서브메뉴
검색
Optimal Vehicle Control Under Friction Uncertainty - From Driver Assistance to Drift Transitions
Optimal Vehicle Control Under Friction Uncertainty - From Driver Assistance to Drift Transitions
Detailed Information
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211152118
- ISBN
- 9798384341536
- DDC
- 600
- 서명/저자
- Optimal Vehicle Control Under Friction Uncertainty - From Driver Assistance to Drift Transitions
- 발행사항
- [Sl] : Stanford University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 123 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
- 주기사항
- Includes supplementary digital materials.
- 주기사항
- Advisor: Gerdes, J.
- 학위논문주기
- Thesis (Ph.D.)--Stanford University, 2024.
- 초록/해제
- 요약Automobiles play a significant role in the lives of people worldwide. They provide a relatively accessible means to transportation, impacting a person's economic and social opportunities. Unfortunately, operating automobiles carries a risk. The World Health Organization estimates that 1.2 million people were killed in automobile crashes in 2019 alone. After decades of progress reducing vehicle fatalities in the United States, the National Highway Traffic Safety Administration reports that 2021 showed the largest increase in vehicle related deaths since they began recording them.Following the passage of the National Highway and Motor Vehicle Safety Act in the 1960's, automotive researchers and manufactures have created many systems designed to reduce the risk of driving. These range from passive systems like seat belts to active systems like the ABS and ESC. The technologies played a crucial role in decreasing traffic fatalities in past decades. More advanced technologies are entering the market with the ability to help prevent rear end crashes and stop drivers from departing the driving lane. Despite these advances, state of the art production systems perform in a fairly confined ODD. Each system may only have authority over one or two actuation systems and can only sense a specific type of emergency. For the greatest impact in righting the trend in vehicle fatalities, the next generation of ADAS must be able to reason more generally about the safety of the vehicle in the context of its current operating state and environment.This dissertation contributes three novel systems in furtherance of an general ADAS. Chapter 2 contributes an NMPC formulation to keep a vehicle safely on the roadway while matching a driver's commands. It builds on a body of previous work beginning with linear MPC formulations for stability control and environmental safety. These methods lack the ability to influence vehicle speed which becomes critically important near the vehicles handling limits. Later work introduces MPC controlling both steering and tractive forces. We improve on previous work by improving the objective function to better match drivers commands, introducing a concept of safe reference speed to improve robustness to unmodeled disturbances, and a method to minimize delays between driver commands and vehicle actuator movements.Chapter 3 contributes an NMPC that blends two different model fidelities to lower computational costs and allows a driver to explore the limits of vehicle friction. This formulation builds on Chapter 2 leveraging experience gained in a challenging ice test track on a frozen lake. This environment challenged our controllers with highly variable tire road friction as the tires traversed patches of polished ice and packed snow. Previous approaches proved too conservative for a drivers comfort and safety. What's more, the approach given in Chapter 2 lacked the long prediction horizon needed to set up slowly evolving ice maneuvers. To address these challenges we developed a controller that allows the driver to explore past the modeled limits of friction in a tunable manner. We employ serially cascaded models over the prediction horizon to lower computation costs while increasing our predictive ability. This inflection provides an opportunity to link the change in objective function (from matching the driver to tracking a safe path) to a dynamically meaningful metric.All of the work up to this point prioritizes stabilizing the vehicle (bounding or minimizing the angle between its velocity and heading). Work in the field of autonomous drifting illustrates that we are not limited to this choice. Chapter 4 contributes an optimal control approach to planning and executing drift transitions which can adjust to goal state changes in real-time. By operating away from the stable handling region, we enable maneuvers unachievable by the previous controllers. These techniques could empower future ADAS to better avoid accidents and improve overall safety.
- 일반주제명
- Friction
- 일반주제명
- Tires
- 일반주제명
- Mechanical engineering
- 기타저자
- Stanford University.
