Subproject 9: Multi-variable robust control for spatial deformations of I-FRC
Table of contents
Motivation
In the case of spatial deformation, there are interactions between physical properties in different deformation directions that can be decoupled by multi-variable control. Mechanical properties can change depending on the direction (e.g. force direction during twisting), thus causing a strongly non-linear behavior in terms of deformation. The state of deformation can be detected using integrated sensor systems and adjusted by robust, real-time control algorithms in a targeted and reproducible manner.
State of the art and preliminary research
According to the literature, there are numerous approaches for one-dimensional position control with actuators based on shape memory alloys [1]. Within the 1st cohort, different models for one-dimensional deformation of fiber rubber composites were developed based on theoretical as well as experimental process analyses. In addition to models for nominal behavior, models for uncertain factors resulting from manufacturing tolerances, changing operating conditions (tension, pressure, temperature) and aging were also generated. Based on these system descriptions, robust controllers were developed and experimentally tested using a variety of fiber rubber composite samples [2]. Robust control unit design is done offline, and the control algorithms to be implemented function in real-time [3]. Additional research is required for the generalization of these design methods for multi-dimensional position control.
Test stand (left), structural image multi-variable control (right)
Scientific questions and project objectives
Spatial deformation can be specifically induced by several separate control actions. In terms of control technology, this is considered a multi-variable system where coupling between different controlling variables must be taken into account. Functions of the multi-variable controller include the targeted decoupling between controlled systems [4]. The modeling of uncertainties results in increased degrees of freedom in the case of multi-variable control. Hence, for robust control design, suitable and practical model structures are essential for real-time self-sufficient position control. In terms of the desired large spatial deformation, there is a focus on non-linearities from a geometrical point of view. Therefore, innovative control structures must be developed and parameterized. For the recording of measurement data, the focus will be on integrated sensors in composite structures so that additional filters and optimized algorithms must be developed and tested.
References
| [1] | Ren, Z., Zarepoor, M., Huang, X., Sabelhaus, A. P., & Majidi, C. (2021). Shape Memory Alloy (SMA) Actuator With Embedded Liquid Metal Curvature Sensor for Closed-Loop Control. Frontiers in Robotics and AI, 8, 9. |
| [2] | Keshtkar, N., & Röbenack, K. (2020). Unstructured uncertainty based modeling and robust stability analysis of textile-reinforced composites with embedded shape memory alloys. Algorithms, 13(1), 24. |
| [3] | Zhou, K., & Doyle, J. C. (1998). Essentials of Robust Control. Upper Saddle River, NJ: Prentice Hall. |
| [4] | Skogestad, S., & Postlethwaite, I. (2007). Multivariable feedback control: analysis and design. New York: Wiley. |
Contact
Institute of Control Theory (RST), Faculty of Electrical and Computer Engineering at TU Dresden
© K. Röbenack
Professor
NameMr Prof. Dr.-Ing. habil. Dipl.-Math. Klaus Röbenack
Director of the institute
Send encrypted email via the SecureMail portal (for TUD external users only).
Certificate of DFN-PKI for encrypted email communication.
SHA1 Fingerprint: 5D:0B:FB:6F:92:2B:87:36:F5:F5:F0:48:64:77:66:8C:DB:58:56:7A
Institute of Control Theory
Institute of Control Theory
Visiting address:
Institutsgebäude S7a, Room 405 (secretariat 406) Georg-Schumann-Straße 7a
01187 Dresden