How self driving cars capable of detecting wet road conditions using artificial intelligence

How self driving cars capable of detecting wet road conditions using artificial intelligence -

Detection of bad weather is included in one of the challenges in building self driving cars. According to the report released by the US Department of Transportation, wet conditions caused 959.760 accidents and the death of 4789 people in 02 and 2012. This figure amounts to 74% of all accidents related to weather in the US and made up 23% of all vehicle accidents in the country.

Fortunately, researchers at the IEEE (Institute of Electrical and Electronics Engineers) understood how to detect a slippery road by analyzing the audio of the car's tires comments and recurrent neural networks (RNN) , an artificial intelligence computer network. The experiment was carried out by fixing a shotgun microphone rear tire 2014 Mercedes CLA. The car traveled at different speeds around the greater Boston area in Massachusetts.

The first test results show precisions with unweighted average return (UAR) of 93.2% for all vehicle speeds. The microphone is also able to receive audio feedback vehicles passing beside him.

This is the first time that artificial intelligence is used to detect the road conditions. However, it is not the first time someone has tried to detect road conditions.

Technical University of Madrid in 2014 experimented using support vector machine (SVM) to analyze sounds when the tire meets the road and classify the different sounds made by asphalt. However, as there were limited number of surface types and audio input irrelevant as the noise of tires bouncing against the stones could lead to false predictions.

University of Toyama in 2014 tried to use surveillance cameras on cars in search of the road reflection from the headlights of other cars, however, this way requires other cars to be present on the same road at the same time. Furthermore, this method works perfectly in the detection of the condition of roads in foggy conditions, snow and poor road lighting.

Next Post »
0 Komentar