How to train your robot: Wear a tiny one in a baby carrier
Briefly

Humanoid robots will initially be operated remotely to gather training data for Vision-Language-Action Models aimed at autonomous bipedal movement. Teleoperation techniques include exoskeletons and augmented reality with game controllers. The CHILD robot, designed by researchers at the University of Illinois Urbana-Champaign, allows operators to control movements through a baby carrier setup. This system supports intuitive control and has diverse applications. A related project, PAPRLE, connects input devices for robotic teleoperation. A paper detailing CHILD will be presented at the IEEE RAS HUMANOIDS Conference in October 2025.
Teleoperation has become a popular approach for collecting data to solve robotic tasks in the context of Physical/Embodied AI. CHILD is one of the solutions we proposed to enable users to provide direct commands via teleoperation for specific target tasks.
CHILD is integrated into this pipeline to support humanoid forms and dual-arm configurations, allowing portable and wireless control of target robots. As a general-purpose humanoid controller, CHILD has a wide range of potential applications. It enables intuitive robot control by general users.
Read at Theregister
[
|
]