Model training method, device and storage medium for multiple-configuration robot
By constructing an initial task dynamic representation model and training a visual action network and a VLA policy network using a robotic arm operation dataset, the problem of the robot model's versatility and robustness across various robot configurations was solved, resulting in better execution performance.
CN122275009APending Publication Date: 2026-06-26BEIJING HUMANOID ROBOTICS INNOVATION CENTER CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING HUMANOID ROBOTICS INNOVATION CENTER CO LTD
- Filing Date
- 2025-10-30
- Publication Date
- 2026-06-26
Smart Images
Figure CN122275009A_ABST
Abstract
This application provides a model training method, device, and storage medium for robots with various configurations, relating to the field of robot control technology. The method includes: constructing an initial task dynamic representation model; training an initial unified visual-motor network (VLA) based on a dataset of robotic arm operations for robots with various configurations to obtain a target unified visual-motor network (VLA); using the target VLA, extracting features from the robotic arm operation image sequence and the corresponding robotic arm state sequence to obtain a latent space visual motion representation; the latent space visual motion representation is a concatenation of the latent space visual representation and the latent space motion representation; and using the latent space visual motion representation as an auxiliary supervision signal to train an initial VLA policy network to obtain a target VLA policy network. This application enhances the robustness and generality of model policy learning, improving the execution performance of robots with various configurations.
Need to check novelty before this filing date? Find Prior Art