Nvidia Jetson Orin combines CPU, GPU, and deep learning accelerators to power real-time SLAM and object detection
NavLive has chosen the Nvidia Jetson Orin system-on-module as the core of its latest handheld construction scanner, enabling ‘advanced real-time processing’ in a lightweight, portable device.
The NavLive scanner integrates five sensors, including three cameras, LiDAR, and an IMU, all streaming data simultaneously to support simultaneous localisation and mapping (SLAM) and computer vision tasks on site.
Previous generations required separate microcontrollers for each sensor, but with the Jetson which features integrated CPU, GPU, and dedicated deep learning accelerators, all processing is consolidated into a single platform.
“By moving to Jetson Orin NX, we reduced weight and power requirements while increasing capability,” said Dr. David Wisth, CTO and Co-founder at NavLive. “Customers immediately notice that the device is easier to hold, yet it delivers much more in terms of performance.”
The scanner uses the Jetson to power real-time SLAM and object detection of building features such as windows and doors. The hardware also leaves room for scaling, including anonymisation features like face detection to support compliance requirements.
