Event
HC2Robot: A Humanoid Parkour Framework for Predicting Foot Landing Points
1 reports · 1 sources
Wenhao Xu and other authors submitted a preprint paper proposing HC2Robot, a deployable perception-based humanoid robot parkour framework composed of a Predictive Foresight Module (PFM) and a Hyperconnected (HC) Transformer. PFM predicts the timing, location, and support quality of the next touchdown event for each foot from deployable observations, converting privileged supervision into a deployable latent representation. The authors state that the deployed actor relies only on onboard perception to operate; on stairs, gaps, high platforms, and slopes, HC2Robot achieves an average traversal success rate of 97.60% in simulation, with safe landing rates consistently higher than strong perceptive locomotion baselines. These results are all simulation data.
Report timeline
- HC2Robot: Multi-Terrain Humanoid Parkour with Predicted Foresight and Hyper-Connection Transformers
HC2Robot proposes a deployable perceptive humanoid parkour framework composed of a Predicted Foresight Module (PFM) and a Hyper-Connection (HC) Transformer. The PFM predicts, from deployable observations, the timing, location, and support quality of each foot's next touchdown event, converting privileged supervision into a deployable latent representation.
arXiv cs.RO ·