SICSA AI Research Seminar: Prof Emma Hart “Towards the Autonomous Evolution of Robotics Ecosystems”

Map Unavailable

Date/Time
Date(s) - 10/12/2021
2:00 pm - 3:00 pm


You are invited to the first Christmas series talk from the AI theme. The Christmas talks will take place every year in December featuring a high-profile academic in Scotland. The first speaker of this series is Emma Hart, who will be talking about the Autonomous Evolution of  Robotics Ecosystems.

Abstract
Robot design is traditionally the domain of humans – engineers, physicists, and increasingly AI experts.    However,  if the robot in intended to operate in a completely unknown environment (for example planetary exploration) then it is very difficult for human designers to predict what kind of robot might be required. Evolutionary computing is a well-known technology that has been applied in various aspects of robotics for many years, for example to design controllers or body-plans. When coupled with advances in materials and printing technologies that allow rapid prototyping in hardware, it offers a potential solution to the issue raised above, while also addressing issues associated with the infamous reality-gap. However, it also brings new challenges. The additional constraints introduced by the need for example to manufacture robots autonomously,  to explore rich morphological search-spaces and develop novel forms of control require some re-thinking of “standard’ approaches in evolutionary computing and some novel approaches to engineering. In this talk I will discuss some of these challenges and propose and showcase some methods to address them, that have mainly been developed in the context of the collaborative EPSRC funded ARE project. I will also touch on some ethical issues associated with the notion of autonomous robot design.

Bio
Emma Hart is a Professor at Edinburgh Napier University. After gaining an undergraduate degree in Chemistry from the University of Oxford, she switched fields to Computer Science, gaining an MSc in Artificial Intelligence from the University of Edinburgh where she focused her research on nature-inspired methods for solving difficult optimisation problems. At ENU. she leads the Nature-Inspired Intelligent Systems group. Recognised as a world-expert in Evolutionary Computation, she has been invited to give keynotes at major international conferences, mostly recently at  IEEE CEC in New Zealand. She is Editor-in-Chief of international journal Evolutionary Computation (MIT Press), and an invited member of the UK Operations Research Society Research Panel.  In Scotland, she previously led the SICSA theme in Artificial Intelligence and was a member of the Steering Group that contributing to the development of Scotland’s recently published AI Strategy.  She is a member of the REF2021 panel for Computer Science.

Her work finds applications in a wide variety of domains and has attracted over £2.5 million in from EPSRC, Leverhulme, EU and UKRI Knowledge-Transfer Projects.

Much of it focuses on developing novel algorithms inspired by biological systems that are used in the context of optimisation and learning: this ranges from evolutionary methods to design algorithms that predict tree-damage in forests to minimise economic losses due to storm-damage, through software systems that optimise routing to minimise carbon-emissions to evolving robots that continuously adapt to their environment. A common theme throughout her research is to be able to develop software systems that autonomously improve over time:  learning from experience, adapting to new user goals and developing new behaviours in response to changes in the environment that they operate in order to remain fit for purpose. Her recent work in applying these ideas to robotics has attracted significant media attention, for example in national newspaper such as the Guardian and Telegraph; she will be presenting a talk on this topic at TEDWomen in December 2021.

Join Zoom Meeting:
https://zoom.us/j/94051082262?pwd=STgzRnM3TmdWWkNEeTFGcHpHMFpRZz09

Meeting ID: 940 5108 2262
Find your local number: https://zoom.us/u/acqBcLk7xi

This entry was posted in .