Groundbreaking tracking technology that reveals new insights into how desert ants navigate their complex worlds could inspire the next generation of smart, efficient robots.
An international research collaboration involving the University of Sheffield has developed a new tracking technology that uses computer vision—a field of computer science that programs computers to interpret and understand images and video—to track the individual ants in the desert throughout their life in search. The device documents an ant’s journey from when it first leaves its nest until it finds a place to eat and returns to its colony.
Their new data reveals that ants quickly learn—memorize—their routes back home after a successful trip. But interestingly, their outward routes evolved over time showing different strategies for exploration versus exploitation. The high-precision data also revealed an underlying oscillatory movement invisible to the human eye, which could explain how ants create complex search patterns that adapt to current conditions.
While the new software works on animal models, and uses video captured with standard cameras, it has already been adopted by many international research groups, and is well suited to citizen science projects. . The high-precision data collected is essential to understanding how brains can guide animals in their complex worlds, inspiring a new generation of bioinspired robots.
The new technology and dataset—developed by Dr. Michael Mangan, a Senior Lecturer in Machine Learning and Robotics at the University’s Department of Computer Science with Lars Haalck and Benjamin Risse at the University of Münster, Antoine Wystrach and Leo Clement at the Center for Integrative Biology in Toulouse and Barabara Webb at the University of Edinburgh- shows a new study published in the journal Science Advances.
The study describes how CATER (Combined Animal Tracking & Environment Reconstruction) uses artificial intelligence and computer vision to track the position of an insect in video captured using off-the-shelf cameras. The system can even detect small objects that are difficult to see with the naked eye, and is robust to background clutter, obstructions and shadows allowing it to function in the animal’s natural habitat where other systems fail.
Dr. Michael Mangan, Senior Lecturer in Machine Learning and Robotics at the University of Sheffield, said, “We got this data during a summer field trip, but it took 10 years to build a system that could take the data, so you could. said to be a decade in the making.
“I am always amazed how these insects can travel long distances – up to 1 km – in such harsh landscapes where the temperature is above 50 degrees celsius.
“Until now, desert ants have been tracked manually with pen and paper, which involves creating a grid on the ground with wire and stakes and monitoring their behavior within the grid. Another method used to get around it is to use the Differential Global Positioning System (GPS)—but the equipment is expensive and the accuracy is low.
“The lack of a cheap, robust way to capture accurate insect tracks in the field has led to gaps in our knowledge about desert ant behavior. which could simplify the task.”
CATER’s new visual tracking method addresses these challenges by capturing high-resolution footage of ants in their natural environment and using imaging technology to identify individual ants based on movement alone. . A new image mosaicking technique is then used to reconstruct, or stitch, the scene from the high-resolution imagery. This new method bridges the gap between field and laboratory studies, providing a unique insight into the navigational behavior of ants. Such data could be crucial in revealing how animals with brains smaller than a pinhead navigate their complex environments so effectively.
Such insights have already been turned into commercial products by pioneering University of Sheffield spin-out company Opteran, which reverse-engineers insect brains to create more robust autonomy using sensors in low cost and computation.
Dr. Mangan said, “Desert ants are the ideal inspiration for the next generation of robots—they navigate long distances, through difficult environments, and don’t rely on pheromone trails like other ants, or GPS and 5G like today’s robots.
“We hope that our tool will allow us to create a more complete picture of how insects learn to navigate their habitats, bringing new scientific knowledge and informing engineers if how they are able to create similarly capable artificial systems.”