392251 ISY Project: Closed loop evaluation of a spiking neural network for collision avoidance (Pj) (SoSe 2020)

Spiking neural networks convince due to their real time capability and energy efficiency. This makes spiking neural networks a perfect candidate for autonomous robotics. We have implemented a spiking neural network for collision avoidance onto a robotic platform called spikebot equipped with the Dynamic Vision Sensor (an event based camera) and the SpiNNaker board (neuromorphic hardware). Some preliminary tests have already been done with the spikebot. Your task includes the characterization and evaluation of the spikebot. To achieve that, you will record and characterize the spikebot's behaviour exposed to different scenarios in the Telewerkbank. You will further improve the spikebot's behaviour by fine tuning the neural network.
In case this would not find enough interest for a team project, this project proposal would be also offered (in reduced/modified form)
[x] as individual project
[x] as project for 2-3 students