Data-Driven Model Predictive Control for the Contact-Rich Task of Food Cutting
Paper in proceeding, 2019

Modelling of contact-rich tasks is challenging and cannot be entirely solved using classical control approaches due to the difficulty of constructing an analytic description of the contact dynamics. Additionally, in a manipulation task like food-cutting, purely learning-based methods such as Reinforcement Learning, require either a vast amount of data that is expensive to collect on a real robot, or a highly realistic simulation environment, which is currently not available. This paper presents a data-driven control approach that employs a recurrent neural network to model the dynamics for a Model Predictive Controller. We build upon earlier work limited to torque-controlled robots and redefine it for velocity controlled ones. We incorporate force/torque sensor measurements, reformulate and further extend the control problem formulation. We evaluate the performance on objects used for training, as well as on unknown objects, by means of the cutting rates achieved and demonstrate that the method can efficiently treat different cases with only one dynamic model. Finally we investigate the behavior of the system during force-critical instances of cutting and illustrate its adaptive behavior in difficult cases.

Author

Ioanna Mitsioni

Royal Institute of Technology (KTH)

Yiannis Karayiannidis

Chalmers, Electrical Engineering, Systems and control

Johannes A. Stork

Örebro University

Danica Kragic

Royal Institute of Technology (KTH)

2019 IEEE-RAS 19TH INTERNATIONAL CONFERENCE ON HUMANOID ROBOTS (HUMANOIDS)

2164-0572 (ISSN)

Vol. 2019-October 244-250

IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)
Toronto, Canada,

Subject Categories

Production Engineering, Human Work Science and Ergonomics

Robotics

Control Engineering

Computer Science

DOI

10.1109/Humanoids43949.2019.9035011

More information

Latest update

7/17/2024