An insect's tiny brain is an unlikely source of biological imitation, but researchers at the University of Groningen in the Netherlands and the University of Bielefeld in Germany believe it is an ideal system for the way robots move. Fruit flies (Drosophila melanogaster) have very simple but effective navigation skills. They use very little brainpower to fly quickly along invisible straight lines, and then make corresponding adjustments - flying tilted to the left or right - to avoid obstacles.
In an era where robotics technology is becoming increasingly advanced, a team has gone against the trend, seeking inspiration from the pinhead-sized brain of a small flying insect, and created a robot that can cleverly avoid collisions with very little effort and energy.
Scientists believe that fruit flies have very small brains and very limited computing resources available during flight. This biological model can be used in the "brain" of robots to achieve efficient, low-energy consumption and obstacle-avoiding movement.
"It's like being on a train," says Elisabetta Chicca, a physicist at the University of Groningen. "Nearby trees seem to move faster than distant houses. Insects use this information to infer the distance of things. What we learn from this is: If you don't have enough resources, you can use your behavior to simplify the problem."
In the Drosophila brain, the movement of surrounding objects is processed through optical neurons T4 and T5. With the help of neurobiologist Martin Egelhaaf of the University of Bielefeld, the team used algorithms to simulate this neural activity in the small robot's "brain", giving it the ability to process directional information to move efficiently and avoid collisions with any obstacles in its path.
"A lot of robotics isn't focused on efficiency," Chica said. "We humans tend to learn new tasks as we grow, and in robotics this is reflected in current trends in machine learning. But insects are able to fly immediately after birth. Efficient flight is hard-wired into their brains."
Ultimately, the tiny robot achieved one main goal - steering to the area with the least detected movement. Thorben Schoepe of the University of Groningen, who designed the hardware, conducted a series of tests on the wheeled robot and found that it could find a center point between objects and dexterously adjust its path to guide itself around obstacles—much like an insect in flight.
Chica said: "This model is very good, once you set it up, it can work in a variety of environments. That is the beauty of this result."
The research team believes that this is the first study of its kind to focus on obstacle avoidance, and it is a major step forward in the development of neuromorphic hardware for robots. In the future, such machines could be used to navigate cluttered terrain such as disaster scenes, with very low energy output, and could be equipped with different types of sensors depending on their purpose, such as radar to detect unstructured objects.
Chica said: "The robot we developed was inspired by insects. It has remarkable abilities to travel in dense terrain, avoid collisions, cross gaps and choose safe passages. These abilities are achieved through a neuromorphic network that guides the robot towards areas with less surface motion. Our system leverages knowledge about visual processing and obstacle avoidance in insects."
The research was published in the journal Nature Communications.