Intelligent dual control supervision for vehicle crash system

Abstract EN

The work presented in this paper demonstrates an intelligent selfoptimized dual controller framework, which is used to govern the mechanics of a car crash non-linear model.

The crash model considered here consists of a single mass coupled with a nonlinear spring and damper model.

An PBF neural network deployed to learn and then tackle the nonlinearities of the system.

Dual controller which can operate as Proportional-Integral-Derivative (P1D), or as zero-pole placement are intelligently alternated and tuned using a fuzzy logic adaptive supervisor.

Sample simulation results are used to illustrate the effectiveness of the intelligent dual controllers framework with respect to tracking desired speed changes and achieving the desired speed of response, whilst penalising excessive control action.