자료유형  

 학위논문

Control Number  

0017164569

International Standard Book Number  

9798384046059

Dewey Decimal Classification Number  

629.2

Main Entry-Personal Name  

Alan, Anil.

Publication, Distribution, etc. (Imprint  

[S.l.] : University of Michigan., 2024

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Physical Description  

165 p.

General Note  

Source: Dissertations Abstracts International, Volume: 86-03, Section: B.

General Note  

Advisor: Orosz, Gabor.

Dissertation Note  

Thesis (Ph.D.)--University of Michigan, 2024.

Summary, Etc.  

요약As safety is an ever more pressing requirement for modern control systems deployed into real-world environments, there has been an increase in demand to design controllers with mathematically rigorous safety guarantees. Control barrier functions (CBFs) provide one such framework, which takes our understanding of the system into consideration, and utilizes theoretical properties of sets to avoid unsafe states. Its intuitive nature leads to practical solutions for designing safe-by-design controllers, and their efficacy has been demonstrated experimentally in real-world conditions. A limitation inherent in the CBF framework is its dependence on the accuracy of the model capturing the underlying dynamics of the system, which consequently raises the question of resilience against model imperfections. The main contribution of this dissertation is a methodical investigation of a range of viable solutions to improve the resilience of CBF-based controllers. We study solutions in a categorical approach, and present our contributions on improving multiple aspects such as performance and conservativeness.The theoretical investigation is introduced in two main categories: input-to-state safety (ISSf) and robust control barrier functions (RCBFs). While methods in the former group looks to treat uncertainties to a more manageable level, methods in the latter group aims a complete cancellation approach. As a result, safety guarantees provided by these approaches land on a spectrum ranging from arbitrarily small (graceful) safety degradation to absolute robust safety. This dissertation delves into both approaches with a scope on controller design, where we focus on maintaining the beneficial properties of the CBF framework while improving the robustness. Another contribution of this dissertation is on the application front, where we utilize practical examples to support the theoretical results. In particular, a case study of designing safe and energy-efficient controllers for an autonomous truck is introduced as the main application. Hard-to-model nature of intricate relationships between complex subsystems makes the autonomous truck an ideal platform to evaluate the benefits of the CBF-based controllers. We deploy controllers (with and without the robustness feature) on a full-scale truck, and evaluate the theoretical guarantees experimentally in challenging conditions such as emergency brake. 

Subject Added Entry-Topical Term  

Automotive engineering.

Subject Added Entry-Topical Term  

Mechanical engineering.

Subject Added Entry-Topical Term  

Robotics.

Subject Added Entry-Topical Term  

Systems science.

Index Term-Uncontrolled  

Safety-critical control

Index Term-Uncontrolled  

Control barrier functions

Index Term-Uncontrolled  

Robust control

Index Term-Uncontrolled  

Autonomous vehicles

Index Term-Uncontrolled  

Conservativeness

Added Entry-Corporate Name  

University of Michigan Mechanical Engineering

Host Item Entry  

Dissertations Abstracts International. 86-03B.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:656724