- 기본자료저록
- Dissertations Abstracts International. 86-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017162970
■00520250211152118
■006m o d
■007cr#unu||||||||
■020 ▼a9798384341536
■035 ▼a(MiAaPQ)AAI31460343
■035 ▼a(MiAaPQ)Stanfordqz036kd2292
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a600
■1001 ▼aTalbot, John Andrew.
■24510▼aOptimal Vehicle Control Under Friction Uncertainty - From Driver Assistance to Drift Transitions
■260 ▼a[Sl]▼bStanford University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a123 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-03, Section: B.
■500 ▼aIncludes supplementary digital materials.
■500 ▼aAdvisor: Gerdes, J.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2024.
■520 ▼aAutomobiles play a significant role in the lives of people worldwide. They provide a relatively accessible means to transportation, impacting a person's economic and social opportunities. Unfortunately, operating automobiles carries a risk. The World Health Organization estimates that 1.2 million people were killed in automobile crashes in 2019 alone. After decades of progress reducing vehicle fatalities in the United States, the National Highway Traffic Safety Administration reports that 2021 showed the largest increase in vehicle related deaths since they began recording them.Following the passage of the National Highway and Motor Vehicle Safety Act in the 1960's, automotive researchers and manufactures have created many systems designed to reduce the risk of driving. These range from passive systems like seat belts to active systems like the ABS and ESC. The technologies played a crucial role in decreasing traffic fatalities in past decades. More advanced technologies are entering the market with the ability to help prevent rear end crashes and stop drivers from departing the driving lane. Despite these advances, state of the art production systems perform in a fairly confined ODD. Each system may only have authority over one or two actuation systems and can only sense a specific type of emergency. For the greatest impact in righting the trend in vehicle fatalities, the next generation of ADAS must be able to reason more generally about the safety of the vehicle in the context of its current operating state and environment.This dissertation contributes three novel systems in furtherance of an general ADAS. Chapter 2 contributes an NMPC formulation to keep a vehicle safely on the roadway while matching a driver's commands. It builds on a body of previous work beginning with linear MPC formulations for stability control and environmental safety. These methods lack the ability to influence vehicle speed which becomes critically important near the vehicles handling limits. Later work introduces MPC controlling both steering and tractive forces. We improve on previous work by improving the objective function to better match drivers commands, introducing a concept of safe reference speed to improve robustness to unmodeled disturbances, and a method to minimize delays between driver commands and vehicle actuator movements.Chapter 3 contributes an NMPC that blends two different model fidelities to lower computational costs and allows a driver to explore the limits of vehicle friction. This formulation builds on Chapter 2 leveraging experience gained in a challenging ice test track on a frozen lake. This environment challenged our controllers with highly variable tire road friction as the tires traversed patches of polished ice and packed snow. Previous approaches proved too conservative for a drivers comfort and safety. What's more, the approach given in Chapter 2 lacked the long prediction horizon needed to set up slowly evolving ice maneuvers. To address these challenges we developed a controller that allows the driver to explore past the modeled limits of friction in a tunable manner. We employ serially cascaded models over the prediction horizon to lower computation costs while increasing our predictive ability. This inflection provides an opportunity to link the change in objective function (from matching the driver to tracking a safe path) to a dynamically meaningful metric.All of the work up to this point prioritizes stabilizing the vehicle (bounding or minimizing the angle between its velocity and heading). Work in the field of autonomous drifting illustrates that we are not limited to this choice. Chapter 4 contributes an optimal control approach to planning and executing drift transitions which can adjust to goal state changes in real-time. By operating away from the stable handling region, we enable maneuvers unachievable by the previous controllers. These techniques could empower future ADAS to better avoid accidents and improve overall safety.
■590 ▼aSchool code: 0212.
■650 4▼aFriction
■650 4▼aTires
■650 4▼aMechanical engineering
■690 ▼a0548
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g86-03B.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162970▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
Preview
Export
ChatGPT Discussion
AI Recommended Related Books
Подробнее информация.
- Бронирование
- не существует
- моя папка
- Первый запрос зрения
- Non-Book Loan Application
- Nighttime Book Loan Application
Available after logging in.


