University of Twente Student Theses
Explaining system behaviour in radar systems
Postema, J.T. (2019) Explaining system behaviour in radar systems.
|
PDF
1MB |
| Abstract: | Radar systems are large, complex, and technologically advanced machines that have a very long life-span. This inherently means that there are a lot of parts and subsystems that can break. Thales Nederland develops a whole range of radar systems, including the SMART-L MM, one of the world’s most advances radar systems, capable of detecting targets at a distance of up to 2.000 kilometers. In order to aid the maintenance and repair of the radar it is equipped with a wide range of sensors, which results in a total of 1.100 sensor signals. The sensor signal are currently processed by two programs, a Built-In Test system, which gives alarms based on a set of rules and an outlier detection algorithm. In the case of the anomaly detection algorithm the main shortcoming is the lack of explanations. Even though an outlier might be detected, there is still no explanation or label assigned to it. In order to resolve this shortcoming Thales wants to create a system which is capable of recognizing and grouping previously seen behaviour. This results in the following research questions: 1. To what extent can the system state be diagnosed automatically? (a) Which techniques are available to diagnose the outliers and which are most suitable given the case described in Section 1.1 and the available data? (b) How to assess the quality of the methods used to provide a diagnosis? (c) How do the methods selected in RQ 1.a stack up against each other based on the metric found in RQ 1.b and training and diagnostic speed? |
| Item Type: | Essay (Master) |
| Clients: | Thales, Hengelo, The Netherlands |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science MSc (60300) |
| Link to this item: | https://purl.utwente.nl/essays/79379 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page