RoboBoom
Humanoids Summit· October 19, 2026· 16:39

Advanced Bionic Hands for Humans & Robots | Aadeel Akhtar (PSYONIC)

Aadeel Akhtar of PSYONIC presents at Humanoids Summit on bionic hands that serve prosthetic users and help robots learn manipulation tasks.

In this talk

  1. 0:09

    Human skills can train robots

    Akhtar says prosthetic users have learned everyday and specialized tasks, including tying shoes, stocking shelves, and pipetting. He argues that their demonstrations can help robots improve at manipulation, where they often struggle despite performing showy movements.

  2. 0:58

    A capable, durable hand

    Akhtar says more than 350 patients in the United States use PSYONIC’s FDA-approved hand, which is covered by Medicare, and more than 90 robotics companies use it. The hand weighs 490 grams, has six motors and 30 touch sensors, detects weights from about 10 grams to one kilogram, and can lift up to 63 kilograms. He demonstrates its grip options and describes durability tests, including drops and impacts, and fingers that can be replaced in about 10 minutes.

  3. 4:05

    One hand across people and robots

    Akhtar shows the hand on industrial robot arms and humanoids, including an Apptronik Apollo assembling car parts at Mercedes and NASA-related demonstrations involving space-station tasks. He says the same hand can move between a person and a robot. A universal fabric socket developed with London-based Koalaa can be adjusted for different limbs and fitted with a data streamer.

  4. 7:06

    Collecting data to teach tasks

    Akhtar describes paying prosthetic users for data that can be used to train robots, including users who may have limited access to affordable devices. In one example, a person pipettes liquid between beakers and a Universal Robots arm, Unitree robot dog, and Unitree G1 humanoid perform the task. The data can combine hand sensors with egocentric video and be sent to NVIDIA tools, including GR00T; a phone app also streams motor and touch data to the cloud.

  5. 11:13

    Connecting human and machine control

    Akhtar names Arm, NVIDIA, and ABB as partners for hand computing, model training, and robot deployment. He also describes research with microwire electrodes that could record nerve signals and stimulate sensation, as well as a brain implant demonstration in which Scott Embry controlled a robot hand and detected finger touch. He closes with a patient who returned to baseball with a bionic hand and hit what Akhtar calls the first bionic home run.

Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.

More talks