Mohamed Hamed

Developing a new approach to detect malicious attacks in the CAN bus of an automobile using deep neural networks with a special focus on RNN where by learning to characterize and to model the normal behavior of in-vehicle network traffic, RNN is able detect deviations from this normal behavior.

Investigating different software hacking scenarios on vehicles and different approaches to handle the problem

- Developing an annotation tool for preparing supervised learning data.

- Developing a waveform generator for MRI emerging matrix coils. - Implementing a user friendly GUI in Tkniter.

Applying Kalman-Filter to predict the movement of an object using 3D-cartesian coordinates and neuron cells data

- Using a neural network model to predict the probability of a potential costumer for a video game ’wack-a-doo’. - Implementing cross validation metrics. - Adding flexibility to easily choose different inputs for the model.

Investigating different human-like robot mimic approaches

- Machine learning - Deep learning - Reinforcement learning - Statistical pattern recognition - Artificial Intelligence