Tom Zhang, founder and CEO of Daxo Robotics, discusses the company’s tendon-driven robotic hand and its design, capabilities, and possible applications with the RoboPapers hosts.
Zhang describes starting with a soft robot for apple picking, then pivoting after seasonal work made product development difficult. The team sold 600 apple-stem clippers, while the soft robot became part of his PhD research. In late March, they shifted to building a robotic hand using the same continuum robot mechanisms.
The hand has 108 motors and uses tendon-driven fingers that can bend continuously, with grip strength and stiffness that can be adjusted. Zhang reports a $1,200 bill of materials and cites demonstrations including phone use, circle drawing, and teleoperated pen spinning. He describes soft robotics as a spectrum of compliance and rigidity, rather than a sharply defined category.
Zhang says the hand drew circles with roughly one to two millimeters of consistency, and reports that adding tendons reduced trajectory noise up to a point. Inline sensors measure tension in each tendon, giving 108 force feedback signals that can help infer contact location; encoders are planned to add shape information. He says the modular design is intended to make motors, tendons, and fingers replaceable, while the system can keep operating with some motors lost.
Zhang sees repeated, simple components as a way to create complex behavior and address difficult, long-tail manipulation tasks such as wrapping wire harnesses. He imagines a platform where people can train and share physical skills across different embodiments. Over the next two years, Daxo plans to pursue a practical use case and build a skill acquisition pipeline, while exploring research collaborations.
Zhang says it remains unclear whether skills trained on one hand will transfer to another, though he sees similar dynamical attractors as a possibility. He expects to integrate the hand with an off-the-shelf arm and is considering how to provide versions to researchers. He also discusses artificial muscles as a possible future route to a smaller form factor, while noting their high power needs today.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Tom Zhang of Daxo Robotics discusses robotic hand design, the company’s technology and its path from student research and agricultural products to a dexterous hand at a GRASP Lab talk at Penn.
