Event
OpenDoge controls quadruped robot without maps or contact sensors
1 reports · 1 sources
Guyuehome analyzed OpenDoge's quadruped robot locomotion control method: this hybrid internal model, without a terrain map or contact sensors, encodes observation history into a velocity estimate and a 16-dimensional latent state, which are used to determine gait phase and external disturbances. The article, combined with real inference data, examined the information carried by and missing from the latent state, but it did not provide a conclusion on the method's performance in real-machine locomotion.
Report timeline
- OpenDoge uses observation-history implicit state estimation to control quadruped robot locomotion
OpenDoge proposes a hybrid internal model for quadruped robot locomotion control that, without terrain maps or contact sensors, encodes observation history into a velocity estimate and a 16-dimensional latent state to determine gait phase and external disturbances. The article combines real inference data to examine what information the latent state carries and what it lacks.
Guyuehome ·