He NEO research team, from the Department of Languages ​​and Computer Sciences of the University of Malaga (UMA) has developed a ‘smart’ system that controls the traffic light networks of this Andalusian city and of Paris (France) to reduce traffic jams and polluting emissions.

The objective, extrapolated to other cities with more than 200,000 inhabitants, is to find the best traffic light planning and red-yellow-green cycle configuration, in addition to improving the flow of vehicles, also reducing economic losses. All this without spending for municipalities or neighbors.

Bioinspired algorithms

As published by the Discover Foundation, the system is based on so-called bioinspired algorithms, in this case based on the evolution of species. The city is the environment where these traffic light programs evolve, surviving those that produce an improvement in travel times and a decrease in pollution.

In this way, these evolutionary algorithmsthat is, they adapt to the environment as ‘mutations’ occur, changes such as the passing of a red light, which can affect the flow of traffic on a street.

The results are displayed on a computer screen after collecting the data of the change cycles of the traffic lights of both cities. They are subsequently transferred to managers, who can implement the changes without the need for works or large investments, or monitoring the drivers through new applications or regulations that force them to change their habits.

More steps towards smart mobility

As reflected in the study entitled ‘A better understanding on traffic light scheduling: New cellular GAs and new in-depth analysis of solutions’, until now, the scientific approach to so-called ‘intelligent mobility’ has revolved around providing ad algorithms. -hoc to reproduce what researchers had previously calculated on issues such as traffic regulation in green areas or how drivers can get to their destinations as soon as possible.

In this UMA study, the researchers have first designed two new computer programs to create the map of these traffic lights, with their respective change intervals, which will be modified as these algorithms mark what may be best for city traffic. . “It is very difficult to come up with any solution, even if it is very basic, because we must bear in mind that we have to put a lot of relevant data from the cities into the computer”, the UMA researcher Enrique Alba has indicated to the Discover Foundation, who has led the work.

Enrique Alba, researcher at the UMA.

Number of vehicles and speed limit

Among the data included is the number of vehicles, the various types that circulate daily, the signs, and the speed limits of the roads of the cities. If a traffic light in Alameda de Málaga opens, the system will search for different alternatives, once the change has been registered.

When this map of the variables that influence traffic light changes has been configured, the information is entered into SUMO, the most widely used simulator in this scientific field due to its capacity for microanalysis. Thus it can be interpreted by public entities that want to apply it to their cities following the examples of the models tested, not only in the Andalusian city, but also in another example of very different physiognomy and size such as the French capital. “Traffic in a city cannot be encapsulated in a mathematical formula, however complex it may be, hence the system takes into account the characteristics of each city,” the researcher has summarized.

European cities have not historically evolved based on a guide, as others such as the American ones may have, and, in this sense, they are the ones that can most benefit from this type of artificial intelligence studies, which introduce solutions automatic and adapted for each city.

Genetic algorithms

The research applies the so-called celulares cellular genetic algorithms ’, which end up selecting the best proposals for their urban networks, taking into account their own historical evolution. These solutions are going to be the support to program the traffic lights of the city.

The study by the Andalusian research team has shown that, between the two types of genetic algorithms used, synchronous and asynchronous, the latter are more effective, being more realistic in terms of simulating the evolution of species, and also of cities like Malaga or Paris. “This is usually classified as ‘swarm intelligence’, since emerging behavior solves problems better than separate individuals,” Alba explained.

This ‘swarm intelligence’ is reflected in advanced mathematical formulas, which work like the algorithms’ genotypes, with the keys to Malaga or Parisian traffic. They are the ones that the scientists will finally transfer to the interested city council or manager, who will be able to use them as an de instruction manual ’, with a particular and optimal way of programming the traffic lights in the city, reducing pollution or delay in traffic.

The work, which has been financed by the European Regional Development Fund and the Ministry of Science and Innovation, has come from a collaboration developed by this department of the UMA with the National University of Southern Patagonia, in Argentina.

References

Villagra A, Alba E, Luque G. “A better understanding on traffic light scheduling: New cellular GAs and new in-depth analysis of solutions”. Journal of Computational Science.

I like this:

I like Loading …