Reproduce “PneuTac: Tactile Manipulation with Soft Pneumatic Robots via Unified MPM-Gaussian Splatting Simulation” 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
Shaohong Zhong, Marco Pontin, Joe Watson, Perla Maiolino, Ingmar Posner
PneuTac simulates a soft robot and its tactile gel together, then supplements real demonstrations with simulated ones for tactile manipulation. It uses material point method models, with policies fine-tuned on real data. On egg-carton opening, Sim+Real reached 90.0 ± 5.8% success versus 26.7 ± 8.8% for Real, both trained with 10 real demonstrations. Across three tasks, more real data matched or exceeded it on one task.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.