A simulation of a real robot workspace must preserve task-relevant interactions, while policies developed in it must operate on observations available to the real robot. Yet scene reconstruction and policy development are often treated separately. We present Agentic Real-to-Sim-to-Real (Agentic RSR), a framework that l...
Yi-Han Li, Ya-Ting Feng, Sheng-Jiu Sun et al.· 1 citation
A data-efficient diffusion RL post-training framework - GQRM (Group Q-score Reweighted Matching), which achieves state-of-the-art cross-embodiment visual navigation performance and introduces two complementary designs: a self-bootstrapped exploration strategy with behavior perturbation that preserves the pretrained pol...
Tian-Yu Yang, Yi-Ming Zeng, Wen-Zhe Cai et al.· arXiv.org· 0 citations
Mobile manipulation requires generating whole-body action chunks that jointly satisfy goal reaching, collision avoidance, base kinematic constraints, manipulator joint limits, and trajectory smoothness. Flow-based generative policies provide an efficient paradigm for learning multimodal and temporally consistent motion...
Unordered Landmark Visual Navigation (ULVN) is proposed, a unified RGB-only framework free from temporal and odometric priors that constructs a robust 2D topological map directly from unstructured images via calibrated geometric verification and maximum spanning forest refinement.
Hao Ren, Junzhe Zhu, Yihan Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
