Reproduce “NeuroVLA: A Brain-inspired Embodied Intelligence for Fluid and Fast Reflexive Robotics Control” 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
Weiyu Guo, He Zhang, Pengteng Li, Tiefu Cai, Ziyang Chen, Yandong Guo, Xiao He, Yongkui Yang and 2 more
NeuroVLA separates semantic planning from feedback-driven motor control, using a vision-language model to provide motor intent, a recurrent module to condition it on proprioceptive and force history, and stateful spiking neurons to produce actions. In evaluated tasks, it reduced mean absolute commanded jerk by 75.6% and achieved 54.8% recovery versus 0.0% for four VLA baselines in a tested obstruction scenario.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.