Reproduce “Model-Based Geometry-Aware Generative Optimization for Constrained Locomotion Planning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhilin He, Xinyuan Wang, Changliu Liu
2GO guides locomotion planning by using task-weighted constraint geometry to shape diffusion refinement and filter exploration noise, while adapting reverse updates to risk and candidate diversity. In quadruped stepping-stones, its planning Safe-and-Success Rate is 0.88 versus MDOC at 0.68, with lower foothold-distance violation but longer planning time. These results support the combined method, not universal superiority or guaranteed execution safety.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.