Browsing by Author "Sağlam, Harun Buğra"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Master Thesis Analysis of lane changes for vehicle strings on highways: String stability, driving safety and comfort(Çankaya Üniversitesi, 2017) Sağlam, Harun BuğraAn important aim of intelligent transportation systems (ITS) is the full or partial replacement of human driver functionality. Cooperative adaptive cruise control (CACC) is a recent technology for automating the longitudinal vehicle motion. Fulfilling the condition of string stability, CACC enables safe vehicle following at small inter-vehicle spacings and hence supports the formation of tight vehicle strings for improving the road capacity. In its classical realization, CACC is limited to the case where vehicle strings are already formed and all vehicles in a string follow each other on the same lane of a road. However, practical driving situations include the case of vehicles entering or leaving a string and performing maneuvers different from only vehicle following. This thesis is concerned with the effect of additional maneuvers due to lane changes (vehicles entering or leaving) on the safety of vehicle strings. Lane changes include gap opening and closing maneuvers and are subject to measurement inaccuracies and sensor errors due to changes of the vehicle locations. Accordingly, the effect of these maneuvers on the longitudinal vehicle motion has to be analyzed. As the first contribution, the thesis argues that the described measurement inaccuracies during lane changes can be modeled by input signal impulses of the respective vehicle. Moreover, opening/closing gap maneuvers can be realized by the generation of suitable feedforward input signals that are nonzero for a limited time. Respecting that multiple lane changes can occur in a vehicle string, the thesis proposes to study the effect of repeated input signals (impulses or time-limited input signals) on the output signal norm of LTI systems. The second contribution of the thesis is extending the definition of string stability to additional disturbances that can be applied to any vehicle in the string. Respecting the same idea, the third contribution of the thesis shows that a bound on the output signal norm of stable LTI systems exists if the repeated input signals (impulses or time-limited signals) are separated by a non-zero dwell-time. Additionally, an original computational procedure for finding a tight bound on the output signal norm is provided. The fourth contribution is the adaptation of these computational methods to the case of stable LTI systems with multiple inputs and outputs. The fifth contribution is the application of the obtained results to vehicle strings. It is shown that suitable analytical bounds for the relevant output signals such as distance error or acceleration can be determined and the results are validated by simulations. The last major contribution is the development of new numerical methods for bounding the matrix exponential function for large LTI systems based on the Jordan canonical form and the Schur decomposition. The evaluation of such norms is needed when computing the output signal norm of large LTI systems such as long vehicle strings.Article Citation - WoS: 1Citation - Scopus: 1Outputs Bounds for Linear Systems With Repeated Input Signals: Existence, Computation and Application To Vehicle Platooning(2018) Schmidt, Klaus Werner; Sağlam, Harun BuğraThis paper investigates the effect of repeated time-limited input signals on the output excursion of stable, linear time-invariant systems. It is rst shown that the maximum norm of the output signal remains bounded if the repeated input signals are separated by a nonzero dwell time. Then a novel method for computing a tight bound on the output signal norm is proposed. The setting of the paper is motivated by a vehicle platooning application, where vehicles repeatedly open/close gaps in order to perform lane changes. The developed method analyzes driving safety by computing a bound on the spacing error between vehicles when performing repeated open/close gap maneuvers.

