Safer, Smarter Construction Sites with Edge AI and Caterpillar Autonomous Machines - Ep. 285

What It Covers

Caterpillar's VP of AI Brandon Hootman explains how the century-old equipment manufacturer deploys NVIDIA edge AI systems in construction machines, achieving 100-millisecond supply chain optimization and autonomous capabilities through digital twins and onboard intelligence.

Key Questions Answered

• •Edge AI in Cabs: Caterpillar integrates NVIDIA Thor compute platforms with Riva voice services into excavator cabs, enabling operators to access machine controls and company knowledge through natural language without taking eyes off jobsites, developed in months not years.
• •Manufacturing Optimization: Using NVIDIA Omniverse digital twins and cuOpt models, Caterpillar calculates clear-to-build status across 30-day windows in 100 milliseconds at facility level, processing complex multi-layer supply chain signals that previously required extensive manual analysis by teams.
• •Autonomous Safety Approach: AI-based autonomy requires millions of simulation hours using digital twins of jobsites and machines before physical testing. Foundation models get reinforced with Caterpillar's century of operational data to handle construction-specific scenarios that generic models cannot address.
• •Sensor Innovation: Quadruped robots with thermal, acoustic and visual sensors walk factory floors to perform predictive maintenance on legacy equipment without shutting down production for months to install permanent sensors, simultaneously updating digital twin data in real-time.