Researchers at the American MIT have developed an algorithm that allows self-driving cars to change more like people from lanes. Up to now, this has been done with self-driving cars mostly on the basis of static information.
According to the researchers, most existing algorithms are not sufficient, because there are two drawbacks. For example, there are algorithms that are entirely based on detailed static models of the environment of driving. These are difficult to analyze spontaneously and directly. There are also models that are so simple that it only leads to conservative, prudent, impractical decisions, which often leads to the decision not to change jobs.
The new model of MIT researchers is based on buffer zones. to the vehicle. This technique is already being used in existing algorithms, but according to the scientists, this is limited to the pre-calculation of the zones. The MIT algorithm also uses this technique, but if it leads to inhumane decisions, new buffer zones are calculated in real time. Especially in heavy traffic, according to the researchers, only the use of pre-calculated zones is too restrictive.
The algorithm was tested in a simulation with sixteen autonomous vehicles. Each car was thereby set to a different risk profile, so that different driving styles emerged. The cars all drove side by side without conflicts or collisions. The research will be presented at the International Conference on Robotics and Automation.