Earlier this month I attended one of the regular BRLSI philosophy talks. Andreas Wasmuht prepared an excellent introduction to the work of Martin Heidegger, specifically his work Being and Time (first published in 1927). This post is not specifically about that work, but it got me thinking again about how much more … Continue Reading ››
Delighted to be part of the programme @IETevents new webinar series on Responsible AI, exploring cutting-edge technology and future applications over seven sessions in November and December. Booking is open now: http://ow.ly/Kcdg30rWpTu
This year has been hectic, to put it mildly. The significantly increased workload of online teaching, combined with students' increased anxieties and the consequential increase in pastoral activities, combined to become an all consuming void that could not be filled --- however many hours one worked. On top of all that, the uncertainties of rule changes and the chronic lack of advance notice provided to HE by Government increased the number and frequency of staff and teaching related meetings. All in all a not-to-be-repeated experience. Work/Eat/Sleep/Repeat for 9 months solid.
Having said all that, there were some highlights, and I thought it worth mentioning them in this blog. Firstly, I was fortunate to have some really great PGT students this year. They threw themselves into the online teaching with exceptional engagement, and as a result many achieved some excellent coursework and exam results. They had to learn robotics software and hardware design completely remotely using laptop based simulation tools, accommodating Mac, PC, and sometimes even linux environments. I've written a blog about teaching with simulation, and it includes some video where students demonstrate their work. Here is the first of two videos:
From January though to May I ran our third year undergraduate group design and business project within the department of electronic & electrical engineering. Students work in groups, producing technical feasibility studies, designs and ultimately full business plans for real-world projects. Highlights included an online Dragon's Den event, with representatives of our industry advisory board acting as Dragons, and also an in depth assessed final group presentation and design exhibition, again all carried out online. I really enjoy working with well motivated students, and our third years handled the online delivery comfortably.
Remarkably, I've also squeezed in some time to continue participation on the IEEE P7001 standards working group, and we've just published a Frontiers in Robotics and AI journal article explaining our approach to producing the standard. We've been working on this for several years now, so to be close to a published standard and to have this paper out is particularly satisfying given the timing. Huge thanks must go to Prof Alan Winfield for leading this work.
Overall, it's been a long hard year, but nevertheless there have been some highlights, and some lasting success.
"I firmly believe that any man's finest hour, the greatest fulfilment of all that he holds dear, is that moment when he has worked his heart out in a good cause and lies exhausted on the field of battle - victorious." --- Vince Lombardi, 1913-1970
After many months of writing, proof reading and waiting for printing, I'm delighted that my book is now available. It's a very practical book, explaining why transparency is so important, followed by the details of experiments with various forms of transparency.
The book is based on my PhD research, but is expanded and extended, including an additional chapter to explain the importance of transparency within the wider context of accountability, responsibility and trust (ART). Here is a short extract from that new chapter:
Transparency as a Driver for Trust
.... I argue that although trust is complex, we can use system transparency to improve the quality of information available to users, which in turn helps to build trust. Further, organisational transparency drives both accountability and responsibility, which also bolster trust. Therefore transparency is an essential ingredient for informed trust. These relationships are illustrated in Figure 2.3.
System Transparency helps users better understand systems as they observe, interact or are otherwise affected by them. This informed understanding of system behaviour in turn helps users make appropriate use of systems.
System Transparency also supports accountability, by providing mechanisms that allow the system itself to offer some kind of ‘account’ for why it behaves as it does. It also provides mechanisms to facilitate traceability....
Organisational Transparency supports and encourages organisational accountability, and helps to develop a culture of responsibility....
Trust is built from a combination of informed system understanding, together with the knowledge that system providers are accountable and behave responsibly. Trust ultimately drives greater product acceptance and use....
In this book I also argue for the creation of transparency standards applicable to Autonomous Intelligent Systems (AIS) of all kinds. Standards will encourage transparency, and regulation may enforce it. This encourages business to develop cultures that embrace transparency in their processes and products.
Wortham, Robert H., Transparency for Robots and Autonomous Systems: Fundamentals, Technologies and Applications, The Institution of Engineering and Technology, 2020
ISBN-13: 978-1-78561-994-6 (eBook ISBN: 978-1-78561-995-3)
This year I've been developing a project with a final year undergraduate in Computer Science at Bath, integrating the Instinct Reactive Planner with the Robot Operating System (ROS). The project has gone really well, resulting a flexible and powerful framework to enable the integration of ROS based robots with Instinct. The target platform used for the project is the Husarion ROSbot (shown). There is also a short video. For further details please contact me.
This page has been created for the AISB Tutorial/Hackathon event, part of AISB2017 at the University of Bath, but may be used by anyone wanting to get practical hands-on experience with reactive planning for robot control.
This week I am primarily focused on preparing my talk for the TEDx event in Frome on 28th January. I am giving a talk about how humans interact with robots, and why it is important that we design our robots to be understandable and transparent, rather than mysterious and deceptive.
However, as I started … Continue Reading ››
Over the last year or so, one of my main research interests has been machine transparency. We've been using the R5 Robot in various 'robot encounter' experiments, both online using video recordings and in a live environment. We've shown that people do not naturally form good mental models of robots on first encounter. They … Continue Reading ››
I've just spent a couple of days at the ICAPS PlanRob workshop. About 35 of us met together at King's College in London, and my paper on the Instinct Planner was one of 23 accepted papers.