NWO Vici grants for Guido de Croon and Marios Kotsonis)

Neuromorphic Artificial Intelligence for autonomous robots

Prof.Dr. Guido de Croon, Full Professor Bio-inspired Micro Air Vehicles

Autonomous robots promise to transform our economy and society, thanks to the quick progress in artificial intelligence. However, state-of-the-art AI needs a lot of computational effort, which leads to extra robot weight, costs, and energy consumption. This makes robots with AI large, heavy and dangerous, and hence unsuitable for many applications. 

The aim of this project is to create an orders-of-magnitude faster and more energy efficient robotic AI. To reach this goal Guido de Croon and his team will tackle the daunting challenge of developing a neuromorphic AI that gives full autonomy to small robots. De Croon: “This intelligence will be based on neural networks that come closer to the real neurons in the brains of animals. The neurons have a more complex dynamics and asynchronously send pulses to communicate with each other - something that leads to much less energy consumption and much faster information processing.” To best demonstrate this technology’s efficiency, the AI will be modelled after insect intelligence and instilled on a resource-restricted drone. This will unlock innumerable new applications and even provides opportunities to learn about nature itself.

Together with partners from the industry De Croon and his team will also work on the application of this technology, focussing on small autonomous drones performing various tasks in greenhouses, such as monitoring crop and chasing pests. Having autonomous drones performing such tasks will save on resources and reduce the use of pesticides. Success in this challenging application will set the stage for using the developed efficient AI also for other types of robots and applications.

The project is funded with 1,5 million euros. Additionally, industrial partners provide in-kind support amounting to 100.000 euros. The industrial partners involved are: IMEC, PATS, Royal Brinkman, KeyGene, Wageningen University & Research, XOSight, and Opteran.

Running up that hill: reducing the aerodynamic drag of aircraft

Prof. Marios Kotsonis - Flow Control

Reducing the aerodynamic drag of aircraft can play a key role in reducing emissions. The transition of air flow from a (smooth) laminar state to a turbulent state over the aircraft wing is a major contributor of aerodynamic drag. Often, the transition from laminar to turbulent occurs due to certain modifications of the wing surface, such as panel joints or skin deformations. Professor of Flow Control Marios Kotsonis and his team have recently made a surprising and counterintuitive discovery: when air flow is running up a smooth small ‘hill’ on the aircraft wing, which they’ve since called the Delft Laminar Hump, it remains laminar for much longer, thus making the wing very efficient. This novel finding can revolutionise laminar wings, potentially saving millions of tons of fuel and harmful emissions every year.

However, how the Delft Laminar Hump, does this is currently unclear. Marios Kotsonis: “We simply lack a clear physical understanding of the interaction between laminar-turbulent transition and surface modifications.” In the project, called ‘Running up that hill’, the physics of transitional flows interacting with Hump-like wing surface modifications will be explored and described using a combination of advanced numerical, experimental and theoretical methods. The support and in-kind contributions from leading aerospace partners will help to demonstrate laminar flow at commercial flight conditions using Hump- like modifications.

The project is funded with 1.5 million euros for the coming 5 years. An additional 130k euros is contributed as in-kind funding by industrial partners such as KLM, DNW and Deharde.