Natural materials robots - cleaning oil spills and exploring planets
Robotics continues to make headway towards stronger, faster and more naturalistic contraptions. But what role does natural materials robotics have to offer? Khai Trung Le investigates its esoteric potential.
Oil spills are among the most high-profile industrial and environmental disasters, and there are a number of conventional measures available to combat their harmful effects. Booms – floating physical barriers made from a variety of materials including polymers and metals deployed using mooring systems, slow or contain the spread of surface oil. Ships can remove oil from the sea surface by towing a boom, otherwise known as skimmers or vessels of opportunity. Finally, in-situ burning can be employed when the oil is fresh and the weather is calm to prevent the oil reaching the coast.
However, Professor Jonathan Rossiter, Professor of Robotics at the University of Bristol, UK, thinks natural materials robotics could be employed to ambitious results. ‘Imagine cleaning the oil slick not with 10 robots that you bring back when they die, but you put a million, or a billion, out there, thrown from an aircraft, that do their job and degrade into nothing.’
While much of conventional robotics preoccupies itself with emulating humanity, natural materials robotics has moved in other directions, and proponents position the specialist sector as fulfilling more unusual purposes including oil spill cleanup, non-disruptive space exploration and deliverers of in vivo healthcare. Lofty goals, but how close is the reality?
The belly and its members
Rossiter’s team at the Bristol Robotics Laboratory, a collaboration between the Universities of Bristol and the West of England, reached several milestones during the 2012 project A robot that decomposes. These include creating biodegradable actuators – soft polymers coated with compliant electrodes activated by placing a voltage across the electrodes – and an energy source in the form of microbial fuel cells capable of converting waste matter, such as urine, into energy. ‘The goal is to have a robot that, in its total energy and environmental cost, is less than the benefit it does to the environment,’ said Rossiter.
Specialising in soft robotics, moving away from electro-mechanical motors to develop components akin to biological organisms, Rossiter believes that the initial value in natural materials robotics comes from the capabiliy to perform unusual tasks that would be challenging for conventional robotics, leaning on the example of an oil spill cleanup operation.
‘You will make it, put it out into the environment and if it includes unpleasant chemicals or materials – typically what happens when you’ve got batteries or interesting motors – at the end of its product life, you have to collect it. Because of monitoring and retrieval, you’re restricted to the number of robots you can set out. Worse, if you don’t bring them back, it could be that the impact on the environment is greater than the oil spill itself.’
Despite the milestones reached, the Bristol Robotics Laboratory team failed to meet the two primary objectives of its 2012 project – building a robot entirely out of biodegradable materials, and testing functionality in the natural environment. ‘Robotic organisms require three things – an energy source, essentially a stomach, and an energy sink, the body,’ said Rossiter, referring to the microbial fuel cells and actuators. ‘Then they need a brain or nervous system to control them. But we haven’t quite solved the problem of the biodegradable brain. We currently use a low-power silicon microprocessor, which means that at the end of the life cycle, you have a tiny microprocessor that falls to the bottom of the sea.’
Paper wasps and fungus
Lynn Rothschild, an evolutionary biologist at the NASA Ames Research Center, USA, recalls how the temporary loss of a drone by a colleague at an Alaskan coastline prompted the idea of constructing the drone from natural materials to mitigate potential problems. ‘Not just because you’re harming the environment, nor because you’re losing this property, but also because of the various instruments that you might not want other people to pick up. That got me thinking, what if it was biodegradable? If it just dissolved away?’ Rothschild suggested the concept of the biodegradable drone to her joint Stanford and Brown Universities team for the 2014 International Genetically Engineered Machine (iGEM) competition.
Starting with the shell, the team worked with New York-based biomaterials company Ecovative to develop a drone chassis made from fungal mycelium. A mould is filled with mycelia spores and fed until reaching the desired density. Described by Rothschild as ‘a little bit like cardboard – somewhat porous, not waterproof and not very tough’, fungal mycelium was selected due to its inert status, and because it would not contaminate or introduce living cells into an environment upon decomposing.
In overcoming the unsuitable characteristics of mycelium, the chassis was coated in lightweight and biodegradable cellulose acetate. The vast majority of cellulose-based biomaterials are not hydrophobic and in order to find one that was, the team looked to the European paper wasp (polistes dominula), which makes its nests by collecting cellulose from nearby plants and mixing it with protein-rich saliva. The cement created is hydrophobic, and the team isolated the waterproofing protein from the saliva to create the solution with which to shield the cellulose acetate. The iGEM team website notes, ‘Never before have the proteins in wasp saliva been identified or applied as functional biomaterials.’
Biosensors were investigated during the competition for various uses, including determining the level of oxygen or toxins in the atmosphere, but presented a contamination risk. Rothschild said, ‘Up until now, everything on the drone is a bioproduct, inert, no longer alive and not dangerous. But for the biosensor, we wanted something alive. The George Church Lab, Harvard helped us recode a particular organism, an E-coli bacterium with an altered codon. If the cells escaped the drone on impact, they would not be able to cross-speak to any other organism outside of the Church Lab.’ This prevents the spread of genetically-modified organisms being introduced into an alien environment – especially important when surveying interstellar habitats that are yet to be fully understood.
