Abstract

Control Engineering and the principle of negative feedback have had a significant impact and enabled some of the most important advances in the engineering of man-made systems – from aerospace to energy. The future holds significant promise in the application of control engineering methods and tools in other domains which suffer from unprecedented levels of uncertainty and data availability including biology and automation. New feedback control methods and tools are needed to address these new challenges.

This talk will first focus on the role feedback in biology in natural systems but also the advantages it can offer in synthetic systems. I will discuss the implementation of feedback regulation inside cells and present different feedback approaches/architectures to achieve this. I will then talk about automation platforms and new ways of closing the loop, using input-output data.

Data-based approaches sometimes use a controller in the form of a Neural Network. These have become very popular, but some neural networks are very fragile and off- the-shelf methods may not be ideal in many situations. I will talk about ways to verify properties of Neural Networks and Neural Feedback Loops, using sum of squares approaches taking advantage of sparsity. The talk will close with future research directions at the crossroads of control and applications in biology and autonomy.

Both seminars are hosted by Malcolm Smith

We strongly encourage you to attend the seminars in person. You can also connect via Zoom:

https://newnham.zoom.us/j/92544958528?pwd=YS9PcGRnbXBOcStBdStNb3E0SHN1UT09 Meeting ID: 925 4495 8528 Passcode: 203369