Fully Train-centric System Principle for Autonomous Train Control
- 주제(키워드) Autonomous Train Control , Coordinate System Transformation , Route Composition , Route Factorization , Train-centric
- 주제(DDC) 621.39
- 발행기관 아주대학교 일반대학원
- 지도교수 Young-Jong Cho
- 발행년도 2026
- 학위수여년월 2026. 2
- 학위명 박사
- 학과 및 전공 일반대학원 컴퓨터공학과
- 실제URI http://www.dcollection.net/handler/ajou/000000035344
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
Current moving-block train control increases capacity by removing track-circuit dependencies and using bi-directional radio, but conventional designs still rely on wayside interlocking and zone/infrastructure controllers, creating mediator-centric control flows that add latency and complexity. Recent train-centric variants reduce this overhead through direct train-to-train (T2T) exchange, yet they continue to finalize track resources through a wayside controller and to store exhaustive route tables onboard, which impairs scalability and maintainability during network changes. This dissertation proposes the Autonomous Train Control System (ATCS), a fully train-centric principle that relocates route generation and track-resource ownership onto the train itself. An onboard interlocking function models the railway using a Super Balise Group (SBG) coordinate system and four canonical switch sub- graphs, enabling route factorization and composition via lightweight string-matrix operators. Generated routes are verified to satisfy simple-path and switch-safety constraints. Trains secure exclusive authority over plain-track segments and external operable resources (e.g., point machines, platform screen doors, level crossings) through an onboard resource- controller that treats such objects as mutual-exclusion resources, eliminating wayside finalization for plain track. ATCS introduces a braking model that couples the follower’s worst-case stopping distance with the leader’s best-case stopping distance and formulates an extended Movement Authority (exMA) with a routed segment, a distance limit, and a speed limit. Because the exMA distance references the leader’s inevitable braking distance, the terminal MA speed need not collapse to zero, enabling shorter safe gaps without sacrificing provable safety. Safety is argued both structurally—through route-verification rules that preclude disconnected and V-shaped switch traversals—and formally, by showing that the occupied track sets of successive trains remain topologically disconnected under bounded sensing and actuation delays. The overall routing pipeline runs in time linear in the track- database size, supporting scalable deployment. A case study on a real Korean station layout demonstrates that ATCS consistently reduces minimum headway relative to conventional and existing train-centric systems across straight, curved, and graded segments, while also improving junction and turn-back operations through overlapped route release/setting. A reliability analysis further indicates higher fleet-level robustness by removing wayside series dependencies. By generating routes from onboard track maps (rather than distributing exhaustive route tables), directly managing resources via T2T coordination, and minimizing movement authority update latency, ATCS simplifies architecture, enhances maintainability, and unlocks additional capacity from moving-block operation. These results position ATCS as a practical evolution path for CBTC/ETCS and a strong candidate for future urban and high-speed rail standards.
more목차
Chapter I Introduction 1
Chapter II Limitations of Current Moving Block Systems 5
II.A Traditional Moving Block System 5
II.B Existing Train-centric Moving Block Approaches 9
II.C Existing Interlocking Approaches 12
II.D Structural Limitation Analysis 14
II.D.1 Control Performance Constraints 14
II.D.2 Scalability and Maintenance Constraints 14
Chapter III System Architecture and Operational Framework of Fully Train-centric ATCS 16
III.A ATCS Architecture 16
III.B Definition of ATCS Track Resources 19
III.C Operational Scenarios 21
III.C.1 Mission from ATS and Route Generation 21
III.C.2 Resource Lock/Unlock 22
III.C.3 Resource Control and Movement Authority Update 23
Chapter IV Fully Train-centric Principle 24
IV.A Onboard Route Generation via Route Factorization and Composition 24
IV.A.1 Characteristics of Railway Switch 24
IV.A.2 Track Modeling with SBG Coordinates 26
IV.A.3 Graph and Matrix Formulation 28
IV.A.4 Route Factorization and Composition 33
IV.A.5 Route Verification 36
IV.A.6 Management Resource Conflicts in Overlapping Routes 39
IV.B Fully Train-centric Train Control 41
IV.B.1 Preceding Train Determination via Train-to-Train Coordinate System Transformation 41
IV.B.2 ATCS Train Braking Model 45
IV.B.3 ATCS Movement Authority 52
Chapter V Evaluation and Analysis 54
V.A Route Verification and Analysis 54
V.B Safety Analysis of ATCS Braking Model 59
V.C Minimum Headway Analysis 64
V.C.1 Minimum Line Headway Analysis 64
V.C.2 Minimum Junction Headway Analysis 78
V.C.3 Minimum Turn-Back Headway Analysis 86
V.D System Reliability Analysis 97
V.E Comparative Analysis with Existing Approaches 104
V.E.1 Operational Mechanism Comparison 104
V.E.2 Performance Comparison 105
Chapter VI Conclusion 107
Bibliography 109
