Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Manipulation via Force Distribution at Contact

Haegu Lee, Yitaek Kim, Casper Hewson Rask, Christoffer Sloth

TL;DR

This work tackles the limitations of point-contact models in contact-rich manipulation by introducing the Force-Distributed Line Contact (FDLC) model, which distributes contact forces along a line via two points connected by a virtual spring-damper. A bi-level optimization framework combines a lower-level contact-force optimization with an upper-level iLQR trajectory optimizer, integrating FDLC into the dynamics to enable direct torque generation through force distribution. Through simulation and real-world experiments on a box-rotation task, the method demonstrates lower control effort, reduced end-effector travel, and improved robustness to slip and model uncertainty compared to point-contact models. The findings suggest that FDLC can achieve human-like dexterity in manipulation by leveraging distributed line contacts, with practical impact for efficient, robust robotic manipulation of soft or compliant objects.

Abstract

Efficient and robust trajectories play a crucial role in contact-rich manipulation, which demands accurate mod- eling of object-robot interactions. Many existing approaches rely on point contact models due to their computational effi- ciency. Simple contact models are computationally efficient but inherently limited for achieving human-like, contact-rich ma- nipulation, as they fail to capture key frictional dynamics and torque generation observed in human manipulation. This study introduces a Force-Distributed Line Contact (FDLC) model in contact-rich manipulation and compares it against conventional point contact models. A bi-level optimization framework is constructed, in which the lower-level solves an optimization problem for contact force computation, and the upper-level optimization applies iLQR for trajectory optimization. Through this framework, the limitations of point contact are demon- strated, and the benefits of the FDLC in generating efficient and robust trajectories are established. The effectiveness of the proposed approach is validated by a box rotating task, demonstrating that FDLC enables trajectories generated via non-uniform force distributions along the contact line, while requiring lower control effort and less motion of the robot.

Manipulation via Force Distribution at Contact

TL;DR

This work tackles the limitations of point-contact models in contact-rich manipulation by introducing the Force-Distributed Line Contact (FDLC) model, which distributes contact forces along a line via two points connected by a virtual spring-damper. A bi-level optimization framework combines a lower-level contact-force optimization with an upper-level iLQR trajectory optimizer, integrating FDLC into the dynamics to enable direct torque generation through force distribution. Through simulation and real-world experiments on a box-rotation task, the method demonstrates lower control effort, reduced end-effector travel, and improved robustness to slip and model uncertainty compared to point-contact models. The findings suggest that FDLC can achieve human-like dexterity in manipulation by leveraging distributed line contacts, with practical impact for efficient, robust robotic manipulation of soft or compliant objects.

Abstract

Efficient and robust trajectories play a crucial role in contact-rich manipulation, which demands accurate mod- eling of object-robot interactions. Many existing approaches rely on point contact models due to their computational effi- ciency. Simple contact models are computationally efficient but inherently limited for achieving human-like, contact-rich ma- nipulation, as they fail to capture key frictional dynamics and torque generation observed in human manipulation. This study introduces a Force-Distributed Line Contact (FDLC) model in contact-rich manipulation and compares it against conventional point contact models. A bi-level optimization framework is constructed, in which the lower-level solves an optimization problem for contact force computation, and the upper-level optimization applies iLQR for trajectory optimization. Through this framework, the limitations of point contact are demon- strated, and the benefits of the FDLC in generating efficient and robust trajectories are established. The effectiveness of the proposed approach is validated by a box rotating task, demonstrating that FDLC enables trajectories generated via non-uniform force distributions along the contact line, while requiring lower control effort and less motion of the robot.
Paper Structure (13 sections, 5 equations, 9 figures, 1 table)

This paper contains 13 sections, 5 equations, 9 figures, 1 table.

Table of Contents

  1. Introduction
  2. Contributions
  3. Related Work
  4. Method
  5. Force-Distributed Line Contact Model
  6. Bi-level Optimization for Trajectory generation
  7. Simulation
  8. Control Effort and Travel Distance
  9. Contact Forces and Robustness
  10. Experimental Results
  11. Experiment Setup
  12. Open-Loop Controller and Force Distribution
  13. Conclusions

Figures (9)

  • Figure 1: Overview of the proposed framework combining iLQR, optimization-based dynamics with the Force-Distributed Line Contact (FDLC) model. Changes in force distribution at contact is shown
  • Figure 2: Illustration of the Force-Distributed Line Contact (FDLC) model, approximated by two points connected by a virtual spring–damper system, allowing for non-uniform force distribution.
  • Figure 3: Execution of trajectories generated by the point contact model (left) and the FDLC model (right) in the box rotation task.
  • Figure 4: Control input from generated trajectory for each $\theta_\textnormal{goal}$. The row (a) shows the control input with point contact model, while the row (b) shows the control input with FDLC model.
  • Figure 5: Control effort comparison between the FDLC model and the point contact model for different target angles.
  • ...and 4 more figures