Not only drivers have a harder time driving during rainfall. It turns out that artificial intelligence also has problems with driving in such conditions. It is a conclusion of the research carried out by the Motor Transport Institute (ITS) and carried out under the project “AV-PL-ROAD – Polish road to road transport automation” in collaboration with Ministry of Infrastructure and the Warsaw University of Technology.

The artificial intelligence system tries to track and predict what other computer-driven or human-driven cars will do. This does not mean, however, that in rain or snow, smart vehicles with all extensive electronics, will cope better than the average human driver – claims Prof. Marcin Ślęzak, ITS director. He adds that difficult weather conditions are an additional challenge for smart AI vehicles, as well as for people. This is caused by the distortion in the functioning of LIDAR (the most accurate laser sensor on board a self-driving car) and less accurate radars and cameras due to bad weather conditions. This is why most road tests of such vehicles are carried out in sunny places, e.g. in California, Arizona or Texas – notes Prof. Ślęzak.