The landmark papers by Pelrine et al. marked the beginning of the use of dielectric elastomer (DE) actuators (and sensors) for soft robotics. They showed that a voltage applied through the thickness of a soft material could produce large strains as a result of the Maxwell stress between charges on the two electrodes. Subsequent works have shown that DE-based actuators are particularly attractive for many soft robotic applications because they exhibit large energy densities with large strains and muscle-like response times. In recent years, we’ve been focusing on developing new configurations, models, fabrication, and design methods for long-life-cycle, high-power-density dielectric elastomer artificial muscles. I’ll present some of our recent work on such aspects and their applications in wearables, microrobotics, and other devices that require compact and high-power actuators.

Computational modeling has long been central to the success of robotics, enabling precise, agile motion in rigid-body systems ranging from aerial vehicles to humanoids. In contrast, soft and hybrid soft–rigid robots rely on continuously deformable structures whose dynamics are significantly more complex, and for which traditional rigid-body modeling pipelines are insufficient. Although soft robot simulation has advanced significantly, developing models that are both accurate and efficient, and compatible with large-scale gradient-based optimization, remains a fundamental challenge.
This talk presents an analytically differentiable dynamics framework for hybrid soft–rigid robotic systems. By deriving structured analytical derivatives of the governing equations, the framework enables efficient, well-posed solutions to both simulation and motion planning problems. The approach is demonstrated across multiple case studies, including implicit time integration and trajectory optimization, highlighting substantial computational gains and improved scalability for highly deformable systems. These results establish a unified computational foundation for soft robotics, narrowing the long-standing gap in model-based simulation and optimization between soft and rigid-body systems.

Dr. Federico Renda is an Associate Professor in the Department of Mechanical and Nuclear Engineering at Khalifa University in Abu Dhabi, UAE. Before joining Khalifa University, he was a Post-Doctoral Fellow at the BioRobotics Institute of Scuola Superiore Sant’Anna, where he received his Ph.D. degree in 2014. Dr. Renda has been a visiting professor at the National University of Singapore (NUS), the National Institute for Research in Digital Science and Technology (INRIA, Lille), and others. He has been appointed as an Editor and Program Committee Member of prominent robotics journals and conferences, including the International Journal of Robotics Research (IJRR) and the International Conference on Robotics and Automation (ICRA), among others. Dr. Renda’s research interests encompass the study of multibody flexible systems, with an emphasis on the modeling and control of soft and underwater robots.

The bioactuators that have been realized so far are all driven by ATP chemical energy, but they use the muscle contraction function of cells, which are the smallest unit of living organisms. However, a millimeter-scale hybrid device in which the precise mechanism generated by the mechanical interaction of biomolecular motors generated inside cells has been integrated and controlled on a large scale in vitro has not yet been realized. In this talk, firstly, I will show our recent result of the large scale molecular motor assembly process artificially controlled in vitro by the concept of 3D printing by microfluidic manipulation and stereolithography. We have demonstrated in situ flexible production system for dynamically reconfigurable microrobots, which can continuously manufacture a large variety of microrobots in small quantities and integrate and assemble actuators in one place. It can be expected to develop innovative energy-saving new principle devices that are fundamentally different from conventional energy-consuming manufacturing. Secondly, I will show one of the project to build AI platform using cyborg insect as a tool to design biohybrid intelligent matter, inspired by self organization process in natural behavior of insects.

Keisuke Morishima is a Professor, Department of Mechanical Engineering, and The Center for Advanced Medical Engineering and Informatics, The University of Osaka, JAPAN; he graduated from Nagoya University where he received his PhD in Engineering in 1998. In 1997, he was JSPS Postdoctoral Research Fellow. From 1998 to 2001, he was a Postdoctoral Research Associate, Department of Chemistry, Stanford University, USA. He joined Kanagawa Academy of Science and Technology as a Research Scientist in 2001. In 2004, he was a Visiting Research Fellow at Lund Institute of Technology, Sweden. In 2005, he joined Department of Mechanical Systems Engineering, Tokyo University of Agriculture and Technology as an Associate Professor. In 2007, he joined Department of Bio-Mechanics and Intelligent Systems. In 2011, he moved to Department of Mechanical Engineering, Osaka University as a Professor. He is mainly engaging in the research fields of Micro-Nano Robotics and its application to the micro-nanomanipulation, BioMEMS, living machine, soft & wet nano robotics.