Very recently, some advanced predicted the arrival of autonomous cars and in the next decade we will meet millions of them in cities or roads.

One of the biggest challenges that need to overcome the system is the relationship between human drivers and Artificial Intelligence.

Humans have a series of particular features that a machine cannot understand yet. Many positive, but not so many others. Simply, we are unpredictable especially when we take the car and become aggressive beings at the wheel.

Machine learning still needs to understand all this.

Researchers at the Laboratory of Computer Science and Artificial Intelligence (CSAIL) at MIT say they have found a solution to the problem or rather a development base because there is a lot of work ahead.

It is an algorithm that can predict the behavior of drivers with respect to features such as impatience and selfishness.

In real-world tests, MIT researchers observed 25% more accuracy in predicting the behavior of other cars. In other words, autonomous vehicles were able to make solid judgments with the new algorithm.

Researchers are using the ‘Social Value Orientation’ of psychology. Basically, it prioritizes a driver over others by analyzing their social behavior in an external environment.

For this to work, the algorithm initially analyzes the most effective driver responses on the road. Here, the system analyzes the value of a driver's own benefit against others under that orientation of social value.

The algorithm classifies the behavior of the driver and all the drivers around him and makes intelligent decisions while we are on the road. Researchers will continue working on the new algorithm to prepare it for real-world driving and will try to apply the model to other actors such as pedestrians or cyclists.

We leave you a video which explains in broad strokes the algorithm used in this type of vehicles.

Are autonomous cars really the future?

Autonomous cars (together with electric / non-polluting) are a trend in an automobile industry that is going through a great process of technological immersion.

There are dozens of research groups working on AI models and machine learning applied to the motor world and great progress is expected in the next decade although its incorporation into society will be gradual until reaching level 5 of the phase, which is when the vehicle It works totally independently and will not need any kind of human intervention to operate in a totally safe way.