Syslogic adds RTK capability to AI computer for localization

Listen to this article

hero image of new syslogic NANO enclosure.

Syslogic adds RTK capability to AI computers. | Credit: Syslogic

Syslogic has combined the powerful NVIDIA Jetson Orin Nano with its reliable Rugged Computer series to create the Rugged Computer RPC RS A4NA. The unit is equipped with four GMSL2 inputs and is designed to be compatible with GMSL2 cameras and GMSL2 stereo cameras like the Sterolabs ZED-X and Intel RealSense D457 stereo depth camera. The embedded system targets computer vision applications, including autonomous navigation for farm or construction equipment.

In 2023, the company released its first version of rugged computers targeted at field robotics applications.

A purpose-built carrier board has been incorporated into this new generation of Rugged Computer to ensure it meets shock and vibration requirements. To safeguard the electronics from damage caused by moisture, humidity, or dust, the rugged PC is equipped with a housing that includes secure screw-on M12 connectors. The device has a waterproof rating of IP67 ensuring it can survive deployment into agricultural vehicles, excavators, dump trucks, and mining vehicles.

As artificial intelligence (AI) continues to advance, edge computing is becoming more powerful to support AI processing requirements in the field. Outdoor mobile robot applications require precision, and one of the key localization technologies is RTK (Real-Time Kinematic). While GPS can provide location data accurate up to +/- 2 meter (6 feet), RTK can provide sub-centimeter (sub-inch) accuracy. This is generally sufficient for ag and construction-related use cases.

RTK requires the deployment of an additional signal (local antenna or RTK service) that provides local offsets from the GPS satellite signals so that the individual machines can compensate for any GPS error, and remain localized at all times.

Syslogic leverages the Global Navigation Satellite Systems (GNSS) technology of u-blox to add RTK to the solution. The AI Rugged Computer RS A4NA can now be configured with up to two integrated u-blox receivers. This allows RTK functionality to be implemented without additional hardware. As a further highlight, Syslogic has integrated a heading function. This allows not only the position, but also the rotation angle of a vehicle or machine to be precisely determined.

The Rugged Computer uses u-blox ZED-F9P and -F9H interconnected modules. The F9P is an accurate GNSS receiver that receives positioning data from satellites. In combination with the ZED-F9H and u-blox’s own PointPerfect correction data service, it is possible to determine the angle of rotation and position with an accuracy of up to three centimeters.