Quad rotor vehicles are gaining prominence as Unmanned Aerial Vehicles (UAVs) owing to their simplicity in construction and ease of maintenance. They are being widely developed for applications relating to reconnaissance, security, mapping of terrains and buildings, etc. The control of the quad rotor is a complex problem. As a precursor to developing a model based design, the simulation of the mathematical model of the quad rotor is implemented. This will facilitate easier implementation of the model based design using Matlab Simulink©.

Machine learning models are susceptible to adversarial attacks which dramatically reduce their performance. Reliable defenses to these attacks are an unsolved challenge [11]. In this work we present a novel evasion attack: the 'Feature Importance Guided Attack' (FIGA) which generates adversarial evasion samples. FIGA is model agnostic, it assumes no prior knowledge of the defending model's learning algorithm, but does assume knowledge of the feature representation. FIGA leverages feature importance rankings; it perturbs the most important features of the input in the direction of the target class we wish to mimic. We demonstrate FIGA against eight phishing detection models. We keep the attack realistic by perturbing phishing website features that an adversary would have control over. Using FIGA we are able to cause a a reduction in the F1-score of a phishing detection model from 0.96 to 0.41 on average. Finally, we implement adversarial training as a defense against FIGA and show that while it is sometimes effective, it can be evaded by changing the parameters of FIGA.

For mission critical applications like Avionics, dependability is to avoid consequences of catastrophic results. Traditionally fault tolerance is implemented using hardware redundancy, the higher the redundancy, greater the cost and possibilities of more failures occurring. In this paper, an adaptive fault tolerant scheduling mechanism developed earlier with augmented performance capability and online fault recovery for a dual redundant system has been extended for an avionics mission system. An algorithm has been developed, simulated and evaluated on the practical case study vis -a-vis a traditional dual redundant system. This paper also roposes an extension of earlier scheme to schedule arrival of either/and critical and non -critical aperiodic tasks. The augmented scheme helps to achieve full functionality when no fault occurs, a fail safe mechanism for a single fault and performance metrics highlights the better computational performance. It elucidates that the use of this adaptive model leverages better in terms of enhanced performance and throughput compared to the existing dual redundant systems.

No abstract

No abstract