NaVILA: Legged Robot Vision-Language-Action Model for Navigation
It addresses the challenge of flexible human-robot interaction and navigation in cluttered scenes for legged robots, representing an incremental advancement by integrating vision-language-action models with locomotion skills.
This paper tackles the problem of Vision-and-Language Navigation for legged robots by proposing NaVILA, a framework that translates human language instructions into low-level leg joint actions via mid-level language-based commands, achieving substantial improvements on existing benchmarks and demonstrating advantages in new realistic benchmarks with real-world experiments.
This paper proposes to solve the problem of Vision-and-Language Navigation with legged robots, which not only provides a flexible way for humans to command but also allows the robot to navigate through more challenging and cluttered scenes. However, it is non-trivial to translate human language instructions all the way to low-level leg joint actions. We propose NaVILA, a 2-level framework that unifies a Vision-Language-Action model (VLA) with locomotion skills. Instead of directly predicting low-level actions from VLA, NaVILA first generates mid-level actions with spatial information in the form of language, (e.g., "moving forward 75cm"), which serves as an input for a visual locomotion RL policy for execution. NaVILA substantially improves previous approaches on existing benchmarks. The same advantages are demonstrated in our newly developed benchmarks with IsaacLab, featuring more realistic scenes, low-level controls, and real-world robot experiments. We show more results at https://navila-bot.github.io/