Answers are grounded in this paper and cite their sources.
Engineering project demonstrates voice-driven waypoint navigation in a quadruped robot, highlighting seamless integration of natural language control and autonomous localization.
This project investigates natural-language control of autonomous mobile robots through the development of a voice-command navigation system for the Boston Dynamics Spot robot. I developed a Python-based pipeline that converts spoken input to text, interprets a user’s requested destination, matches natural-language location names to predefined waypoints, and generates navigation commands that interface with Spot’s GraphNav system. I created and processed a mapped environment in Sitterson Hall as a graph of waypoints and edges, allowing Spot to localize itself within the map and autonomously navigate between named destinations. The system integrates speech processing, language-based command interpretation, map management, localization, and autonomous navigation into a single workflow. I also began migrating the system from a laptop-based implementation to an NVIDIA Jetson connected to Spot, with the goal of supporting onboard processing and a more self-contained robotic system. This work provides a foundation for connecting human friendly interfaces with autonomous navigation systems and for future integration of external geographic data, such as latitude/longitude coordinates and routes, with Spot’s internal representation of physical environments.
No takes yet. Share an insight, caveat, or question.
Alexa Tinajero (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: