SPRING 2022 GRASP Seminar - Robert J. Wood, Harvard University
Robert J. Wood, a robotics researcher at Harvard University, discusses designing soft manipulators for delicate and dexterous tasks, including deep-sea sampling, in a GRASP Lab seminar.
Robert J. Wood, a robotics researcher at Harvard, describes work on the mechanical side of artificial intelligence at a CMU Robotics Institute seminar.
Wood frames his lab’s work as designing robotic bodies, materials, and manufacturing methods that make perception, control, and autonomy easier. He introduces flapping insect robots and legged microrobots, whose size and constraints call for custom fabrication rather than off-the-shelf parts.
To speed up prototyping, the lab developed layered, pop-up fabrication: flat materials are patterned, laminated, and raised into mechanisms with a single input. Wood explains how a flapping robot uses one actuator for coupled wing motions, and how motion capture and feedback controllers help stabilize its otherwise unstable flight. Small robots can also tolerate crashes and use electroadhesion to perch or climb.
Wood contrasts conventional metal tools with soft, fiber-reinforced grippers that curl around delicate organisms using seawater pressure. Foam in the fingers provides contact-force feedback through its material response, helping operators grasp organisms with limited visual feedback. He also describes a rotary-actuated enclosure designed to examine midwater animals and release them unharmed.
Wood describes the Octobot, which combines peroxide fuel, a microfluidic oscillator, and soft actuators in an untethered all-soft system that moves its limbs. He raises the question of how fluidic circuits could support richer behavior, and shows how embedded printing and soft lithography can create channels and articulated structures. He also points to muscle-like dielectric elastomer actuators and self-folding robots.
In the discussion, Wood says fabrication has become more repeatable, while designing motion-producing structures remains difficult; the lab’s Pop-Up CAD software is an early aid. He argues that softness can make contact easier to handle, but complex systems still need modeling and control. For wearable motion sensing, he says machine learning can interpret signals while soft materials may reduce the burden of understanding deformation and contact.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Robert J. Wood, a robotics researcher at Harvard University, discusses designing soft manipulators for delicate and dexterous tasks, including deep-sea sampling, in a GRASP Lab seminar.