Despite work on the different components of the biodrone, a full unit culminating their work was not completed during the iGEM competition. A demonstration drone comprising of the mycelium shell with conventional non-biodegradable motors and propellers was constructed. However, Rothschild was keen to stress that subsequent work by iGEM team members and beyond could be repurposed for a finalised drone, including bioplastics for the propellers, printing circuits using silver nanoparticle ink and degradable cameras developed by Sony.
Origami for the body
Looking internally, Dr Shuhei Miyashita, who joined the newly created University of York Microbotics Group, UK, has designed a miniature self-folding robot that he hopes will one day perform micro-operations within the body. The York PhD project page states, ‘Imagine swallowing a pill of microparts which spontaneously self-assemble into a microbot inside the human body, subsequently navigating and repairing.’
Miyashita began his approach with a flatsheet indented to assemble to the desired shape when exposed to magnetic fields emitted by a small electromagnetic coil system at the centre of the sheet. The influx of magnetic fields also enables movement from the robot, tensing and flexing according to field variance. Three models have been developed, made from a mixture of PVC, polystyrene and aluminium-coated polyester. Two of the models are water and acetone-degradable.
Miyashita stated that creating a truly biodegradable flatsheet would not be a challenge, ‘We can replace the PVC with a polyolefin-based material called biolefin. The body could be any biodegradable material, even a brand name food-based structure. It’s not a big problem,’ although he declined to elaborate, inferring that further information on the materials approach will be available in a new paper scheduled for publication in May/June 2016.
The magnet remains a more significant material obstacle. Miyashita proposed replacing the magnet with another actuation method or making the magnet harmless to the human body, ‘encapsulating the magnet so it can be taken out from the intestines’ or safely passed through.
At around 20mm, the current origami robot is too large to perform in vivo tasks, but Miyashita does not envision the unit being any smaller ‘than a few hundred micrometres’. Current models are able to assemble within 60 seconds, and are capable of pre-programmed walking, swimming, carrying loads twice the weight of the robot, as well as block delivery, which Miyashita equates to being directed to a wound and undertaking drug delivery.
With natural materials robotics still in its infancy, it is little surprise that all three projects have yet to deliver on their potential. But they continue in various forms. Rothschild notes, ‘I applied to NASA headquarters for additional funding. This is something that has caught everyone’s attention – even the Secretary of Defense [Chuck Hagel in 2014] asked for my slides!’ The biodrone will be making further appearances in 2016, having resolved a number of issues that persisted during the 2014 iGEM competition.
Miyashita believes the field of in vivo operations is an ideal fit for natural materials robotics, deeming conventional automated robotics unsuitable for the task, and he hopes to ‘develop robotics away from traditional motors and pincers. The next step is to focus on more intelligence and kinematics – greater autonomous behaviour – the material itself, and downscaling.’ Ex vivo tests of the self-folding origami robot are expected to begin within five years.
Rossiter continues to explore soft robotics and analyse the expected variables less prolific in conventional robotics. ‘A structure that is designed to degrade and disappear means its ‘death’ is gradual. Performance might start off at 100% capabilities, but like us, as it degrades, it moves less and less.
‘That’s an interesting model to look at in predicting the behaviour of these models in environment, this property of degradation. Some might degrade at different rates. What about activity within a population of older or newer models – how can you optimise the spread? How many do you need to achieve your mission? The very notion of degradation impacts on the way we plan to use natural materials robotics.’
The biological connection
Rossiter and Rothschild conceded that conventional robotics are yet to be beaten by natural materials robotics on traditional scales – strength, toughness, speed – but both believe that the gap in capabilities can be closed within the next decade due to the advantages nature provides.
Rothschild said, ‘Our bodies last for a hundred years. I’m looking at trees outside my window that last thousands. Your cellphone has a lifespan of two years. You really think you’ll have a robot that is good for a hundred? Biology has an enormous number of tricks up its sleeve for both making structural materials as well as repairing and rebuilding. I do believe there is easily a competition. Every single cell on the planet is a better nanotechnologist than anyone in this room. People in the engineering space can look at biology for enormous advantages.’
Crediting a recent drive in focus on biomaterials in education, Rothschild has observed renewed enthusiasm within the wider engineering space. According to Rossiter, the drive for biomaterials is equally mirrored in the fashion and consumer device industry and, in conjunction with advances from conventional automation, could amount to a change in the way we use robotics.
‘The notion of the throwaway robot means you can change the way things are used. I can look out of the window and see all the chewing gum on the pavement. What if you fly a drone across the city and drop millions of little robots that crawl around and eat the gum? The next rainfall, they all get washed away. Pie-in-the-sky stuff, but that’s the notion of this no longer being a niche technology but something more ubiquitous.’