Real state information, which influences the switching times of a light signal system, is supplied as input signals to a first neural network in a fixed time cycle. The first neural network calculates estimated state information as a replacement for real state information or parts of the real state information which are not received in good time or are received incorrectly in the fixed time cycle. This estimated state information is output to artificial intelligence which predicts the switching times. The first neural network allows the artificial intelligence to also make good predictions for the switching times of signal groups when one of the many communication paths involved fails or is overloaded. It is therefore possible to predict signal group states in the fixed time cycle in real time with a high degree of robustness and tolerance with respect to gaps in the time cycle of the real state information provided.
VERFAHREN UND ANORDNUNG ZUR VORHERSAGE VON SCHALTZEITPUNKTEN EINER SIGNALGRUPPE EINER SIGNALANLAGE ZUR STEUERUNG EINES VERKEHRSFLUSSES
METHOD AND ARRANGEMENT FOR PREDICTING SWITCHING TIMES OF A SIGNAL GROUP OF A SIGNAL SYSTEM FOR CONTROLLING TRAFFIC FLOW
PROCÉDÉ ET AGENCEMENT POUR PRÉDIRE DES TEMPS DE COMMUTATION D'UN GROUPE DE SIGNAUX D'UN SYSTÈME DE SIGNAL POUR COMMANDER UN ÉCOULEMENT DE TRAFIC
01.05.2024
Patent
Elektronische Ressource
Deutsch
Europäisches Patentamt | 2021
|Europäisches Patentamt | 2021
|Verfahren und Vorrichtung zur dynamischen Steuerung einer Signalanlage
Europäisches Patentamt | 2021
|