Detailed Schedule

Modeling soft robots: a discussion on capabilities and limitations of numerous techniques

Caleb Rucker, University of Tennessee, Knoxville, TN
“Cosserat Rod Models for Soft Robots: When and How to Use Them”

Biography:

Abstract:

Student/Scientist Presentation, TBD
“TBD”
Abstract:

Girish Krishnan, University of Illinois
“Modeling and Inverse Design Framework for Fiber Reinforced Actuators”
Biography:

Abstract:

Student/Scientist Presentation, TBD
“TBD”
Abstract:

Madhusudhan Venkadesan, Yale University
“Rheology of Muscle and Lessons for Tunable Materials”
Biography:

Abstract:

Giorgio Oliveri, Luuk van Laake, Johannes T.B. Overvelde, AMOLF
“Decentralized Reinforced Learning of Emergent Behavior in Robotic Matter”
Biography:

Abstract: Soft robots have the potential to be more robust, adaptable, and safer for human interaction than traditional rigid robots. State-of-the-art developments push these soft robotic systems towards applications such as rehabilitation and diagnostic devices, exoskeletons for gait assistance, and grippers that can handle delicate objects. However, despite these exciting developments, their inherent non-linear response limits the number of actuators that can be accurately controlled simultaneously, especially in complex or unknown environments. To enable modularly scalable and autonomous soft robots we have developed a new type of soft robot that is assembled from identical 1D building blocks with embedded pneumatic actuation, position sensing and computation. In this robotic system, motility might emerge from local interactions, rather than from a central brain. Here we shows that we are able to implement decentralized learning in this system. Using a stochastic optimization approach, each building block individually adjusts its actuation phase continuously, in order to find the fastest way to move in a predefined direction. We show that even for larger number of modules, this robotic system is still capable of learning. As a result, the system is robust to damage, as it will adjust its behaviour accordingly.

Student/Scientist Presentation, TBD
“TBD”
Abstract:

Student/Scientist Presentation, TBD
“TBD”
Abstract:

Marc Killpack, Brigham Young University
“Learned Models for Optimal Model-based Control of Soft Robots”
Biography: Marc Killpack has been an assistant professor in the department of Mechanical Engineering at Brigham Young University (BYU) since 2013. He was awarded a NASA Early Career Faculty award which has funded research on soft robots and control of underdamped robot arms. Further soft robot research is currently being funded under an NSF EFRI award with collaborators at the University of Tulsa. His current research interests relate to improving modeling and control for soft and compliant robots. This includes applications to space exploration, search and rescue, disaster response and human-robot interaction. Marc completed his Ph.D. in Robotics from the Healthcare Robotics Lab (HRL) at the Georgia Institute of Technology. Prior to joining HRL, Marc completed Masters’ degrees in Mechanical Engineering in 2008 from both Georgia Tech and AM ParisTech (formerly ENSAM) in Metz, France. In 2007, Marc graduated with a Bachelor of Science in Mechanical Engineering from Brigham Young University.

Abstract: Model-based control can allow more aggressive and potentially high-performing systems. However, developing tractable and accurate models to improve control performance for soft robots is a difficult task. Over the past five years we have been developing model-based control methods for large-scale soft robots. The models we have used range from first-principle lumped parameter dynamic models to learned non-linear discrete-time models. In this presentation we will present a process to develop learned models that enable linear and nonlinear model predictive control for soft robots. We will discuss trade-offs in different approaches to learned models and present results relative to model-based control.