Berkeley's Robot Learning Breakthroughs Could Soon Be in Your Home
From folding laundry to assembling furniture, robots are getting smarter fast.
UC Berkeley's BAIR researchers will be out in force at the Conference on Robot Learning (CoRL), held November 9–12, 2026, in Austin, Texas. Per the article, coverage of the schedule includes 33 main-conference papers, 4 late-breaking results posters, and the listed workshop contributions, with paper inclusion defined as at least one author affiliated with UC Berkeley (joint affiliations qualify).
The program also features an Early Career Keynote from Professor Negar Mehr. Session topics listed include generalist policies, memory and transfer; reinforcement learning and motion reasoning; dexterous learning, grasping and assembly; humanoid data, control and skills; humanoid locomotion and whole-body control; and optimization, agentic planning and evaluation. According to the article, the main conference demo track is not covered, and main-conference poster numbers follow an unofficial paper index, so check the official program on the day.
- UC Berkeley is presenting over 30 robot learning studies at CoRL 2026, covering dexterous manipulation, humanoid control, and generalist policies.
- The research aims to make robots that learn tasks by observation and practice, reducing the need for custom programming.
- Practical impact could include affordable home robots that assist with chores and caregiving within the next decade.
Why It Matters
This research could lead to affordable robot helpers that save you time on household chores and caregiving within the next decade.