Control Systems Design for Automatic Drive of a Passenger Car in Critical Scenarios
Total Page：16
File Type：pdf, Size：1020Kb
University of Windsor Scholarship at UWindsor Electronic Theses and Dissertations Theses, Dissertations, and Major Papers 2016 Control Systems Design for Automatic Drive of a Passenger Car in Critical Scenarios Jerome Blanc University of Windsor Follow this and additional works at: https://scholar.uwindsor.ca/etd Recommended Citation Blanc, Jerome, "Control Systems Design for Automatic Drive of a Passenger Car in Critical Scenarios" (2016). Electronic Theses and Dissertations. 5802. https://scholar.uwindsor.ca/etd/5802 This online database contains the full-text of PhD dissertations and Masters’ theses of University of Windsor students from 1954 forward. These documents are made available for personal study and research purposes only, in accordance with the Canadian Copyright Act and the Creative Commons license—CC BY-NC-ND (Attribution, Non-Commercial, No Derivative Works). Under this license, works must always be attributed to the copyright holder (original author), cannot be used for any commercial purposes, and may not be altered. Any other use would require the permission of the copyright holder. Students may inquire about withdrawing their dissertation and/or thesis from this database. For additional inquiries, please contact the repository administrator via email ([email protected]) or by telephone at 519-253-3000ext. 3208. Control Systems Design for Automatic Drive of a Passenger Car in Critical Scenarios by Jerome Blanc A Thesis Submitted to the Faculty of Graduate Studies through the Department of Mechanical, Automotive, and Materials Engineering in Partial Fulfillment of the Requirements for the Degree of Master of Applied Science at the University of Windsor Windsor, Ontario, Canada 2016 c 2016 Jerome Blanc Control Systems Design for Automatic Drive of a Passenger Car in Critical Scenarios by Jerome Blanc APPROVED BY: Dr. N. Kar Department of Electrical and Computer Engineering Dr. B. Minaker Department of Mechanical, Automotive, and Materials Engineering Dr. J. Johrendt, Advisor Department of Mechanical, Automotive, and Materials Engineering 29 August 2016 Declaration of Originality I hereby certify that I am the sole author of this thesis and that no part of this thesis has been published or submitted for publication. I certify that, to the best of my knowledge, my thesis does not infringe upon anyones copyright nor violate any proprietary rights and that any ideas, techniques, quotations, or any other material from the work of other people included in my thesis, published or otherwise, are fully acknowledged in accordance with the standard referencing practices. Furthermore, to the extent that I have included copyrighted material that surpasses the bounds of fair dealing within the meaning of the Canada Copyright Act, I certify that I have obtained a written permission from the copyright owner(s) to include such material(s) in my thesis and have included copies of such copyright clearances to my appendix. I declare that this is a true copy of my thesis, including any final revisions, as approved by my thesis committee and the Graduate Studies office, and that this thesis has not been submitted for a higher degree to any other University or Institution. iii Abstract Advanced Driver Assistance Systems (ADAS) aim at supporting the driver's task in order to improve vehicular safety. One of the most promising and most studied technologies in this direction is Autonomous Driving (AD). While control systems for AD based on lane markings have been proposed in the liter- ature, few have addressed the problem of coping with the absence of lane references on the ground. Moreover, many of these solutions resort to complex software and/or hardware. In this project a relatively straightforward way of restoring suitable knowledge of the position of the vehicle when the output of the Lane Recognition Camera (LRC) is not available or degraded is presented. This is done exploiting a relatively new approach to variable recovering which results in a so-called \virtual sensor". In order to show the potential of this solution, then, a control system based on a LRC is first developed in the Simulink R environment. Subsequently, the virtual sensor for precise vehicle position reconstruction is implemented and evaluated against the first solution. Simulations considering realistic driving conditions showed comparable levels of performance between the two systems, demonstrating the effectiveness of this new approach. iv To my family, for their relentless support. v Acknowledgements This thesis is the final outcome of a two year Double Degree Master program made possible only thanks to the collaboration and organizational efforts of two universities, University of