This paper introduces a method to represent Takagi-Sugeno Fuzzy Control Systems (FCSs) as computational graphs, so they can be adjusted through a supervised training process based on gradient descent. It has been tested both with artificial (i.e. a known fuzzy controller) and naturalistic (i.e. driver's data extracted from the vehicle and the environment) data. The results achieved show high conformance to synthetic data, and seem to describe a car-following behavior with quite good precision, which suggests that it is possible to model the driver's behavior in a longitudinal model based on if-then type rules.
Fuzzy Controller Inference via Gradient Descent to Model the Longitudinal Behavior on Real Drivers
01.06.2019
2635702 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
FUZZY CONTROLLER INFERENCE VIA GRADIENT DESCENT TO MODEL THE LONGITUDINAL BEHAVIOR ON REAL DRIVERS
British Library Conference Proceedings | 2019
|A Recognition Method of Drivers Characteristics Using Fuzzy Inference
British Library Online Contents | 1995
|An Adaptive Network Fuzzy Inference System Controller for Car Following Behavior
Trans Tech Publications | 2011
|Derivation and Tuning of PID Controller via Direct Gradient Descent Control Method
British Library Online Contents | 2000
|ROUTE CHOICE BEHAVIOR MODEL USING FUZZY INFERENCE
British Library Conference Proceedings | 2003
|