Biomimetica en neurale netwerken: als een vis in het water

Description

The lateral line system of fish provides them with a unique sense to detect obstacles and flow, even in dark or murky conditions. The anatomy and structure of the biological sensors in this lateral line system inspired us to create bio-mimetic sensors that can measure a range of fluid parameters using fibre optics. We use these sensors to measure flow speeds and temperature, and demonstrate nearby object localisation and identification using Neural Networks.
Neural Networks can be used for a range of tasks, e.g. modeling, prediction, and fault detection, and they are powerful tools if used correctly. However, it is important to be aware of the 'black box' nature of some neural networks and also the 'black holes' in the knowledge/data that you use to train the neural network. Neural networks, and AI in general, may make you more aware of patterns in your data and system, but a domain expert's view is, for the time being, necessary to interpret results and draw actionable conclusions.

Period	10-May-2019
Event title	KNW & Wateropleidingen voorjaarscongres | Kunstmatige intelligentie in de watersector: Verwatert on brein? Maakt kunstmatige intelligentie ons dommer of slimmer?
Event type	Conference
Location	Amersfoort, Netherlands
Degree of Recognition	National