Windsor and Politecnico di Torino, and a prestigious industrial partner such as Fiat Chrysler Automobiles. Therefore I would like to express my sincere appreciation to the persons who represent the aforementioned institutions and, in particular, to Mohammed Malik from FCA Canada. I would like to extend my gratitude to my academic advisor at the University of Windsor, Dr. Jennifer Johrendt, who has supported me throughout the project and to my industrial advisor from Centro Ricerche Fiat, Dr. Pandeli Borodani, for his help, encouragement and suggestions. That said, my deepest recognition goes to my parents, for the education they gave me and for their continuous moral and economic support throughout my studies. A huge \thank you" goes to my brother, for being my biggest fan, just as much as I am his. Last but not least, special recognition goes to my colleagues, house-mates and friends, Marco \Div" Di Vittorio, Davide \Bore" Borello, Mirko \una brava persona" Pesce, Marco \the tall one" Gerini e Davide \duro ma onesto" Pezzetti, for making this experience unique and unforgettable. vi Table of Contents Declaration of Originality iii Abstract iv Dedication v Acknowledgments vi List of Tables x List of Figures xi List of Abbreviations xvi List of Symbols xx Chapter 1: Introduction 1 1.1 Problem statement . 2 1.2 Objectives . 4 1.3 Methodology . 5 1.4 Thesis Organization . 10 Chapter 2: Theory 11 2.1 Vehicle Dynamics . 11 2.1.1 Tire Contact Modeling . 12 2.1.2 Longitudinal Dynamics . 14 2.1.3 Lateral Dynamics . 15 2.2 Virtual Sensors . 17 2.3 Neural Networks and System Identification through ARX . 21 2.3.1 Artificial Neural Networks . 22 2.3.2 Non-linear ARX . 24 vii TABLE OF CONTENTS Chapter 3: Background and Literature review 27 3.1 General Concepts . 28 3.1.1 Vehicle Safety . 28 3.1.2 ADAS . 30 3.1.3 Levels of Automations . 39 3.1.4 Look-down and Look-ahead Approaches . 42 3.1.5 Data Fusion . 44 3.1.6 V2X . 47 3.2 Model Components . 50 3.2.1 Electric Power Steering . 50 3.2.2 Lane Recognition Camera . 52 3.2.3 Global Positioning System . 58 3.2.4 Control Systems . 60 3.2.5 Virtual Sensors . 64 3.3 Ongoing Research and AD in Urban Scenarios . 66 3.3.1 Example of ADAS for Unmarked Urban Scenarios . 70 Chapter 4: Description of the Model 74 4.1 Blocks Modeling . 74 4.1.1 Modeling Softwares . 74 4.2 Vehicle Dynamics . 77 4.2.1 Longitudinal Dynamics . 78 4.2.2 Lateral Dynamics . 79 4.3 EPS - Electric Power Steering . 80 4.3.1 Experimental Data . 80 4.3.2 Neural Network Models . 82 4.3.3 Non-linear ARX Models . 90 4.4 LRC - Lane Recognition Camera . 93 4.5 Controllers Design . 96 4.5.1 Longitudinal Control . 97 4.5.2 Lateral Control . 101 viii TABLE OF CONTENTS 4.6 Virtual Sensor Design . 109 4.6.1 Used Data . 109 4.6.2 Procedure . 110 4.6.3 Implementation . 113 Chapter 5: Simulation Procedures 115 5.1 Simulations Description . 115 5.2 Model Validation . 116 5.3 Vehicle Trajectory in the X-Y Plane . 117 5.3.1 Input Coordinates Transformation . 117 5.3.2 Output Coordinates Transformation . 118 Chapter 6: Results and Discussion 119 6.1 Uncontrolled Model Performance . 119 6.2 Controlled Model with LRC . 121 6.2.1 Longitudinal Control . 121 6.2.2 Lateral Control . 122 6.3 Controlled Model with Virtual Sensor . 130 6.3.1 Longitudinal Control . 132 6.3.2 Lateral Control . 133 Chapter 7: Conclusion and Recommendations 140 Bibliography 144 Vita Auctoris 154 ix List of Tables 3.1 Advantages of V2X in terms of safety and convenience to drivers . 49 3.2 High-level comparison between mono and stereo camera ADAS systems [62] 54 4.1 Effects of increasing a parameter independently . 100 4.2 Ziegler-Nichols' heuristic tuning technique for PID controllers and related variants . 101 x List of Figures 1.1 Road deaths in the European Union [3]. Solid blue: proposed objective; solid red: actual data; dashed red: projected estimate . 2 1.2 Considered vehicle: 2015 Fiat 500x ....................... 6 1.3 The proposed Control Scheme . 8 2.1 Longitudinal tire dynamics . 12 2.2 Typical µ(λ) curves for different road conditions. If scaled by the appropriate vertical load Fz the ordinate axis represents the tractive force Fx . 13 2.3 Tire contact patch deformation in a bend. Notice the (Side)slip angle α [58] 13 2.4 Forces acting on the vehicle in longitudinal motion [58] . 14 2.5 Cornering motion studied with the bicycle model [58] . 15 2.6 Vehicle Sideslip Angle β (positive in clockwise direction) [58] . 17 2.7 Standard observer design approach (from [82]) . 19 2.8 One step approach for virtual sensor design (from [82]) . 20 2.9 ANN with 3 inputs, 4 hidden nodes, 2 outputs . 22 2.10 A simple neuron [103] . 23 2.11 nlarx structure . 25 2.12 nlarx model in a simulation scenario . 26 3.1 ESP intervention in an extreme steering maneuver . 30 3.2 Lane Departure Warning system . 33 3.3 Adaptive Cruise Control system .