Legal Considerations for Trusted Defence Autonomy

This is another short extract from the Thesis that I thought was particularly relevant given recent news coverage of the dangers of autonomy and AI, particularly in the field of “killer robots”. Legal Considerations in Design Trust If there is one key feature of the application of robotics and autonomy to the defence field that separates it from applications in commercial and civil fields, it is the potential direct impact on life and safety. ...

September 28, 2017 · Andrew Bolster

Human Factors related to Trusted Operation of Autonomous Systems

Preface It’s nearly a year to the day since I passed my Ph.D Viva (And since I last updated the blog…), so I thought it’d be fun to very-gently tidy up one of my appendices that’s a bit relevant to current stories about the end of the world and machines taking over and such. It’s a piece of work that I enjoyed researching and had originally had as a significant part of the main thesis, but it just didn’t fit in anywhere sensible, so it got stripped to it’s bare minimum and kicked to the end. ...

September 21, 2017 · Andrew Bolster

So long and thanks for all the fish

I’m leaving Northern Ireland, and I’m moving to Liverpool. Via Portsmouth. I have spoken with a few people about the situation I find myself in, and a few more people about my decision, but I want to get it all down somewhere. Where did this come from? The short answer is that a series of unexpected events mean that it makes sense to move my research. Recently, my PhD supervisor, Prof Alan Marshall, was offered a research Chairship at the University of Liverpool, which he gladly accepted. It’s a great move for him (Another Russell Group, one of the original ‘redbrick’ universities, etc etc), but that’s his life not mine so all it does is provide an option. (Disclosure: I also work with Alan on his spinout Wireless Security company, TOM Ltd) ...

June 20, 2013 · Andrew Bolster

LitReview: Communication in a behaviour-based approach to target detection and tracking with Autonomous Underwater Vehicles

Biblio Sorbi, Toni, Dio De Capua, Rossi TERA, Genova Grade 1 What is this work about? Collaborative target tracking behaviours What are the main findings of this work? No surprise; two AUVs with a comms link are better at tracking, and that tracking is easier when the targets are stupid/random. What gap in our understanding does this work fill? Highlights MLO tracking operations, and provides insight into globally aware control systems ...

June 12, 2012 · Andrew Bolster

LitReview: An Overview on Behaviour based methods for AUV Control

Biblio Carreras, Batlle, Ridao, and Roberts; An Overview on Behaviour based methods for AUV control, Girona, Spain What is this work about? A review and analysis of four AUV behaviour based reactionary control architectures, presenting four control architectures; Scheme, Subsumption, PDL, and Action Selection Dynamics, i.e. cooperative (schema, PDL) and competitive (Sub, ASD) What are the main findings of this work? Competitive models are often easier to design as only one behaviour is resident at a time, however this leads to ’jaunty’ and often sub-optimal results, as in this case where authors say the resultant pats for both competitive models are non-optimal ...

June 12, 2012 · Andrew Bolster

So, what is it you do again?

Update: I got asked to do a simplified version of this post for the University of Liverpool, it lives here (Backup) I’m technically in a third year of a PhD, and most of the time, when someone asks me what it is I’m actually doing, I fluff it and say something about “autonomous submarines” or “collaborative autonomy” or “Emergent properties of communities” or something similarly vague. In the spirit of setting the record straight in a less-academic way, I thought it’d be worth while to edit a presentation I recently made to the Association for the Advancement of Artificial Intelligence last month in Stanford and make it a little more digestible. ...

Andrew Bolster

Unfeeling Fire

This is an approximate transcript from my July 2018 talk at Digital DNA’s AI NI Community Panel on wether the use of AI in defence and surveillence was inherently evil Yes, It’s been sitting in my drafts folder for months because I completly forgot about it, sorrynotsorry Hello folks, I’m Andrew Bolster, most everyone calls me Bolster. And nobody calls me Doctor. I’m a Data Scientist at Alert Logic, a cyber security firm based Texas but with a research office in Weavers Court where we monitor, analyse and identify malicious and suspicious internet activity, protecting thousands of companies with advanced sequence and pattern matching sensors deployed across the world. ...

Andrew Bolster