Next generation distributed processing platform for sensor signal and avionics data processing
Reference number | |
Coordinator | Saab AB - SAAB Aktiebolag Aeronautics |
Funding from Vinnova | SEK 4 200 000 |
Project duration | November 2017 - September 2022 |
Status | Completed |
Venture | National Aeronautical Research Program 7 |
Call | 2017-02942-en |
Important results from the project
A new type of hardware platform architecture is needed to meet future demands for more processing power, ability to reuse a single platform for different types of processing, and to enable evolution of scalable and future-proof avionics systems (compatible with disruptive technologies). This project took an important step towards solutions for future distributed hierarchical network-based processing platforms to meet current system design trends that moves towards re-configurable platforms for all types of processing capabilities, easily scaled to increase performance.
Expected long term effects
The project has created methodology for early assessment and evaluation of properties of a hierarchal network of computational nodes based on SoC technology. The project has developed a processing node as well as ADC and DAC chips as examples of building blocks in a distributed computing platform and demonstrated the applicability of the method with a use case with a safety-critical flight data function and a high-performance active electronically scanned array (AESA) radar. The results are promising, but more research is needed to reach a higher degree of maturity for use in industry.
Approach and implementation
The project is within the Avionics platform technology cluster that addresses future systems and the need for computing power, robustness, security and development cost. Two parallel project within the cluster have supported this project. The project has demonstrated advantages with new design methods for future aircraft design in two areas: avionics functions, demanding sensor functions of the AESA radar type, and taken a step towards machine learning that will be an important part of future autonomy functions with extensive data processing needs by defining a new use case.