The main objectives of vehicle motion control on an automated highway system are stable and safe automatic longitudinal and/or lateral path following in a platoon of vehicles. Various controllers can be used to satisfy the same objectives, but they may require different variables to be sensed or different conditions to be met. Supervision can select a controller and switch to a different controller depending on the conditions. Specifically, a fuzzy supervisory expert system checks for various system conditions and chooses a controller from PID, PI, sliding mode and a fuzzy controller or gives a distress signal. The choice between PID and PI controllers is based on the availability of the error derivative. Robust and complex sliding mode control can counter the external disturbances which the PID and PI controllers cannot handle. The fuzzy controller is used when the sensors are not working perfectly but the sensor values are still reliable enough to define corresponding fuzzy linguistic variables.
Agogino, A. M.
A Comprehensive strategy for longitudinal vehicle control with fuzzy supervisory expert system.
1995 IEEE International Conference on Systems, Man and Cybernetics, 1
Institute of Electrical and Electronics Engineers.