Search papers, labs, and topics across Lattice.
This paper introduces a novel architecture for safe robot navigation that integrates natural language safety rules and operator preferences by translating them into Signal Temporal Logic (STL) specifications. These specifications, combined with VLM-based scene understanding, guide robot planning and navigation in unstructured environments. The approach grounds persistent rules in a 2D cost map and uses runtime STL monitoring to enforce dynamic requirements, demonstrating a system that satisfies both hard constraints and soft preferences.
Guaranteeing safe robot navigation in unstructured environments just got easier: translate human language rules into formal logic, ground them with VLMs, and let the robot navigate.
We propose an architecture for integrating high-level, human-provided safety rules and operator-aligned semantic preferences into autonomous robot navigation in unstructured outdoor environments. In our approach, natural-language rules are translated into Signal Temporal Logic (STL) specifications that guide planning and navigation during runtime. Persistent, environment-centric rules and terrain preferences are grounded into a 2D cost map, while temporally dynamic requirements are expressed as STL specifications to be monitored during runtime. We hypothesize the use of Vision-Language Models (VLMs) for zero-shot scene understanding, enabling mapping between human instructions, semantic features, and environmental constraints. Within this framework, we construct an illustrative navigation model that is designed to satisfy a set of STL-encoded specifications and soft operator preferences through formal satisfaction metrics embedded into environmental properties and runtime monitoring.
