One of the most interesting conundrums in AI is that it can easily solve adult-level thinking problems but struggles with basic motor controls and reactions that toddlers are capable of.
AI can defeat a world-class chess champion because it has clearly defined rules to work with and can come up with patterns more easily than any human alive.
But when it comes to navigating a robot across a floor with simple obstacles, the matter is not so simple.
Developing AI’s capabilities is essential for business, humanitarian, and the general benefit of mankind. We already rely on it heavily, especially in businesses. For example, 26% of marketers recognize that it’s a major development in marketing. And 79% of the best-performing businesses have been using automation and AI for lead generation for three-plus years.
One helpful way to advance AI is to rely on animals and their behavior to inform this field. Let’s break down some ground-breaking studies of animal behavior applied to Artificial Intelligence.
How Animal Behavior Can Be Applied to AI
The variety of animal species and unique behaviors in different groups can radically alter robotics if such activities can be tracked. Here are some of the most interesting examples and their potential applications.
Dogs to Help Develop Spatial Intelligence
Dogs navigate human environments all the time and with ease. They can tell the difference between babies and adults. They’re also capable of navigating obstacles and identifying and spaces to walk, run, and play.
Dogs have senses that can detect danger and they instinctively protect humans. While this last characteristic will take time to develop, there’s already an interesting study to develop training data from dogs.
Cameras attached to a dog going about its daily activities observed how dogs react to obstacles in its path. Machine learning was then applied and the result was that the neural network could differentiate between outdoor and indoor spaces and spaces to walk on with fair accuracy.
This is a tremendous time-saving activity that can help develop robotics faster.
Group Behavior to Train Drones and Robots
Jackdaws are birds that fly in groups for foraging purposes. They have complex social structures that enable them to carry out ‘flocking’ i.e. moving in groups to fly or forage. We see similarities in how fish swim in shoals, how insects swarm, and how land animals move in herds.
Jackdaws’ social groups use cues from other birds to maintain distance, save energy, and forage for food. In a groundbreaking study on bird flocking applied to AI, the researchers found that mating birds behave differently from others. They rely on cues from their partners rather than the whole group and are able to forage more efficiently.
For drones, this kind of knowledge can help groups of drones maintain the right distance from each other and move in formation. As this type of research goes ahead, it will be possible to do surveillance of structures like dams, animal migrations, and climate-related concerns.
Ants also have social group structures and transmit information to each other that allows them to find food and then inform the colony about the route to the food source. When something in the environment, like a fallen branch, interrupts the route, ants need to find new routes and communicate this to the colony. Applied to AI, this can help swarm robots that can react and deal with environmental changes such as winds, accidents, other animals appearing, etc.
This is extremely useful for robots maneuvering situations where unexpected changes can occur such as in a burning building or dangerous exploration conditions in mines or caves.
There are more examples of how animals can help improve robotics and AI. The maneuverability of a hummingbird can help create a drone that navigates tight spaces and move in different directions. Studying how the stickleback fish species respond to climate change can help develop AI that detects significant environmental changes.
Nature has already created animals with incredible abilities to adapt and thrive in harsh and unpredictable conditions. Being able to learn from them can transform human lives and save the earth by building robots and other AI technology that respond to challenging and difficult environmental events.