Please use this identifier to cite or link to this item:
http://hdl.handle.net/11375/27335
Title: | SmartCruise: A Highway Driver Assist Technology |
Authors: | Ghasemi Dehkordi, Shiva |
Advisor: | Emadi, Ali |
Department: | Mechanical Engineering |
Publication Date: | 2021 |
Abstract: | Driver fatigue is one of the leading causes for motor vehicle accidents. The riseof automated vehicle technologies and autonomous features, not only significantlyreduces the effects of driver fatigue in turn reducing motor vehicle accidents, it alsoreduces fuel consumption, traffic congestion, travel time, and significantly improvesvehicle accessibility for people who are not able to drive a car themselves.In this thesis, a new highway driver assist technology called SuperCruise ispresented. SuperCruise takes over parts of the driving task by controlling thethrottle, braking, and steering of the vehicle.For the longitudinal control of the vehicle, a new Adaptive Cruise Control(ACC)system is introduced which solves the problem of repetitive switching between thetwo modes of the classic ACC system. To achieve this, a novel switching algorithmis implemented that creates a hysteresis between the two modes by implementing aset of logical comparisons which smoothens the transition between the two modes,and reduces repetitive switching. This results in lower longitudinal accelerationand jerk, ensuring a more comfortable ride.For the development of the lateral controller which is the Lane Centering As-sist(LCA), a Fuzzy Model Predictive Controller(MPC) is developed by implement-ing fuzzy control logic into the formulation of the MPC. This system automaticallytunes the cost function parameters of the optimization algorithm used in MPC toimprove the lateral stability of the system based on the current lateral deviationand heading angle error of the vehicle with respect to the desired trajectory.iii The longitudinal system is then compared to the classic ACC via three differentdriving scenarios. Results show significant reduction is the switching between thesemodes which results in lower longitudinal accelerations and hence an improvementin driver comfort.The performance of the proposed lateral controller on the other hand, hasbeen evaluated by comparing the path following performance, lateral stability,and driver comfort of the system with two of the well known methods used inliterature; Stanley method and classic MPC in three different driving scenarios.Results show that implementation of Fuzzy control logic into the MPC, improvesthe lateral stability of the system and reduces the unnecessary oscillations in steer-ing which in turn reduces lateral acceleration and increases driver discomfort. |
URI: | http://hdl.handle.net/11375/27335 |
Appears in Collections: | Open Access Dissertations and Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ShivaGhasemiDehkordi_Thesis_FinalVersion_RevisionsImplemented.pdf | 6.55 MB | Adobe PDF | View/Open |
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.