MiniVLA-Nav v1: A Multi-Scene Simulation Dataset for Language-Conditioned Robot Navigation
This dataset provides a standardized benchmark for language-conditioned navigation in simulation, but it is an incremental contribution as it follows existing dataset paradigms without introducing new methods or breakthroughs.
The authors introduce MiniVLA-Nav v1, a simulation dataset with 1,174 episodes for language-conditioned robot navigation across four photorealistic environments, providing synchronized RGB, depth, segmentation, and expert action labels. The dataset supports in-distribution and out-of-distribution evaluation splits for benchmarking.
We present MiniVLA-Nav v1, a simulation dataset for Language-Conditioned Object Approach (LCOA) navigation: given a short natural-language instruction, an NVIDIA Nova Carter differential-drive robot must navigate to the named object and stop within 1 m across four photorealistic Isaac Sim environments (Office, Hospital, Full Warehouse, and Warehouse with Multiple Shelves). Each of the 1,174 episodes pairs an instruction with synchronized 640x640 RGB images, metric depth maps (float32, metres), and instance segmentation masks, together with continuous (v,omega) and 7x7 tokenized expert action labels recorded at 60 Hz from a vision-based proportional controller. Trajectory diversity is ensured through three spawn-distance tiers (near: 1.5-3.5 m, mid: 3.5-7.0 m, far: global curated points; Pearson r=0.94 between spawn distance and trajectory length), 12 object categories, 18 training templates, and 12 paraphrase-OOD templates. Five evaluation splits support in-distribution accuracy, template-paraphrase robustness, and OOD object-category benchmarking. The dataset is publicly available at https://huggingface.co/datasets/alibustami/miniVLA-Nav