool distortion during swabbing. Empirical results show SA-KLQR achieving superior force tracking and trajectory stability compared with PID and SMC across multiple trajectories and frequencies, along with improved surface coverage due to the centroid regulation. The work bridges data-driven Koopman modeling with classical optimal control, delivering a practical, real-time solution for deformable tool manipulation in industrial hygiene applications with potential for broader automation in food safety.

Abstract

Deformable Object Manipulation (DOM) remains a critical challenge in robotics due to the complexities of developing suitable model-based control strategies. Deformable Tool Manipulation (DTM) further complicates this task by introducing additional uncertainties between the robot and its environment. While humans effortlessly manipulate deformable tools using touch and experience, robotic systems struggle to maintain stability and precision. To address these challenges, we present a novel State-Adaptive Koopman LQR (SA-KLQR) control framework for real-time deformable tool manipulation, demonstrated through a case study in environmental swab sampling for food safety. This method leverages Koopman operator-based control to linearize nonlinear dynamics while adapting to state-dependent variations in tool deformation and contact forces. A tactile-based feedback system dynamically estimates and regulates the swab tool's angle, contact pressure, and surface coverage, ensuring compliance with food safety standards. Additionally, a sensor-embedded contact pad monitors force distribution to mitigate tool pivoting and deformation, improving stability during dynamic interactions. Experimental results validate the SA-KLQR approach, demonstrating accurate contact angle estimation, robust trajectory tracking, and reliable force regulation. The proposed framework enhances precision, adaptability, and real-time control in deformable tool manipulation, bridging the gap between data-driven learning and optimal control in robotic interaction tasks.

Paper Structure

This paper contains 16 sections, 26 equations, 14 figures, 3 tables, 2 algorithms.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORK
  3. SYSTEM AND FRAMEWORK IDENTIFICATION
  4. Tactile Sensors Utilization
  5. Framework Description
  6. State Adaptive Optimal Angle Alignment and Contact Distribution Control
  7. Operator-Based System Linearization
  8. SA-KLQR Controller
  9. Centroid-Based Fuzzy Force Regulation
  10. EXPERIMENTS, RESULTS AND DISCUSSIONS
  11. Embedded FSR Sensor Behavior Evaluation
  12. Koopman Operator Performance Evaluation
  13. SA-KLQR Force Control and Trajectory Tracking
  14. Tool Force Distribution and Coverage Evaluation
  15. Future Work
  16. ...and 1 more sections

Figures (14)

  • Figure 1: Swab sampling collection model with deformable sponge stick
  • Figure 2: The robotic system setup consists of two different contact sensors (Waterproof FSR sensor and Contact Detector Pad) and a 3D printed swab holder gripper.
  • Figure 4: FSR Array Matrix Pad Contact Distribution Map results based on the sponge coverage contact
  • Figure 5: Control scheme of the proposed overall framework
  • Figure 6: End-effector rolling angle and Koopman-based force regulation for maintaining stable contact pressure at the given location.
  • ...and 9 more figures