Artificial intelligence brings automated driving closer

Collecting data

To test this, three types of data are collected. The car is equipped with cameras that capture images of the surroundings, but also with a Lidar sensor. These sensors are used to measure distances by emitting a kind of pulse, which is reflected by an object. In addition, the car also has metadata about its own speed, weight, etc. With an AI application, it should be possible to combine this data into insights that can make traffic safe.

Automated driving

One of the most important megatrends of recent years has been the increasingly sophisticated capabilities of automated driving, or self-driving cars. Object detection, for example for suddenly crossing people, other vehicles or objects, is a technology that can help to increase safety in traffic. A smart navigation system could possibly also make interventions based on this data, which could make the difference in emergency situations. Although this is still 'future music', the technology already offers many possibilities.

Object detection & memory

The central challenge is that a system, like AI, has no memory like we humans have. If the camera on a car takes two pictures of another car, an AI cannot simply conclude that it is the same car. In the video, Raymond talks about the challenges he and his team faced with the project, but also the complex ethical dimensions that come with applying AI in situations like traffic.

The project is a collaboration between Fontys ICT and partner company HERE, part of the SPARC partner collective. Together with these partners, students in our Fontys ICT InnovationLab are working on applied research in and with the working world.