Re-thinking robotics

"We take inspiration from nature to build robots which can cope with complex environments and adapt to new environmental conditions. Our aim is to fuse technology with biology."
In the future, new components for soft robotics will be researched together with the ITT and the Institute of Biomaterial Science. Photo: ITT

Ms Mazzolai, you and your team are working in the field called soft robotics. What does that mean – how are they different from normal robots?

Conventional robots have solid bodies and limbs and are made predominantly of hard materials, such as metals. As a rule, they are designed to operate in a structured, fixed environment, such as in an industrial factory. This means that for safety reasons, cooperation with humans is limited because the hard, fastmoving arms of the robot can definitely lead to injuries. Soft robots, on the other hand, are made of resilient, flexible plastics such as silicone - materials which recreate natural organic materials. Such robots have the ability to react more flexibly to their environment and their soft limbs allow them to work safely with humans. They cause no damage and could even be used inside the human body for medical purposes. Search and rescue robots are conceivable as further applications, as are small mobile machines for environmental monitoring.

What are the challenges in developing small robots? Which difficulties must be overcome?

The research field is still in its infancy and we still have some basic problems to solve. Control is a major challenge. We have to develop completely new mechanisms so that robots can move in a targeted manner. How can we combine different materials into one system and control them purposefully and reliably? And how can this system be combined with the right sensors? For this we need new materials that fulfil different functions and at the same time act as sensors and actuators. We take inspiration from nature to build robots which can cope with complex environments and adapt to new environmental conditions. Our aim is to fuse technology with biology. We have already achieved some initial successes: in 2012, for example, we presented a robot whose flexible arms were conceived based on octopus tentacles. Although we’re still in the midst of the pioneering phase, industry is slowly becoming aware of the field and is starting to provide some funding support.

You’ve recently started collaborating with the Helmholtz-Zentrum Hereon, with the group led by Andreas Lendlein from the Institute for Biomaterials Science in Teltow near Berlin. What is the goal of this collaboration?

The scientists at IIT have developed a plantoid, a robot that behaves and grows like a plant. It serves the investigation of biological and technological questions.
Photo: Duilio Farina/ llT

Both partners have different, complementary skills. In Pisa, we have the know-how to build a soft robot prototype. The Institute of Biomaterials Research, on the other hand, has expertise to develop the adaptive materials for this robot. In order to be as effective as possible in the field as engineers, we have to work as closely as possible with materials research - we absolutely need to be exposed to the different visions and perspectives provided by other disciplines. As engineers we can‘t design robots all alone in our corner. We need input from other disciplines, such as biology, materials research and chemisttry. This interdisciplinary collaboration is something new for us robotics experts. I find that very exciting because there are very few examples of such cooperation in our field.

Why did you decide to cooperate with the Helmholtz-Zentrum Hereon? What exactly do you seek from this new collaboration?

The challenge is to teach the new robots targeted and directed movements. To date, artificial tendons, for example, have been recreated by inserting wires or fine tubing controlled by air pressure into a silicone matrix. The shape-memory polymer actuators from Teltow promise significant improvements here. The polymer itself can function as an actuator which can be moved extremely accurately – and specifically responds to changing external conditions, such as temperature. The behaviour of the polymer is programmed into its design structure such that its movement range is predetermined to changing environmental conditions, be they increasing temperature or increasing humidity. Incidentally, plants that react to certain environmental stimuli with movement, for example, the Venus flytrap, work in a very similar way. By creating novel components with the new polymer actuators from Teltow and integrating them into our prototypes, we can significantly reduce the complexity of the systems.

Which projects will you work on together first? Which milestones are you aiming for?

Together with the Helmholtz-Zentrum Hereon, we want to introduce novel adaptive materials to robotics and use them to build our first prototypes. We are thinking of a robot with features from plants or a prototype that can hop and swim like a frog. Our long-term vision is that of robots, which can themselves reprogramme their actuators - much like a type of transformer, which independently changes its shape. Prototypes with selfhealing properties and machines which can collect and store energy and are therefore largely self-sufficient are also feasible. To consolidate the fundamental research underlying such systems, we have recently applied, for example, for a Helmholtz International Lab with the Teltow Institute, where we want to build such components for soft robots which can actuate autonomously. We are ready for action.

What could the most important applications be?

This is a close-up of a plantoid’s roots. Each root extremity is movable, equipped with chemical sensors, sensitive to gravity and touch.
Photo: IIT

We can conceive robots, which move skilfully and flexibly through unknown terrain and explore impassable areas. They could then use their sensors to search for toxins and pollutants and play important roles in environmental monitoring. Or they could search for survivors in disaster areas, such as where earthquakes or floods have occured. There are promising perspectives in medicine too. We are thinking along the lines of novel endoscopes for colonoscopy. Today's colonoscopy endoscopes can cause pain when their relatively hard tip hits the intestinal wall. With soft robotics, soft, flexible tips are totally feasible. In this way, the examination could be done much more gently and delicately.


Thank you very much for the interview.