Credit: University of Illinois at Urbana-Campaign

Robots can be very strong, fast and enduring. However, unlike in animals, none of this strength comes from muscle, instead robots mainly rely on electrical motors and other hard and generally inflexible parts. But with all the advantages that conventional robot hardware can deliver, it still does not match the ability of muscle-powered animals to provide an accurate response to different physical environments. To address this downside of robotics, a group of researchers, led by Professor Rashid Bashir, at the University of Illinois at Urbana-Campaign developed tiny walking bio-robots powered by engineered muscle tissue.

The robot consists of a 6 mm long flexible 3D-printed backbone with two strains of muscle attached to each of its ends. The backbone has two little feet and is used both for walking and sustaining the structure. The important thing about the robot is that the muscle tissue used in it is the skeletal muscle, the one humans use to move around, which means that it can be easily turned on by administering electric impulses. Furthermore, by adjusting the frequency of the impulses, the robots’ speed can be modified.

The use of skeletal muscle allows for a better control over the robots’ movements. This significantly differs from the previous study conducted by the same group, where the researchers used heart tissue, which contracts non-stop and with a constant rate.

This technology is an important step on the way to integrating biological tissue in machines, which in some cases can be priceless. For example, muscle-powered robots are perfect for medical applications inside the body: the tissue is a perfect biodegradable material and such robots could run in a nutrient rich fluid without any additional power source. In addition, the use of muscle-powered limbs in biomimetic machine design would open hundreds of new possibilities, especially in the field of soft robotics. Imagine how much more lifelike a robotic starfish or octopus could be if powered by muscle tissue!

With the concept of a muscle-powered robot tested, the researchers are now preparing for the next step: the group envisions equipping their robots with light or chemically sensitive neurons for controlling direction of robot’s movement as well as testing new designs of the backbone to enable a wider range of motions.

The results of the study can found in PNAS in the article called “Three-dimensionally printed biological machines powered by skeletal muscle.”

Chimps, birds and bees are just a few of the many animals that communicate with each other when searching for food. Since everyone’s got to eat, communication during foraging is essential among social animals. In order to study how different types of communication strategies might evolve, the University of Lausanne (UNIL) and the École polytechnique fédérale de Lausanne (EPFL) teamed up to conduct a joint research project.

¨Animal kingdom communication systems are very complex¨ points out Laurent Keller (UNIL), one of the project’s lead researchers along with Steffen Wischmann and Dario Floreano (EPFL). In fact, a lot of these systems are so complex, researchers must devise models in order to understand them more thoroughly. Using 100 groups of 20 robots, researchers studied the evolution of communication strategies among the robots throughout 1000 generations.

Each robot was able to communicate the presence of a virtual food source through the use of a light that could be shone in various colours. They were also equipped with wheels and cameras giving them mobility and awareness of their environment. What was guiding the robots’ behaviour? Genes of course! Just as in nature, these were made to evolve through mutation and selection through out the course of successive generations.

Results of this research stress that random factors seem to be an important part of evolutionary processes; distinct communication systems evolved among groups of robots operating in identical environments. The study also showed that populations that have developed a fairly effective communication strategy are not likely to drop it for the next strategy that comes along because to do so would require significant changes in the way information would be transmitted and received. This is similar to the reason why human languages are unlikely to change rapidly through generations.

You can read more about this research recently published in the PNAS  (Proceedings of the National Academy of Sciences of the United States) magazine and if you’re interested in other European projects involving the development and use of social robots, check out the Robot Companions for Citizens Flagship Initiative which Dario Floreano is also involved in. You may also be interested in the European project EFAA (Experimental Functional Android Assistant) which is  coordinated by the SPECS lab at Pompeu Fabra University.