With a brain the size of a pinhead, insects perform incredible navigational feats. They can avoid obstacles and move through small openings despite their limited brain power. Physicist Elisabetta Chicca of the University of Groningen showcases the potential of understanding an insect’s brain in the development of energy-efficient computing. Her recent achievement is a robot that mimics insect behavior.
Utilizing the visual input received through their eyes to guide their movements is not an easy task for insects. A crucial factor in this process is the perceived motion of objects as the insect moves. Chicca explains that insects use this information to estimate the distance of objects. This method works well for moving in a straight line, but real-life situations are rarely that simple.
Moving in curves presents a complex problem for insects. To overcome this challenge with their limited brain capacity, insects adapt their behavior by flying in a straight line, making a turn, and then continuing in a straight line. Chicca explains, “This teaches us that if resources are scarce, simplifying the problem through behavior is a viable solution.”
Seeking to understand the neural mechanism behind insect behavior, PhD student Thorben Schoepe developed a model of insect neuronal activity and a small robot that utilizes this model for navigation. Under Chicca’s guidance and in collaboration with neurobiologist Martin Egelhaaf of Bielefeld University, Schoepe’s model successfully captured the computational principles used by insects.
Schoepe’s model is based on a fundamental principle: always steer towards the area with the least apparent motion. To test the model, the robot was sent through a simulated “corridor” with two walls featuring random patterns. The robot successfully stayed centered in the middle of the corridor, mimicking insects’ behavior in similar situations.
Chicca highlights the versatility of the model, stating that it performs well in various environments, including those with obstacles or small openings. She states, “The model is so effective that once implemented, it can adapt to different environments. That’s the remarkable aspect of this achievement.”
The significance of the model lies in providing insights into how insects navigate and accomplish tasks with remarkable efficiency. Chicca explains that efficiency is often overlooked in the field of robotics, as humans tend to rely on learning new tasks over time. In contrast, insects possess innate efficiency in their brains, enabling them to fly immediately after birth.
Chicca’s research group has previously developed a chip that demonstrates efficiency in computer systems. The chip, smaller than a keyboard key, exemplifies the potential of incorporating specific insect behavior into hardware. Chicca envisions building tiny, energy-efficient chips that perform specific tasks, rather than relying on general-purpose computers with extensive capabilities.
Elisabetta Chicca is affiliated with the Groningen Cognitive Systems and Materials Center (CogniGron), which aims to develop materials-centric systems paradigms for cognitive computing. Their focus spans from materials capable of learning to devices, circuits, and algorithms.
