서브메뉴
검색
A Topology-Aware Hierarchical Planning and Control Framework for Highly Dynamical Autonomous Ground Vehicles
A Topology-Aware Hierarchical Planning and Control Framework for Highly Dynamical Autonomous Ground Vehicles
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105243
- ISBN
- 9798291569511
- DDC
- 629.2
- 저자명
- Shen, Congkai.
- 서명/저자
- A Topology-Aware Hierarchical Planning and Control Framework for Highly Dynamical Autonomous Ground Vehicles
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 221 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-02, Section: B.
- 주기사항
- Advisor: Ersal, Tulga.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약Autonomous ground vehicles (AGVs) are critical for applications such as military operations, mining, and planetary exploration, where off-road autonomy is essential. These environments are characterized by unstructured terrain with complex, non-planar topologies that invalidate assumptions commonly used in on-road autonomy. Current planning and control strategies restrict AGVs operations in such conditions because terrain topology significantly influences vehicle dynamics, posing mobility challenges and safety risks. The lack of terrain topology consideration in planning and control algorithms reduces mobility and often results in trajectories that are difficult or infeasible for AGVs to execute-compromising both performance and safety. Therefore, the overarching goal of this dissertation is to develop, implement, and evaluate a topology-aware hierarchical planning and control framework for highly dynamic AGVs operating in off-road environments. The algorithm is designed to plan on rigid, relatively smooth, yet non-planar terrain topology, capturing the key dynamics of off-road motion while enabling computationally efficient planning and control. This dissertation has three key objectives. The first is to develop an efficient global planner that incorporates nonholonomic and dynamic constraints to generate feasible trajectories in unstructured environments and avoid dead ends. The second is to design a real-time topology-aware local planning and control algorithm that addresses limitations in existing methods, which either oversimplify terrain effects, are too complex for real-time use, or rely on restrictive assumptions. The third is to establish a hierarchical framework that integrates the global and local algorithms-combining long-term predictive capabilities with real-time optimization to improve both mobility and safety. The proposed framework is evaluated through high-fidelity simulations and real-world experiments. The simulations examine the performance across a variety of terrain scenarios, while physical testing is conducted using a custom-built 1/5-scale remote-controlled (RC) platform to assess feasibility under real-world conditions.The original contributions of this dissertation are as follows:1. Development of an efficient global trajectory planner for highly dynamic, nonholonomic AGVs navigating non-planar terrain. 2. Design of a real-time topology-aware local planning and control algorithm for dynamic off-road operation. 3. Integration of global and local planners into a hierarchical framework that combines long-horizon planning with short-horizon planning and control to enhance mobility and safety in complex environments.
- 일반주제명
- Automotive engineering
- 일반주제명
- Mechanical engineering
- 일반주제명
- Robotics
- 일반주제명
- Computer engineering
- 키워드
- Motion planning
- 키워드
- Optimal control
- 키워드
- Vehicle dynamics
- 기타저자
- University of Michigan Mechanical Engineering
- 기본자료저록
- Dissertations Abstracts International. 87-02B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359968
■00520260202105243
■006m o d
■007cr#unu||||||||
■020 ▼a9798291569511
■035 ▼a(MiAaPQ)AAI32272022
■035 ▼a(MiAaPQ)umichrackham006255
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a629.2
■1001 ▼aShen, Congkai.
■24512▼aA Topology-Aware Hierarchical Planning and Control Framework for Highly Dynamical Autonomous Ground Vehicles
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a221 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-02, Section: B.
■500 ▼aAdvisor: Ersal, Tulga.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aAutonomous ground vehicles (AGVs) are critical for applications such as military operations, mining, and planetary exploration, where off-road autonomy is essential. These environments are characterized by unstructured terrain with complex, non-planar topologies that invalidate assumptions commonly used in on-road autonomy. Current planning and control strategies restrict AGVs operations in such conditions because terrain topology significantly influences vehicle dynamics, posing mobility challenges and safety risks. The lack of terrain topology consideration in planning and control algorithms reduces mobility and often results in trajectories that are difficult or infeasible for AGVs to execute-compromising both performance and safety. Therefore, the overarching goal of this dissertation is to develop, implement, and evaluate a topology-aware hierarchical planning and control framework for highly dynamic AGVs operating in off-road environments. The algorithm is designed to plan on rigid, relatively smooth, yet non-planar terrain topology, capturing the key dynamics of off-road motion while enabling computationally efficient planning and control. This dissertation has three key objectives. The first is to develop an efficient global planner that incorporates nonholonomic and dynamic constraints to generate feasible trajectories in unstructured environments and avoid dead ends. The second is to design a real-time topology-aware local planning and control algorithm that addresses limitations in existing methods, which either oversimplify terrain effects, are too complex for real-time use, or rely on restrictive assumptions. The third is to establish a hierarchical framework that integrates the global and local algorithms-combining long-term predictive capabilities with real-time optimization to improve both mobility and safety. The proposed framework is evaluated through high-fidelity simulations and real-world experiments. The simulations examine the performance across a variety of terrain scenarios, while physical testing is conducted using a custom-built 1/5-scale remote-controlled (RC) platform to assess feasibility under real-world conditions.The original contributions of this dissertation are as follows:1. Development of an efficient global trajectory planner for highly dynamic, nonholonomic AGVs navigating non-planar terrain. 2. Design of a real-time topology-aware local planning and control algorithm for dynamic off-road operation. 3. Integration of global and local planners into a hierarchical framework that combines long-horizon planning with short-horizon planning and control to enhance mobility and safety in complex environments.
■590 ▼aSchool code: 0127.
■650 4▼aAutomotive engineering
■650 4▼aMechanical engineering
■650 4▼aRobotics
■650 4▼aComputer engineering
■653 ▼aAutonomous ground vehicles
■653 ▼aMotion planning
■653 ▼aOptimal control
■653 ▼aVehicle dynamics
■653 ▼aControl algorithm
■690 ▼a0548
■690 ▼a0540
■690 ▼a0771
■690 ▼a0464
■71020▼aUniversity of Michigan▼bMechanical Engineering.
■7730 ▼tDissertations Abstracts International▼g87-02B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359968▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
