Machine learning to detect Cyber intrusion and anomalies

The objective is to design an aircraft onboard security filter with an accuracy level of intrusion detection compatible with cyber protection objectives. The topic focuses on the use of machine learning techniques to enable the detection of cyber intrusion due to vulnerabilities on legacy ground-to-air digital exchanges to avionics. This domain is totally different to machine learning in operational avionics functions like guidance or control laws as the cybersecurity domain concerns defensive capacities and not model based behaviour. Please refer to the full topic descriptions document published in this call.