Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs
Abstract
An elusive goal in navigation research is to build an intelligent agent that can understand multimodal instructions including natural language and image, and perform useful navigation. To achieve this, we study a widely useful category of navigation tasks we call Multimodal Instruction Navigation with demonstration Tours (MINT), in which the environment prior is provided through a previously recorded demonstration video. Recent advances in Vision Language Models (VLMs) have shown a promising path in achieving this goal as it demonstrates capabilities in perceiving and reasoning about multimodal inputs. However, VLMs are typically trained to predict textual output and it is an open research question about how to best utilize them in navigation. To solve MINT, we present Mobility VLA, a hierarchical Vision-Language-Action (VLA) navigation policy that combines the environment understanding and common sense reasoning power of long-context VLMs and a robust low-level navigation policy based on topological graphs. The high-level policy consists of a long-context VLM that takes the demonstration tour video and the multimodal user instruction as input to find the goal frame in the tour video. Next, a low-level policy uses the goal frame and an offline constructed topological graph to generate robot actions at every timestep. We evaluated Mobility VLA in a 836m^2 real world environment and show that Mobility VLA has a high end-to-end success rates on previously unsolved multimodal instructions such as "Where should I return this?" while holding a plastic bin. A video demonstrating Mobility VLA can be found here: https://youtu.be/-Tof__Q8_5s
Cite
@article{arxiv.2407.07775,
title = {Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs},
author = {Hao-Tien Lewis Chiang and Zhuo Xu and Zipeng Fu and Mithun George Jacob and Tingnan Zhang and Tsang-Wei Edward Lee and Wenhao Yu and Connor Schenck and David Rendleman and Dhruv Shah and Fei Xia and Jasmine Hsu and Jonathan Hoech and Pete Florence and Sean Kirmani and Sumeet Singh and Vikas Sindhwani and Carolina Parada and Chelsea Finn and Peng Xu and Sergey Levine and Jie Tan},
journal= {arXiv preprint arXiv:2407.07775},
year = {2024}
}