NavLive has selected Nvidia’s Jetson Orin system-on-module to power its latest handheld AI-driven construction scanner, enhancing real-time performance and efficiency in on-site building data capture. The upgrade consolidates multiple processing units into a single, compact device, revolutionizing how construction professionals model and map real-world environments.
The new NavLive scanner integrates five synchronized sensors — three cameras, LiDAR, and an inertial measurement unit (IMU) — all streaming live data for simultaneous localization and mapping (SLAM) and computer vision applications. This enables precise real-time scanning of structural elements such as doors, walls, and windows, allowing instant feedback on construction accuracy.
Compact Power Through Nvidia Jetson Orin
Earlier scanner models required separate microcontrollers for each sensor, making them heavier and more power-intensive. By shifting to Nvidia Jetson Orin NX, which combines a CPU, GPU, and dedicated deep learning accelerators, NavLive achieved significant gains in both portability and performance.
“By moving to Jetson Orin NX, we reduced weight and power requirements while increasing capability,” said Dr. David Wisth, CTO and Co-founder of NavLive. “Customers immediately notice that the device is easier to hold, yet it delivers much more in terms of performance.”
Smarter and More Scalable AI Integration
The Nvidia-powered platform enables real-time object detection and supports AI-driven anonymization tools, such as face detection, to ensure data privacy compliance. Additionally, the scalable architecture allows for future software enhancements and new AI features as industry needs evolve.
NavLive’s decision underscores the growing importance of AI-powered sensors in the Architecture, Engineering, and Construction (AEC) sector. By leveraging Nvidia Jetson Orin, the company provides professionals with a lightweight, intelligent, and field-ready solution that accelerates digital twin creation and boosts productivity on construction sites.
