23 June 2020

Can We Engineer Artificial Life for the Coronavirus?

Opinion piece by ACSE’s Dr John Oyekan 
Autonomous systems particularly robots where designed to ease humans lives, to do things that are repetitive, boring and dangerous.

Coronavirus Image

Before 2020, dangerous environments were restricted to places with extreme radiation, extreme pressure, extreme temperatures and other environments that were outside the liveable parameters of humans. Examples of such environments included Nuclear, Space, Deep Sea to mention a few. These environments were few and far between and not in the exact domain of everyday human living.

But that has changed in recent months. Now we are waking up to a new world where there is danger lurking everywhere - on the surfaces we touch, air we breathe, on trains, in supermarkets etc. Worse still, the danger is invisible, caused by a tiny organism that cannot be seen without extreme magnification and yet causes death or severe respiratory distress for the unfortunate who come into contact with it. This brings to mind the proverbial story of how a tiny spider was able to overpower a mighty elephant who thought he was the king of the forest. 

When I conducted my PhD, at the University of Essex, on the use of swarm robotics to visualise invisible hazardous substances, my imagination did not stretch to the point that robots could be applied to providing a visual representation of the concentration of viruses in the environment. I was focusing mainly on non-biological invisible airborne pollution or hazardous substances such as sarin gas or nerve gas [1]. 

In developing my approach, I used a bio-inspired approach that made use of the foraging behaviour of bacterium to find pollution sources in the environment and then through the use of the social behaviour of bacteria, provided a visual density distribution of the pollutant in the environment.

Thinking of this further, I’m reminded of how there are many examples in which natural organisms are being used or applied to the environment in order to control the spread of another undesired organism. Such bio control mechanisms are quite effective if engineered appropriately. Take for example, the recent use of ducks to control locust invasion in Pakistan [2]. 

In my case, I was designing a new form of intelligent artificial life and applying it to a problem of finding hazards in the immediate environment. The design of Intelligent artificial life is something that the academic community have always been interested in since the late 1980s [3]. 

Termed Animats (a contraction of animals and materials) [4], the development of intelligent artificial life starts with an understanding of how real biological agents solve a problem in their immediate environment. This is followed by extracting operational rules and then designing algorithms as well as physical robotic agents that mimic those rules. They are then deployed into an environment where they can learn and evolve into higher constructs towards doing their jobs more effectively. 

By following this animat engineering approach, it is even possible to engineer hybrid or chimera organisms as was the case when I developed my bacterium flocking algorithm during my PhD.

Rodney Brooks is one the people that thinks along this line and he introduced this in  his seminal work titled "Intelligence without representation". Rodney Brooks proposed a behaviour based architecture in which reactive behaviours make up the architecture. Each behaviour was an ingredient or gene of the final organism and by adding them together in a “cocktail” and with the right approach, an artificial life engineer could come up with new sophisticated artificial organisms capable of operating intelligently in the environments in which they are deployed. 

As we deal with the coronavirus situation, perhaps such engineered agents might be needed more than ever. The current crisis has definitely highlighted vulnerabilities in the way we conduct the construction of artefacts, manufacturing of goods as well as logistics. The need for us to stay at home to protect ourselves from the coronavirus is costing governments billions.

It might be beneficial to have intelligent agents that could help us highlight viral hot stops based on their programmed behaviour and sanitise those areas. Autonomous mobile agents such as UAVs and UGVs could be deployed to aid us in logistics delivery of crucial medical supplies or keep the flow of goods and food going to support and sustain us while we deal with the health situation rapidly unfolding in our communities.  

Collaborative robots, an area that is actively being researched by the academic community, could be useful as well in such situations [5][6]. Collaborative robots are designed to work alongside humans supporting them in areas that require massive strength and repetition outside the natural ergonomic limits of humans. They are also meant to learn from humans as they work together on various tasks. 

If perhaps they had been deployed earlier into manufacturing systems, they could have continued various manufacturing tasks in our absence using lessons learnt and training sequences. Perhaps now we are staying at home to keep safe, we could have been able to remotely log into them and control them remotely via telepresence [7].

The coronavirus pandemic is perhaps not the only pandemic we will see in our lifetime. The world health organisation has been warning the world that we were on the brink of another pandemic since the early 2000s costing $3 trillion… and cause misery, economic decline, and societal disruptions on a global scale“ [8]. Is the current Coronavirus pandemic the one? Maybe or maybe not. No one knows for sure. Nevertheless, it was not a question of if, but when. 

With increasing world population, we are bound to encroach into forest areas we have not before and with this comes the possibility of opening up the human race to various flu type diseases. We should be prepared for the next time a pandemic hits with more government investment in autonomous systems and the development of artificial intelligent agents. 

Even though artificial intelligent life research is still in its early days, perhaps the country that solves this first might come out on top when the next Pandemic hits. Such a country will be able to ensure continuous production as well as productivity of workers when such disaster strikes again. 

Nevertheless, I’m reminded of various dystopian sci-fi novels I have read in the past. We are only flesh and blood and vulnerable to various calamities that our immediate environment throws at us. However, as I mentioned at the beginning of this article, robots have the capacity to work in extreme environments. Perhaps, after we are long gone, these artificial intelligent agents could be relics and reminders of our once great civilisation. But before that, could we be driven underground or into our houses while the autonomous agents operate outside and above us trying to keep the artefacts of our great civilisation intact for the day when we are able to come out once again. 

References:

[1] Oyekan, J., Gu, D. and Hu, H., 2013. Visual imaging of invisible hazardous substances using bacterial inspiration. IEEE Transactions on Systems, Man, and Cybernetics: Systems43(5), pp.1105-1115.

[2] The Telegraph, Chinese 100,000-strong 'duck army' bound for Pakistan to fight locust plague, https://www.telegraph.co.uk/news/2020/02/27/china-sending-100000-strong-duck-army-pakistan-fight-locust/

[3]Wikipedia, Artificial Life, https://en.wikipedia.org/wiki/Artificial_life, Last accessed 29/03/20. 

[4] Wikipedia, Animat, https://en.wikipedia.org/wiki/Animat, Last accessed 29/03/20.

[5] Oyekan, J.O., Hutabarat, W., Tiwari, A., Grech, R., Aung, M.H., Mariani, M.P., López-Dávalos, L., Ricaud, T., Singh, S. and Dupuis, C., 2019. The effectiveness of virtual environments in developing collaborative strategies between industrial robots and humans. Robotics and Computer-Integrated Manufacturing55, pp.41-54.

[6] The rise of the cobots, robots business review, https://www.roboticsbusinessreview.com/wp-content/uploads/2018/04/RBR_Cobots_WP-Final.pdf

[7] Oyekan, J., Prabhu, V., Tiwari, A., Baskaran, V., Burgess, M. and Mcnally, R., 2017. Remote real-time collaboration through synchronous exchange of digitised human–workpiece interactions. Future Generation Computer Systems67, pp.83-93.

[8] Waiting for the big one: A new flu pandemic is a matter of time, 2008, https://www.newscientist.com/article/2156921-waiting-for-the-big-one-a-new-flu-pandemic-is-a-matter-of-time/

Undergraduate scholarships

Our department offers financial assistance in the form of scholarships and bursaries, in addition to the scholarships offered by the University of Sheffield.

A world top-100 university

We're a world top-100 university renowned for the excellence, impact and distinctiveness of our research-led learning and teaching.