Imperial College London > Talks@ee.imperial > Yiannis Demiris's list > Navigation, Neuroscience and Neural Networks: A Quest to Understand Intelligence and Build Better Technology for Robots and Autonomous Vehicles

Navigation, Neuroscience and Neural Networks: A Quest to Understand Intelligence and Build Better Technology for Robots and Autonomous Vehicles

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Yiannis Demiris.

The goal of my research is to understand the fundamental nature of intelligence, so that we can both shed light on how the brain functions, and create intelligent, autonomous systems that transform society. My group models and mimics the neural mechanisms and behaviours underlying those most tangible aspects of intelligence such as navigation, perception and recognition. Our inspiration? The amazing natural world around us, including rats, insects, primates and even humans. Our tools? Robots, algorithms, innovative sensing modalities, computational neuroscience and artificial neural networks of all types including deep nets, continuous attractors and even spiking networks. This talk will cover the highlights and lowlights of that journey, the advances made, the insights gained, and touch on where we are headed now. I’ll summarize a roboticist’s perspective on the neuroscience underlying one of the most well understood aspects of intelligence – spatial intelligence and perception – and describe how we have adapted this knowledge to create landmark robotics advances including RatSLAM’s mapping of an entire suburb using only a low cost webcam and persistent long term robot navigation autonomy trials. I’ll describe our extensive work in creating robotic perception and navigation algorithms, both algorithmic (SeqSLAM, deep learning-based) and bio-inspired, that function under even the most challenging of environmental conditions. I’ll also give an example of how research can transcend traditional disciplinary boundaries, showing how robotics, neuroscience and deep learning can come together to create new techniques for data compression. Finally I’ll highlight how we are translating fundamental transdisciplinary science into industrial applications, working with major multinational companies in domains including autonomous vehicles and automated hazard detection. Useful Links • Google scholar: http://scholar.google.com/citations?user=TDSmCKgAAAAJ • Linkedin: https://au.linkedin.com/in/michaeljmilford • Math Thrills Startup: http://store.maththrills.com/

Brief Bio: Professor Michael Milford conducts interdisciplinary research at the boundary between robotics, neuroscience, machine learning and computer vision and is a multi-award winning educational entrepreneur. His research models the neural mechanisms in the brain underlying tasks like navigation and perception in order to develop new robotics and computer vision-related technologies, with a particular emphasis on challenging application domains where current techniques fail such as all-weather, anytime positioning for autonomous vehicles. He currently holds the position of Full Professor at the Queensland University of Technology, as well as Australian Research Council Future Fellow, Microsoft Research Faculty Fellow and Chief Investigator at the Australian Centre for Robotic Vision.

This talk is part of the Yiannis Demiris's list series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Changes to Talks@imperial | Privacy and Publicity