Studies in the Nonlinear Dynamics and Control of Soft Robotic Manipulators: Modelling, Analysis and Morphological Computation

Loading...
Thumbnail Image

Persistent link to this item

Statistics
View Statistics

Journal Title

Journal ISSN

Volume Title

Title

Studies in the Nonlinear Dynamics and Control of Soft Robotic Manipulators: Modelling, Analysis and Morphological Computation

Published Date

2021-06

Publisher

Type

Thesis or Dissertation

Abstract

Soft Robotics is an exciting area of research which has been fast emerging over the past decade. Widely considered as a future paradigm for robotics, this area envisions precise control of intricate and flexible movements of soft robotic manipulators made entirely of soft materials. Dynamical and material response characteristics of soft materials necessitate a nonlinear approach to the dynamics and control of soft robots. However, novel mathematical and computational frameworks for modelling and analysis that are essential in this context are currently only emerging in the field. Motivated by this, the first part of this thesis introduces and develops a novel, coupled nonlinear oscillator network model for a generic, planar soft robotic manipulator. The model represents the coupling between translational and rotational degrees of freedom - essential to capture the flexible dynamics of a soft manipulator - and is inspired by a nonlinear oscillator model first studied in the 1970s. Nonlinear equations of motion (EOM) for the single-link case of the model, followed by EOM for double and triple link oscillator networks are all derived using the Lagrangian formulation. Results from detailed analyses of the dynamics are then presented, including the time-domain response, phase portraits and linearization results. In the second part, upon establishing controllability for the single and double link cases, a feedback loop control system with Proportional-Integral-Derivative (PID) controllers is designed to command the end-effector of the linearized double link system to trace a predetermined circular trajectory. A set of control results for a range of coupling values showing the performance of the control design are presented. The third and final part of the thesis employs the developed model to investigate morphological computation - the intriguing idea that the nonlinear dynamics of a physical body can serve to exploit the body itself as a computational device or reservoir. Morphological computation holds great promise for the future of distributed computing and decentralized control. Inspiration for morphological computation is drawn from nature, where the computational architecture that allows organisms such as the octopus to initiate and control their dynamics appears to be decentralized and distributed over their physical bodies. Morphological computation results, showing the performances of the two and three link realizations of the developed model as computational reservoirs in emulating a target output function, are presented. Importantly, it is shown that the three link model outperforms the two link case, suggesting that higher order realizations of the proposed nonlinear oscillator network model will likely enhance the capabilities of a soft robotic manipulator from the standpoint of morphological computation. In summary, the contributions of this thesis are expected to advance the modelling, analysis and control of nonlinear soft robotic manipulators as well as their realizations in the future as morphological computational devices.

Keywords

Description

University of Minnesota M.S.M.E. thesis. June 2021. Major: Mechanical Engineering. Advisor: Subramanian Ramakrishnan. 1 computer file (PDF); x, 185 pages.

Related to

Replaces

License

Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Suggested citation

Harma, Heather. (2021). Studies in the Nonlinear Dynamics and Control of Soft Robotic Manipulators: Modelling, Analysis and Morphological Computation. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/260111.

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.