Date of Award
Master of Science
Soft-bodied robots have become increasingly popular due to their ability to per- form tasks that are difficult or impossible for traditional rigid robots. However, accurately modeling and controlling the movement and behavior of soft robots are very challenging due to their complex and dynamic nature. In recent years, Reservoir Computing has emerged as a promising approach to modeling and controlling soft robots. In this thesis, reservoir computing was used to create a digital twin of soft-bodied robots. Specifically, a digital twin of a spring-mass system was created using echo state network, a popular reservoir computing model. Furthermore, an optimal controller was trained using reservoir computing to drive the spring-mass system to follow a desired trajectory. Extensive simulations were carried out to validate the proposed methods. The results demonstrate the effectiveness of the proposed approach. For example, the digital twin model achieved 2% MAPE and the optimal controller achieved 8.7% MAPE for a 20-node 54-spring system.
Index Terms: Deep Learning, Digital Twin Modeling, Echo State Network, Optimal control, Soft Bodied Robotics, Time series prediction
Prairie View A & M University
Rights© 2021 Prairie View A & M University
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Date of Digitization
J. B Coleman Library
City of Publication
Cain, J. (2023). Digital Twin Modeling And Optimal Control Of Soft-Bodied Robotics Using Reservoir Computing. Retrieved from https://digitalcommons.pvamu.edu/pvamu-theses/1520