New AI Lets Robots Feel Their Way Through Tricky Tasks
Robots that sense force could soon fold your laundry or assemble your furniture.
FORCA is a coding agent that takes visual observations and wrist force/torque measurements as input and writes Python programs for contact probing and force-responsive motion. Between executions, it inspects force records and visual outcomes to diagnose failures and revise both motion and force-feedback control logic.
It requires no demonstrations, task-specific training, or simulation. Across five real-robot tasks, researchers demonstrated its effectiveness in contact-rich manipulation: improving manipulation accuracy, adapting to changes in object or surface properties, and regulating contact under surface variation and object motion. Compared with a vision-only variant of the same agent, FORCA improved grasp accuracy and task success and reduced contact loss in wiping and peeling.
- FORCA is an AI that helps robots use force sensors to perform tasks that require a sense of touch.
- It improved robot performance in tasks like wiping and peeling by learning from mistakes and adjusting its grip.
- This could lead to robots that can assist with household chores or delicate manufacturing work.
Why It Matters
Robots with a sense of touch could soon handle delicate chores and factory tasks, making our lives easier.