Researcher blogs

As neural networks pervade our daily life, we need to make sure they can be relied on. Muhammad Zakwan is using tools from control theory to achieve that.

When it comes to optimising energy efficiency on the railways, speed is of the essence – but not for the drivers! It’s in the computations that faster really is better.

In a world underpinned by shared infrastructure, how do we protect the most vulnerable? Sophie Hall explains why she is focused on finding the theoretical tools to build systems that won’t leave anyone out in the cold.

“Digital twins” aren’t just for industry. Plugging details such as soil conditions and weather into an app could soon help farmers to get tailor-made crop management recommendations, cutting emissions while increasing yield.  

Climate control is a major contributor to trains’ energy usage – which means it offers a major opportunity for savings. Ahmed Aboudonia is chasing that opportunity.

What does stability mean in control theory, and what do shapes have to do with it? Grab a pen, an empty toilet roll and a pair of scissors: Wouter Jongeneel has something to teach you.

This serious condition affects sufferers from babies to the elderly, and yet the treatment hasn’t changed in 70 years. We’re working on that.

Common traffic mitigation solutions have a problem: they’re intrinsically unfair. Charging for access shifts the burden of congestion to those less able to pay for it. Our researchers may have found a way around this issue.

In the changing energy landscape, flexibility is emerging as a new market – but how can it be properly traded when so much is uncertain? Julie Rousseau is creating a model to factor in the complexities of real life.

Even as the possibilities of Internet of Things applications multiply, growth is being held back by energy challenges. One strategy is energy harvesting, but that comes with fundamental limitations. Naomi Stricker is working on solutions.

While neural networks often achieve impressive performance, they sometimes fail to capture simple concepts that humans understand intuitively – a gap that may even put lives at risk. But can we transfer our human knowledge to neural networks to help them learn more meaningful solutions? NCCR Automation researcher Loris Di Natale has a plan to make sure automated machines don’t freeze us out.