A vision-language-action policy for a simulated bimanual lab task, from LIDAR. It controls the upper body of a Vega-U humanoid robot: two 7-DoF arms and two 20-DoF Wuji Hand 2 hands. The task is the first half of a pipetting routine: The policy was fine-tuned from the openpi pi0.5 base model on demonstrations that a scripted teacher generated in the same simulation (Isaac Sim 6.0.1 / Isaac Lab, GPU PhysX), across randomized object layouts. - head (ego) camera; - a right-wrist camera looking back at the thumb and the plunger; - left wrist camera; - right wrist camera. Use the companion simulation repository, vega-wuji-pipette-sim. Its README covers: - installing Isaac Sim 6.0.1, Isaac Lab…
Independent publisher
Peiyu Song
birbirll
Models in Library1
Datasets in Library0
Models on Hugging Face18
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Robotics.
Reported licenses: gemma. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.