Seite - 69 - in Integration of Advanced Driver Assistance Systems on Full-Vehicle Level - Parametrization of an Adaptive Cruise Control System Based on Test Drives

5.3. Object Selection s 4020 60 80 1000 1 1− 0 0 0.2 0.4 0.6 0.8 1 )u,s(P 1 1− 0 s 4020 60 80 1000 (a) (b) Figure 5.7.: PriorityP (s,u) of eq. (5.43) for parametersP0 = 1,PL= 0.7 andPb= 0.01 for the two cases (a)n= 2 and (b)n= 8 5.3.3. Comparison of Object Selection Algorithms Figure 5.8 shows an example with two objects and the lane markings for a curved two- lane road. Here, the ego vehicle is following object 1, both of which are in the left lane overtaking object 2, which is travelling in the right lane at a lower speed than those of the ego and object 1. The grey marked area depicts the predicted path. If the in- lane algorithm of chapter 5.3.1.1 is used, object 2 will be selected as the OTF because is has a smaller s-coordinate than object 1, s2,fl < s1,rr. However, when using the priority algorithm described in 5.3.1.2, object 1 will be selected becauseP (s1,rr,u1,rr)< P (s2,fl,u2,fl). The situation in fig. 5.8 shows the significant advantage of the priority algorithm. Whenovertakingwidevehicles (e.g. commercial vehicles), theOTFof the in- lane algorithm jumps to the vehicle in the neighbouring lane, which in most situations is the wrong decision. This is the result of the error in the path prediction. If the error is almost zero, both algorithms would select object 1, which is the right decision. However, the predicted path would rarely be identical with the real driven path in the future. Therefore, the use of additional information will help to increase the quality of the object selection algorithms. Ifn in eq. (5.43) is set to very high values, the priority algorithm results in the selection of thenearest object in thepredictedpath, asdescribed inchapter5.3.1. This isbecause the shape of the priority function in the lateralu-direction degenerates to a rectangular function, whereall pointsat the sames coordinatehave the samepriorityP. For further investigations in this work, the priority algorithm withn= 2 is used. 69