Seite - 33 - in Integration of Advanced Driver Assistance Systems on Full-Vehicle Level - Parametrization of an Adaptive Cruise Control System Based on Test Drives

2.4. ACC Controller Table 2.2.: Fuzzy control rules, according to [Ga¨c12] er tc c rd tf vf er˙ ms hd d d d a s d d d sd a ss sd sd za a ha f sd sd a ha ha mf za za a ha ha deceleration of atc,f =ac,f =asd=−1.635m/s2. In general, the defuzzification reads ades= ∑ ier ∑ ier˙ aier,ier˙ µier,ier˙∑ ier ∑ ier˙ µier,ier˙ . (2.42) In the given example, the defuzzification results in ades=−1.635m/s2. Figure 2.19(b) shows the output for the described fuzzy controller. The method used is just an example. There are many other possible ways to vary the steps carried out during the design process of the controller, as described in [Ada09a]. 2.4.3.4. Sliding Mode Controller Unless otherwise indicated, this chapter is based on [Ada09b] and [SEFL14]. With the Sliding Mode Control (SMC) method, the controller switches between two different laws. The main advantage is that they are very robust in terms of parameter uncertainties. One disadvantage is that the switching around the desired value generates chattering. For the design, the model x˙SMC=ASMCxSMC+bu (2.43) is used, where ASMC = A of eq. (2.26). The state vector contains the components xSMC = [−er −er˙ vax]T. The switching function, which is defined as a function of the state vector, reads s(xSMC) =r T xSMC, (2.44) where rT xSMC describes the sliding surface. The control law, which is only defined for s(xSMC) 6= 0, reads u(xSMC) = { u+(xSMC) for s(xSMC)>0 u−(xSMC) for s(xSMC)<0. (2.45) Thereachabilityconditionmeansthat thetrajectoryofxSMC reaches thedefinedsurface swithin a finite time. Most literature uses s˙=−qsign(s(xSMC))−k s(xSMC) (2.46) 33