CorNav improves robot navigation safety and success on construction sites
CORNAV: Construction-Aware Reasoning for Robot Navigation on Active Worksites
Robotics
Summary
Construction sites are busy and dangerous places where robots can help but struggle to navigate safely using current methods. The authors created CORNAV, a system that uses blueprints and schedules in addition to recognizing objects, so robots better understand the site layout and timing. This helps robots avoid dangerous areas and find their way more reliably. They tested CORNAV on real sites and saw much higher success rates and safer navigation.
What this means in practice
- •For construction robotics teams: Enable robots to navigate active construction sites safely by integrating architectural layouts and current project schedules to avoid hazards.
- •For building maintenance teams: Use blueprint and schedule grounded robot navigation to support routine inspections and maintenance in complex building environments.
Authors
Parastoo Ali Pour, Deepak Prakash Kumar, Tommy Zhou, Pramod Khargonekar, Mohammad Abdullah Al Faruque
Abstract
The construction industry faces persistent labor shortages, low productivity that costs the global economy over $1.6 trillion annually, and one of the highest injury rates among major industries. These factors motivate the use of autonomous robots to improve efficiency and worker safety. Existing language-grounded navigation systems, however, rely on semantic scene understanding alone and lack access to construction-specific context such as architectural plans, evolving work schedules, and safety constraints. As a result, they localize permanent building features unreliably and cannot safely navigate active jobsites. We present CORNAV, a blueprint-grounded, schedule-aware navigation framework that operates from 2D CAD drawings and project schedules without requiring a Building Information Model. CORNAV aligns architectural blueprints against hierarchical open-vocabulary 3D scene graphs to ground object queries, converts project schedules into time-varying navigation constraints, and validates requests through an LLM-based safety module that escalates hazardous zones before planning. An A* planner then enforces mandatory exclusion zones while preferentially avoiding higher-risk areas. Across an indoor office and a real construction site, blueprint grounding raises task success from 13.0% to 72.2% over semantic retrieval alone, schedule awareness eliminates all hard-zone violations, and the safety module correctly rejects hazardous requests arising from mislabeled project schedules.