Unverified50% confidenceEventExact time
At the GTC venue, the Pollen Robotics team demonstrated the process of training policies in a simulation environment and rapidly deploying them to a physical Microduck robot.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/25/2026
First Seen
Valid until: 11/8/2026
Sources
Related Entities
Related Claims
UnverifiedMicroduck is a small bipedal robot development tool developed by Pollen Robotics specifically for reinforcement learning.77% similarUnverifiedThe Pollen Robotics team said that Microduck's core advantage lies in encapsulating the complex problem of bipedal locomotion control into an interface suitable for training RL algorithms, letting developers focus on algorithm iteration rather than hardware debugging.72% similarUnverifiedDuring training, the agent exposed bugs in the simulator itself.66% similarUnverifiedThe locomotion capabilities of modern humanoid robots are largely attributable to large-scale reinforcement learning training in simulated environments.65% similarUnverifiedThe team expects thousands of people to train robot behavior policies on real hardware.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1097711API
curl https://kongchang.com/api/v1/knowledge/claims/1097711MCP
get_claim(id=1097711)