Connect with us

Artificial Neural Networks

Dragonflies and Missile Defense Systems

Updated

 on

Dragonflies have extremely fast reflexes with little depth perception. Their reaction time to prey that is moving through the air or ground is 50 milliseconds, the same amount of time it takes for information to cross three neurons. Sandia National Laboratories is working on research to figure out how dragonfly brains work and learn the ways they are able to calculate complex trajectories. 

The research is led by computational neuroscientist Frances Chance. She is the one responsible for developing the algorithms, and she will be presenting her research at the International Conference on Neuromorphic Systems in Knoxville, Tennessee. The research has already been presented at the Annual Meeting of the Organization for Computational Neurosciences in Barcelona, Spain. 

Frances Chance specializes in replicating biological neural networks like brains, especially neurons and the process of sending information throughout the nervous system. Brains are more complex and better versions of computers. They are more energy efficient while leaning and adapting at a faster speed. 

“I try to predict how neurons are wired in the brain and understand what kinds of computations those neurons are doing, based on what we know about the behavior of the animal or what we know about the neural responses,” Chance said. 

The research conducted by Sandia National Laboratories included creating a simple environment that had generated dragonflies through computer simulations. They used computer algorithms to make the dragonflies catch prey just like their real-life counterparts. The computer simulated dragonflies were able to process visual information while hunting just like dragonflies in the real environment. This showed that programming in this manner is possible, which could be applied in many different sectors. 

The new research is already being applied to the missile defense sector. Using the same system as the one with the computer simulated dragonflies could improve missile defense systems. Missile defense systems work in a similar way as dragonflies targeting and catching prey. They intercept an object in flight like a dragonfly intercepts prey in the environment. Dragonflies are one of the top predators in the world as they catch 95% of the prey they target.

With these new developments, they are trying to make on-board computers on missile defense systems smaller while still being fast and accurate. The current way missile defense systems work is through established intercept techniques that require a heavy computation load. This is one of the areas a model based on dragonflies and their prey can help. 

The new technology and research could help improve missile defense systems in many ways including reducing the size, weight, and power needs of onboard computers. Then, interceptors could become smaller and lighter which will make it much easier for them to move around. The new systems could also learn new ways to intercept moving targets like hypersonic weapons. Unlike ballistic missiles, these targets do not follow a similar predictive trajectory or pattern. Finally, the system could be able to use simpler sensors rather than the complex ones used now to intercept a target. 

One of the problems with this research and the idea is that missiles and dragonflies travel at very different speeds. This could cause some discrepancies

Outside of missile defense systems, the computation model of dragonfly brains could also help develop better machine learning and artificial intelligence. As the use of this kind of technology and artificial intelligence grows, it is finding its way into more and more sectors. The defense sector is one that is using this to become much more efficient and grow rapidly. This research shows how we can develop complex systems based on those that already exist in our environment, among those are dragonflies and their brains. Our new technology allows us to model this and create a better version.

 

Alex McFarland is a historian and journalist covering the newest developments in artificial intelligence.