Challenge being tackled: The SentiSync emerged from challenge 5, focused on the synchronisation of multi-modal sensor data, proposed by ScoutDI. The target of this challenge was to develop and demonstrate a hardware-based synchronisation and fusion platform enabling accurate alignment of multi-sensor data streams for improved autonomy, navigation, and situational awareness.
Description of the solution: SentiSystems has created a reconfigurable hardware-based sensor timing system designed to simplify system integration to time-synchronising sensor data, without compromising accuracy. By using dedicated hardware features such as input capture to record the time of validity of measurement (TOV) from sensors, temporal errors are minimised, ensuring precise synchronisation between all connected devices.
To realise this design, the company developed SentiBoard, a core hardware component that references measurements to a 100 MHz clock, achieving 10 ns temporal resolution. The board supports a wide range of sensors—including IMUs, Inertial Navigation Systems, magnetometers, Doppler Velocity Logs (DVLs), and GNSS receivers – connected via standard interfaces such as SPI, UART, RS232, and RS422. It also supports high-bandwidth timing synchronisation for modern LiDARs and cameras (electro-optical, hyperspectral, and infrared) using digital input-capture and trigger mechanisms.
An onboard configuration system allows new sensors and equipment to be integrated without modifying firmware or software, ensuring scalability and ease of maintenance. SentiSystems is currently developing the next generation, SentiHUB, a compact (8 cm × 8 cm) version that combines hardware timestamping and synchronisation functionality with a network interface supporting IEEE1588v2 PTP and IEEE 802.1AS TSN. These standards enable collaborative latency control between processing units, facilitating synchronised operation not only within a single platform but also across multiple stationary and moving systems.
The system’s core synchronisation and sensor data parsing functionality are hosted on a System-on-Chip (SoC), making integration with new vehicle platforms straightforward through custom carrier boards. Two such carrier boards have already been produced and tested. SentiBoard 2 ensures sub-microsecond accuracy across distributed processing nodes, providing a unified time reference critical for high-performance data fusion and control.
To complement the hardware, SentiSystems developed SentiUtils, a software library for managing, filtering, and preprocessing synchronised sensor data streams. SentiUtils attaches precise TOVs to all sensor data, supports temporal interpolation and extrapolation between sensor readings, and reduces data traffic by performing local filtering before forwarding data to fusion algorithms or external systems. The software includes native interfaces to ROS and ROS2 (Robot Operating System), making deployment easy in research and industrial environments.
By combining hardware-level synchronisation with software-level data management, SentiSync provides an end-to-end solution for precise, reliable, and scalable sensor fusion. This makes it ideal for applications in autonomous navigation, mapping, monitoring, and inspection—particularly in maritime and industrial environments where accuracy, timing, and robustness are essential.
About the provider: SentiSystems AS develops cutting-edge synchronisation and sensor fusion technology to enhance autonomy, navigation, and situational awareness. With over a decade of combined research and development, the company’s core innovation, the SentiBoard, enables high-precision timing and unified data handling across sensor platforms. SentiSystems provides scalable solutions that empower robotics, drones, and marine systems to operate with improved precision, safety, and efficiency.
Check their LinkedIn here.
The Team behind the scenes:
Frederik Leira – Project Lead / Engineer
Stian Lundli – Systems Engineer
Sigurd Albrektsen – Embedded Development Engineer
Torleiv Bryne – Advisor
Erling Jellum – Embedded Developer
