Design Considerations for the Development of Reliable Humanoid Robots

Josh Farina, IHMC

Talk Title: A Considerations Framework for Designing and Building Reliable Humanoid Robots

Josh is a mechanical engineer and research associate at IHMC focused on leading the mechanical development of the new Nadia humanoid robot platform. Josh's primary research interests include the development and engineering of humanoid robotic platforms, composite materials implementation, low-backlash mechanism design, hydraulic actuator design, and actuator dynamics and control.

Luca Marchionni, CTO, PAL Robotics

Talk Title: Kangaroo: Past, present and future

Luca obtained his BSc in Control Engineering and his MSc in Systems and Control Engineering, both from the University of Rome La Sapienza. He worked in FCA (Fiat Chrysler Automobiles Engineering and Design before joining PAL Robotics in 2009. As Software and Control Engineer he participated in the development of wheeled and biped robots, taking care of dynamic simulations, autonomous navigation, and bipedal walking software. Since 2014, Luca has been the company's Chief Technology Officer, in charge of supervising research and development activities and ensuring that they serve the company's overall business strategy.

Youngwoo Sim, UIUC

Talk Title: Struggling with High-DoF Co-Design? Design Metric and Method for Dynamic Humanoid

Developing humanoid requires a lot of engineering work which gives an impression of ‘more like art than science’. Although, empirical design principles help design process, yet integrating the principles and finding a concrete system design solution are laborious work. Towards more systematic science of humanoid design, I have been developing design metrics and integration methods. In this presentation, I will share findings on system-level design metrics that quantify mass distribution and backdrivability. Also, I will present my recent project on a practical and interactive design software for creating actuation system for humanoids. Finally, I will share an ongoing design process of a new humanoid named Dash.

Youngwoo Sim and is Ph.D student in the Department of Mechanical Science and Engineering at University of Urbana-Champaign working with Joao Ramos. He is focusing on developing design methodology of building agile yet practical humanoid robots. His goal is to find extremely efficient ways to explore vast design space and utilizing intuitions gained from those explorations.  

Asst. Prof. Donghyun Kim, U. Mass Amherst

Talk Title: Humanoid robot with actuated toes

Toes play a critical role in human movement, yet they have not been widely featured in the design of bipedal robots. The majority of existing humanoid robots have stiff, flat feet, which result in a slow, inefficient, and unnatural gait. In this presentation, I will introduce a novel leg design known as HyperLeg, which incorporates a toe mechanism that allows for an extensive range of motion and an enhanced push-off force. On behalf of Prof. Yongjae Kim, the architect of the HyperLeg mechanism, I will outline the targeted design principles and the performance considerations to enable advanced locomotion control for agile movement. Building upon the HyperLeg concept, we recently developed a hopping robot named StaccaToe with a vision of its extension into a humanoid robot, dubbed PresToe. During my presentation, I will delve into the considerations and strategies pursued in the leg design process, the details that shaped the mechanism's development, and the challenges we faced throughout our journey. Moreover, I will share the ongoing challenges and outstanding issues related to robot development and control.

Donghyun is an Assistant Professor at the University of Massachusetts Amherst. Prior to joining UMass, he was a postdoctoral research associate at MIT’s Biomimetic Robotics Lab from 2019 to 2020, and at the University of Texas at Austin’s Human-Centered Robotics Lab in 2018, where he also completed his Ph.D. in 2017. Donghyun's research primarily focuses on developing control architectures for dynamic legged robots with a strong emphasis on the experimental validation of his formulations. He designs control systems that integrate sensing, planning, and feedback controllers, accounting for the dynamics, real-time constraints, and hardware limitations. Currently, Donghyun is broadening his research horizons to include perception and machine intelligence, aiming to increase the adaptability and versatility of robotic systems across diverse terrains. His work has been recognized with a best paper award in Transactions on Mechatronics in 2020. His 2022 ICRA paper was nominated as a finalist for the outstanding dynamics and control paper, and his 2016 paper in Transactions on Robotics was selected as a finalist for the best whole-body control paper and video.

Florian Loeffl, DLR

Talk Title: A survey over principles improving reliability in DLRs robots – do you know more?

DLR‘s robot range spans a wide field. Ranging from space robots over industrial applications, exploration, medical care to service robots. Tough their reliability was not always a named development goal, they can serve as examples for design principles enhancing mechatronic reliability. In this talk a couple of principles to enhance reliability will be presented and hopefully completed in the following discussion.

Florian Loeffl received his M.Sc. Degree in Mechanical Engineering from Technical University of Munich, Germany, in 2008. He joined the German Aerospace Center (DLR) in 2014 as leader of the development team of the elastic bipedal robot C-Runner. Then he joined the development team of the quadruped Bert. He is currently head of the systems design group at DLRs Institute of Robotics and Mechatronics and leads the development of DLR’s new bipedal humanoid. His research interest is in the mechatronic design of (soft-) elastic (legged) robots, including co-design, potential- and inertial shaping, structural elasticity, mechanisms and systems engineering.

Andrew SaLoutos, MIT

Talk Title: Building more than just a robot: designing the MIT Humanoid as a robust and reliable research platform

Developing and deploying highly dynamic humanoid robots is an exciting endeavor with a long time horizon, but the immediate usefulness of these robots is ultimately tied to how well they can be utilized as research platforms for pushing current boundaries in hardware design and control. In this presentation, I will describe the design of the MIT Humanoid, focusing on our system-level design decisions for making our robot into a robust and reliable research platform. I will also share some of our recent results and discuss how these validate (and invalidate) our design decisions. Finally, I will close with a discussion of how we are currently closing the loop on our design process, including a brief overview of our future plans for the platform.

Andrew SaLoutos is a PhD candidate in Mechanical Engineering at the MIT Biomimetic Robotics Lab, supervised by Prof. Sangbae Kim. He is focused on designing dynamic and robust robotic platforms for manipulation and locomotion.

Tomas Slimak, TU Munich

Talk Title: TBD

Prof. Daniele Pucci, IIT

Talk Title: Towards Computational Embodied Intelligence: the ergoCub robot

Embodied Intelligence is that characteristics of the body that expresses a degree of capacity in achieving a specific task in a given environment. In artificial “robot” bodies, this property can be found in many cases.  Think of a passive walker going down a slope without actuation, sensors, and control; or of a hot-air balloon achieving flight by exploiting thermodynamic properties of the surrounding air. When the artificial device has to interact with a biological intelligent being, the degree of complexity in attempting to entail a degree of intelligence inside the body increases. Think of a humanoid robot that must carry a payload with a human being; it raises certain number of questions : what are the optimal robot body characteristics for the robot to achieve the task in collaboration with the human? This talk overviews recent method for providing preliminary answers to these questions by taking advantage of modern tachniques and technologies, like differentiable simulator running on GPUs. Also, it overviews how these techniques have been partially used for the design and development of the ergoCub humanoid robot.