This thesis addresses the problem of controlling the lateral motion of an autonomous vehicle in the presence of parametric uncertainties, disturbances, and hard nonlinearities in the steering system, such as backlash in gears, stiction, hysteresis, and dead zones. The lateral motion of an autonomous vehicle is controlled by two cascaded controllers, the trajectory tracking controller and the steering angle controller. This thesis focuses on the development of both controllers using robust and adaptive control techniques. Two control strategies are developed to control the electric power steering angle, sliding mode control and adaptive backstepping control. The limitation of sliding mode control is first addressed, which is the chattering phenomena, and then a proposed methodology is presented to solve it using variable gain sliding mode control. Self-aligning moment actas as disturbance on the steering system that the controller has to compensate for. A model-based approach to estimate it is first developed and its limitations are addressed, which is tire parameters dependence. Two other approaches are then developed to overcome these limitations, the first one is a sliding mode observer, and the second one is part of a backstepping controller. Two approaches are developed to control the vehicle lateral trajectory, non-adaptive backstepping and adaptive backstepping. The extended matching design procedure is used in the adaptive backstepping controller to avoid the overestimation problem. Road curvature must be accurately know by the controller to follow the planned trajectory. It is usually measured by a camera, but the quality of the measurement is affected by environmental factors. An adaptive law is developed to estimate the road curvature online as part of an adaptive backstepping controller. Two feedforward approaches are presented to compensate for road curvature, one is derived from steady state vehicle lateral dynamics, and another is based on estimating the transfer function dynamics from road curvature to steering angle. Road bank angle is a significant disturbance in vehicle lateral control systems. A vehicle lateral state and disturbance observer is developed to estimate the road bank angle and the vehicle side slip angle, which are expensive to measure in current road vehicles, using extended Kalman filter. The observer combines a dynamical vehicle model with two measurements from inexpensive sensors.

The next generation of autonomous vehicles will provide major improvements in traffic flow, fuel efficiency, and vehicle safety. Several challenges currently prevent the deployment of autonomous vehicles, one aspect of which is robust and adaptable vehicle control. Designing a controller for autonomous vehicles capable of providing adequate performance in all driving scenarios is challenging due to the highly complex environment and inability to test the system in the wide variety of scenarios which it may encounter after deployment. However, deep learning methods have shown great promise in not only providing excellent performance for complex and non-linear control problems, but also in generalizing previously learned rules to new scenarios. For these reasons, the use of deep neural networks for vehicle control has gained significant interest. In this book, we introduce relevant deep learning techniques, discuss recent algorithms applied to autonomous vehicle control, identify strengths and limitations of available methods, discuss research challenges in the field, and provide insights into the future trends in this rapidly evolving field.

This report summarizes the research results of TO4233, "Fault Tolerant Autonomous Lateral Control for Heavy Vehicles". This project represents a continuing effort of PATH's research on Automated Highway Systems (AHS) and more specifically in the area of heavy vehicles. Research on the lateral control of heavy vehicles for AHS has been going on at PATH since 1993. MOU129, "Steering and Braking Control of Heavy Duty Vehicles" was the first project and it was followed by MOU242, "Lateral Control of Commercial Heavy Duty Vehicle". Both projects were concerned mostly with the theoretical portion of the problem, i.e. model development, analysis of the dynamic model from the lateral control point of view, and the lateral controller designs. The first experimental results were shown in MOU289 (MOU313), "Lateral Control of Heavy Duty Vehicles for Automated Highway Systems", where the theoretical model was validated and calibrated to the dynamic behavior of an actual tractor-semitrailer vehicle, which was obtained and instrumented. In addition, preliminary closed-loop experiments were performed. A more comprehensive study on a large variety of control strategies was presented in MOU385 and TO4201, "Robust Lateral Control of Heavy Duty Vehicles". More specifically, three types of nonlinear and adaptive controllers for lateral control of heavy vehicles were analyzed theoretically and compared experimentally. All the research efforts mentioned above have been extremely valuable for the development of automated highway vehicles; however they assume the existence of a fully operational magnet-magnetometer scheme. To be more specific, all the results are based on the assumption that each heavy vehicle is equipped with two banks of magnetic sensors, one mounted on the front bumper and the other mounted on the rear bumper of the trailer. The road is also implanted with equally spaced magnets whose magnetic field is used to measure the vehicle's lateral deviation from the road centerline ("lane-keeping control"). Up to now, no heavy-vehicle-related report has discussed the case of vehicle lateral performance under the existence of faults. This problem is very important, since safety and reliability are the primary requirements for the success of AHS. This report addresses the problem of fault tolerant control of heavy vehicles by proposing a secondary system that implements "autonomous vehicle following" instead of "lanekeeping".

This book explores recent perspectives on type-2 fuzzy sets. Written as a tribute to Professor Jerry Mendel for his pioneering works on type-2 fuzzy sets and systems, it covers a wide range of topics, including applications to the Go game, machine learning and pattern recognition, as well as type-2 fuzzy control and intelligent systems. The book is intended as a reference guide for the type-2 fuzzy logic community, yet it aims also at other communities dealing with similar methods and applications.

Vehicle Dynamics and Control provides a comprehensive coverage of vehicle control systems and the dynamic models used in the development of these control systems. The control system applications covered in the book include cruise control, adaptive cruise control, ABS, automated lane keeping, automated highway systems, yaw stability control, engine control, passive, active and semi-active suspensions, tire-road friction coefficient estimation, rollover prevention, and hybrid electric vehicles. In developing the dynamic model for each application, an effort is made to both keep the model simple enough for control system design but at the same time rich enough to capture the essential features of the dynamics. A special effort has been made to explain the several different tire models commonly used in literature and to interpret them physically. In the second edition of the book, chapters on roll dynamics, rollover prevention and hybrid electric vehicles have been added, and the chapter on electronic stability control has been enhanced. The use of feedback control systems on automobiles is growing rapidly. This book is intended to serve as a useful resource to researchers who work on the development of such control systems, both in the automotive industry and at universities. The book can also serve as a textbook for a graduate level course on Vehicle Dynamics and Control.