Automatically identification and classification of traffic and road events is achieved by collecting metrics regarding one or more monitored vehicles 105 from stationary traffic sensors 1 installed on or proximate to a discrete segment of a road 102, traffic monitoring infrastructure and/or one or more connected vehicles 105. A position and a speed is determined for each monitored vehicle 105 per discrete segment in each lane of said road 102. One or more traffic and road events on said road are then identified and classified. The classifying may be achieved by machine learning algorithms, trained incrementally based on real-life or simulated data, and verified by recursive prediction and outcome assessment. Weather data or other externally acquired data may be used to clarify the nature of the classification. The machine learning algorithm may include a convolutional neural network and a recurrent neural network. An alert may be generated.
Traffic event and road condition identification and classification
2021-01-06
Patent
Elektronische Ressource
Englisch
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
TRAFFIC EVENT AND ROAD CONDITION IDENTIFICATION AND CLASSIFICATION
Europäisches Patentamt | 2022
|Traffic event and road condition identification and classification
Europäisches Patentamt | 2019
TRAFFIC EVENT AND ROAD CONDITION IDENTIFICATION AND CLASSIFICATION
Europäisches Patentamt | 2023
|TRAFFIC EVENT AND ROAD CONDITION IDENTIFICATION AND CLASSIFICATION
Europäisches Patentamt | 2021
|