“We are Working on Robots that Can Reproduce and Then they Can Die”

Interview with Jonathan Rossiter | Professor of Robotics

Jonathan Rossiter would like to fill the oceans with millions and millions of tiny devices so small that they could not be recognized by the human eye, so small that the only thing distinguishable about them would be their work: ending pollution. “The oceans are full of plastics, microplastics, algae and chemical and oil spills. Imagine creating robots that will feed on these components, that convert plastics, spills and algae into electricity,” explains this robotics professor at the University of Bristol (UK). Imagined and done. With that idea in mind, Rossiter founded the Soft Robotics Group at the Bristol Robotics Laboratory, where he deals with emerging fields such as soft robots and intelligent materials. For this Royal Society researcher, robots in the future will be on the sunny side of the story. “It’s our debt. We have to provide technology that will solve the problems that humans have created.”

Question. Your main field of investigation is the soft robots. Are they the next step of the robotic revolution?

Answer. Conventional robots are made of gears, motors, metal and plastics, so they don’t bend and twist and move. In contrast to them, soft robots are made out of compliant material and they behave much more like organic material, like biological organisms. They bend and they twist. That means you can use them in different applications in which you couldn’t use a conventional robot. For example, you can interact with the human body safely, because a soft robot is not going to hurt you.

In addition to robots that feed on pollution, you develop robots that can be eaten. What is their purpose?

Healthcare applications, for example. To get medicine or to perform a procedure inside the body, physicians typically inject or do an operation. With this soft robot edible devices you can eat them, they can go inside and through the digestive tract, from the head to the bottom of the body. So we have an opportunity then to deliver chemicals or to interact with the body physically, to restore some functionality. There are people, for example, who have diseases of their intestines and they have problems digesting food. So we can make, hopefully, soft edible robots which can go in there and they can help with the digestion of food.

“We can imagine the human and the robot societies merging together and, in future, we will have a one robot-human society”

What role can these robots have in the fight against diseases like cancer?

Robots as a whole can be almost any size: robots could be big as a building or could be small as a nanoparticule. So if we have robots which are very, very small, they can go into your bloodstream and they can move around your blood until they find cancer. There they can accumulate, you can put some energy into them and, then, they will kill the cancer. This is the work that a lot of people are doing across the world on nanorobotics. Nanorobots have the potential to treat some of these very serious diseases like cancer.

In addition to the field of medicine, what other applications do soft robots have?

A soft robot can be made out of materials which are biodegradable and biocompatible, because we are not limited by conventional technologies. So you can make soft robots that go out into the environment —and we working on this—, they will eat pollution and they will turn that into electricity. They will use the electricity to drive the soft robotics smart materials and that will encourage the robot to then get more pollution. So in the end, you have a robot that will eat the pollution and when it’s finished its life it will degrade to nothing and disappear.

That’s, for example, the main purpose of one of your creations: ROWBOT.

Its name is ROWBOT because it’s a robot that rows. It is a robot which shows the first demonstration of a single-celled stomach. It has got a body and it is energy-autonomous. That means that it can live in the environment —for example, in the sea or in a pond— and it will eat pollution or organic material from the environment and turn it into electricity. And there’s enough electricity for that robot to move and get more electricity. So, that robot is almost independent. For the moment, ROWBOT uses plastics and motors. The next step is to combine that with these soft materials to make it a truly independent robotic organism. And we them robotic organisms, because these robots are much more like real living organisms than conventional robots.

Some of the activities of living organisms are reproduction and death. Do you think robots will also be able to reproduce and die?

Robots are really interesting because at the moment they can’t reproduce. They are born when we make them in a factory and they die when they stop working and then we throw them away. But when we throw them away it takes a lot of energy to recycle them and to reuse those components. Nature, on the other hand, does that really efficiently: it creates organisms through the process of birth and these organisms live in the environment through a process of homeostasis. So, they keep going, they keep living on a day to day basis. Then, when they’ve run out of energy, they die. But, of course, throughout their life they reproduce… So yes, we are working on robots that can reproduce, live longer and die. Reproduction is coming.

And how would this reproduction be?

There’s some really interesting work that’s been done on a related field which is 3D printing. 3D printers, at the moment, are quite commonly available for us to buy and to take home. And then, at home you can print parts for your printer, using the printer that you’ve just bought. So you can start to think of an organism that exists and it could create or print a child organism. 3D printers and robots are becoming closer in that way.

How long are we going to have to wait to see these robots among us?

We are working on some of these technologies, for example, wearable assist devices which are soft and compliant. They are a little bit like smart trousers or power clothing, that will help people, old people, to move around and keep their independence for a longer time. We’re going to show demonstrations wit hin a few years, of course to move into the market and to actually have them in shops and clinics may take 10 years. But it could be sooner, it depends on the marketing drive.

“You can eat robots and they can move around your blood until they find cancer”

What about the robots we can eat? How long will we have to wait?

In the laboratory, I can make parts of the edible robots now: I can mix up some jelly-like material, and apply electricity to it and the jelly will move. That’s one of the fundamental components of an edible robot. We can put chemicals in there and it will go through the body. Of course, we’ll need to add more control and more sophistication into that. So, I would say it will probably take another five or six years to show the first demonstration of an edible robot that will do a really useful job inside the body.

Is this how you imagine the robots of the future?

We often think in terms of conventional robots as humanoids, as the things that are in factories or as the robots that hoover the floor in our houses. But actually I say that the robots of the future are going to be everywhere. They going to affect every part of your life: from your clothing to the insides of your body.

You often say that robots will be ubiquitous. What do you mean by that?

Ubiquitous is a really good word. It does mean absolutely everywhere and I believe that robot technologies will be everywhere. In fact, I think that in the future robotic technologies will be so ubiquitous that we may not even think of them as robots. The technologies will be part of our daily lives. We can imagine the human and the robot societies merging together into one robot-human society.

 “Soft robots are much more like real living organisms than conventional robots”

What is required to achieve this ubiquity of robots?

Robotic ubiquity requires us to work hard at the core technologies. Those are the technologies to make robots more functional, to make the intelligence of robotics more applied and more generalized, and to put both things together. Now, in terms of the technologies for robotics, we need to work up materials, we need to look at totally different ways of thinking about robots. So lots of working at laboratory will be needed.

If robots are to occupy all sectors of our society, should a tax be imposed on companies that replace humans with working robots?

We are at the start of a new revolution. It’s a little bit like the agricultural, the industrial and the technology revolutions of the past. They did disrupt and these new revolution, the robotic revolution will also disrupt our societies. I think it’s up to us in our societies to manage that and to catch any of the people who may lose their jobs through this. But I think sometimes this has been overstated and that we will ask robots to keep on developing. We will have to manage how they and humans work together.

Beatriz Guillén Torres for Ventana al Conocimiento