Seite - 85 - in Integration of Advanced Driver Assistance Systems on Full-Vehicle Level - Parametrization of an Adaptive Cruise Control System Based on Test Drives

6.3. Validation of the Identified Parameters are smaller than those in the measurements. This indicates that the vehicle movement will be more comfortable than the human driving in the measurement. The difference between measured and simulation is greater in the inter-vehicle distance srOTF than in the inter-vehicle range rate sr˙OTF. This is due to the fact that the desired relative speed sr˙OTF should be controlled to zero. The desired inter-vehicle distance depends on the selected time gap τset, see eqs. (2.22) and (2.23), which could be chosen unbounded by the driver. An error in τset will lead directly to an error between measurements and simulation for the inter-vehicle distance. Since the identified parameter P4 shows the weight of er to er˙, it has the unit s −1. This means that an error of er˙ = 1m/s results in the same desired acceleration as er≈ 5m. This value correlates very well with the values mentioned in the literature. In [WDS09], Winner er al. shows that a weight of P4 = 0.2s −1 will lead to good results. Ga¨chter determined that a good weight is in the range of 0.2s−1 < P4 < 0.25s−1, where increasing the weight leads to a more sporty behaviour of the vehicle, [Ga¨c12]. Thus, small weights make the ACC-equipped vehicle decelerate earlier and have less undershoot in the inter-vehicle distance when approaching a slower OTF, compared to large values forP4. 6.3.3. Full-Vehicle Model with ACC Measurements In this chapter, the performance of the identified parameters is compared with measure- ments with a real ACC-equipped vehicle. The target of this comparison is not to achieve the identical system behaviours but rather for the recorded and simulated data to have similar shapes, especially for the longitudinal acceleration vax. The ego vehicle travels behind the OTF at a set time gap of τset= 1.2s. Both the ego vehicle and the OTF drive at vvx≈ vvOTF,x≈ 58km/h. At time t= 47s, the OTF begins to accelerate until it reaches the speed vvOTF,x = 77km/h, with a maximum longitudinal acceleration of vaOTF,x,max = 1.3m/s 2. It starts to decelerate at time t= 70s with a maximum deceleration vaOTF,x,min=−1.1m/s2 until it reaches the final speed of the OTF, vvOTF,x= 58km/h. The longitudinal speed vvx and acceleration vax of the ego vehicle and the inter-vehicle distance and range rate srOTF and sr˙OTF are recorded. Figure 6.13 shows the measured time histories for both the ego vehicle and the OTF. The simulation is carried out using a commercially available software package called CarMaker,whichisaproductofIPG Automotive GmbH. Itprovidesaninterface inwhich custom controllers can be implemented. In addition, optimal environmental-recognition sensors and traffic objects are already available. For the simulation, the ACC controller of eq. (6.9) with the parameters of eq. (6.31) was implemented in the simulation. As an environmental-recognition sensor, the provided optimal Radio Detection and Ranging (RADAR) sensor was used with a Field of View (FOV) described by the maximum detectionrangerFOV = 200mandanapertureangleϕFOV =±8°. Thewholesimulation was done on a straight road. The measured motion of the OTF was fed into the simulation tool CarMaker. The ego 85