If you follow electric cars as avidly as I do, then you would have come across this. But for my readers and followers who don’t, I wanted to break down the swarm intelligence. As mentioned on our podcast show, AutoCentral, on Monday morning, autonomous cars are the way of the future.

So what is swarm intelligence?

Swarm Intelligence is somewhat like artificial intelligence that aims to simulate the behavior of swarms or social insects. The expression was first presented in 1989 by Gerardo Beni and Jing Wang in regard to cellular robotic systems. It’s commonly used in the IT sector.

However, since the 50’s researchers were looking for ways to create AI (artificial intelligence). A lot of focus was put to developing artificial neural networks, this contributes to building intelligence from scratch. Then the term “artificial swarm intelligence” was labelled for an approach that harnessed groups of human minds to predict real-world events.

The notion behind this concept of swarm intelligence is to actually learn from nature. “It’s why fish form schools and birds form flocks and bees form swarms,” said Dr. Louis Rosenberg, CEO of Unanimous AI. “In a nutshell, it’s allowing the group to make better decisions than individuals could make alone.” And this differs from “flocking,” which is used in robotics.

How humans can apply rules from our natural environments

The honey bee is quite remarkable and often used for reference for this concept. Their decision-making capabilities are what has inspired human systems.

On an annual basis, honey bees have to split from their colonies and send out scout bees in search of a new location to make a home. Hundreds of them are sent out, with some travelling distances of as much as 78 kilometers. Upon return to their hive, assessments are then made about new environments. Between 400 and 600 bees then form a “waggle dance” to convey information to the cluster. That is how they make their decisions.

Where humans are getting it wrong

It can be said that humans have a less accurate approach in decision-making compared to the bee’s “real-time negotiation”. Humans make use of voting and polls, which Rosenberg says is “primitive”. And because they are polarizing, they frequently get it wrong.

According to Rosenberg, a Professor at Cornell University in New York has found that bees pick the ideal site over 80% of the time. “What’s fascinating is that no single bee could possibly make that decision.”

How does swarm intelligence fit in with self-driving cars?

There are many cities around the world choking in exhaust fumes and not able to handle more traffic. Now, this is where “swarm intelligence” comes in. Manufacturers want to make all these cars “think”/work together and share data in efforts of finding a solution. Basically, to act as a collective swarm and this, in turn, will benefit the transport infrastructure.

In recent times I’ve noticed ‘self-driving’ cars is the new term for what was previously headlining as ‘autonomous cars’. Either way, what good are they if they just replace human-operated ones in the simplest way. The world, or well bubbly cities, don’t necessarily need fewer cars but smarter ones. And by smarter, I, of course, mean intelligent.

The impact this will have is far-reaching. From this, carbon emissions will be reduced, there will be less traffic congestion, in addition to a higher level of safety. This is possible by the reduced chance of a collision within the swarm and ‘foreign’ objects.

Data sharing for cars

Data sharing is fast becoming a reality that’s moving beyond infotainment and navigation systems. The automakers are working tirelessly on developing systems for integration on future models that will use data from the surrounding traffic environment. These include systems that allow drivers to maintain an optimal speed as they drive through a succession of green robots and avoid unnecessary braking. Then, there’s a cloud-based slippery-road alert that allows vehicles to share warnings about hazardous surface conditions. There’s one that gives a considerable advanced notice of an approaching emergency vehicle to avoid the element of surprise. But the success of these is somewhat dependent on the size of the active swarm.

It all lies in the hands of the automakers and how many follow suit. If they all can effectively integrate similar technology, then the swarm becomes more intelligent. Through effective use and development of collective intelligence and car-to-car technology, self-driving cars will be able to move in seamless swarms. Imagine a world of constantly-flowing traffic, hazard-free cities, safe optimum average speed – sounds like a dream, right?

Drivers can then safely interact on their phone or tablets, sit back and relax, or multitask while the car does all the necessary actions and taking the driving responsibility.