Advanced soft robots: data-driven development, modeling and control
Submitted to the Croatian Science Foundation
In the last 15 years, the study of soft robots inspired by nature has expanded at an exponential rate. Compared to rigid robots, they have tremendously improved human-machine interaction safety. As a result, they increasingly find application in various medical fields, including robotic rehabilitation. The complexities encountered in deploying soft robots beyond laboratory settings stem from their inherent nonlinearity and formally infinite degrees of freedom, posing significant challenges in their design, development, modeling, and control. Building upon the previous research paper “Control of soft robots with inertial dynamics” authored by the Principal Investigator and published in Science Robotics (IF 2022: 25), the proposed interdisciplinary project aims to address these challenges within a biomechatronic design framework. Leveraging the Koopman operator framework, data-driven modeling and development of controllers applicable to soft rehabilitation robots will be carried out. Experimental data will be obtained using a motion capture apparatus, while research will also involve the development of methodology for modeling and predicting grip strength by measuring muscle activity using electromyographic sensors. This would enable adaptive control of the device and adjustment of the device’s output force for the subject to promote recovery during rehabilitation. Furthermore, an investigation of the prerequisites needed to develop soft structures applicable to rehabilitation robots, including kinematic analyses of wrist, material selection, and design w, will be performed. Numerical models will be developed to optimize the designs and achieve better durability and reliability, and experimental analyses will be performed. Finally, as an example of the application of structures being investigated and developed, a proof-of-concept prototype of an innovative soft robotic glove for rehabilitating patients with reduced hand mobility will be developed (Haggerty et al., 2023).
Integration of EMG sensing devices and hand dynamometer into ROS
An example of the 3D printed tendon-driven rehabilitation glove made on traditional robotics principles [B. Stanić, K. Dangubić, T. Galić, University of Rijeka, Faculty of Engineering, Course: Control of mechatronics systems (E. Kamenar)]
References
2023
SciRobot
Control of soft robots with inertial dynamics
David A Haggerty, Michael J Banks, Ervin Kamenar, Alan B Cao, and 3 more authors
Soft robots promise improved safety and capability over rigid robots when deployed near humans or in complex, delicate, and dynamic environments. However, infinite degrees of freedom and the potential for highly nonlinear dynamics severely complicate their modeling and control. Analytical and machine learning methodologies have been applied to model soft robots but with constraints: quasi-static motions, quasi-linear deflections, or both. Here, we advance the modeling and control of soft robots into the inertial, nonlinear regime. We controlled motions of a soft, continuum arm with velocities 10 times larger and accelerations 40 times larger than those of previous work and did so for high-deflection shapes with more than 110° of curvature. We leveraged a data-driven learning approach for modeling, based on Koopman operator theory, and we introduce the concept of the static Koopman operator as a pregain term in optimal control. Our approach is rapid, requiring less than 5 min of training; is computationally low cost, requiring as little as 0.5 s to build the model; and is design agnostic, learning and accurately controlling two morphologically different soft robots. This work advances rapid modeling and control for soft robots from the realm of quasi-static to inertial, laying the groundwork for the next generation of compliant and highly dynamic robots.
@article{haggerty2023control,title={Control of soft robots with inertial dynamics},author={Haggerty, David A and Banks, Michael J and Kamenar, Ervin and Cao, Alan B and Curtis, Patrick C and Mezi{\'c}, Igor and Hawkes, Elliot W},journal={Science robotics},volume={8},number={81},pages={eadd6864},year={2023},publisher={American Association for the Advancement of Science},doi={10.1126/scirobotics.add6864},}
