한국해양대학교

Detailed Information

Metadata Downloads

A Study on the Development of Collision Prevention Algorithm for Small Vessels Preparing the Generic Technology for MASS

DC Field Value Language
dc.contributor.advisor 박영수 -
dc.contributor.author 이명기 -
dc.date.accessioned 2022-06-23T08:57:44Z -
dc.date.available 2022-06-23T08:57:44Z -
dc.date.created 20220308093438 -
dc.date.issued 2022 -
dc.identifier.uri http://repository.kmou.ac.kr/handle/2014.oak/12847 -
dc.identifier.uri http://kmou.dcollection.net/common/orgView/200000603174 -
dc.description.abstract The objective of this study is to develop a collision prevention algorithm for small vessels and verify its performance and validity to prepare the generic technology for MASS (Maritime Autonomous Surface Ship). To develop an efficient collision prevention algorithm, the distance between two vessels, DCPA (Distance at the Closest Point of Approach), TCPA (Time to Closest Point of Approach), and PARK (Potential Assessment of Risk) model, were used, as well as WAVE (Wireless Access in Vehicular Environments) communication. PARK model is one of the maritime traffic risk assessment models. The collision prevention algorithm was developed in four stages: vessel detection, dangerous vessel identification, continuous monitoring, and alarm generation. Detect vessels within three miles and identify dangerous vessels using DCPA and TCPA. In addition, it continuously monitors based on the PARK model risk and generates a four-step alarm signal depending on the distance. Simulation experiments and scenario-based real ship experiments were performed to verify the alarm operation of the developed collision prevention algorithm. As a result, it was confirmed that when a small vessel encounters a dangerous situation, the developed algorithm generates a step-by-step alarm, and when the dangerous situation is resolved, the alarm is released. It was also confirmed that, despite frequent changes in DCPA and TCPA due to changes in the course of small vessels, alarms are generated continuously, and alarms continue in dangerous situations. The developed collision prevention algorithm was applied to the actual waterway to examine its validity. The risk of the sea was evaluated by the PARK model, ES (Environment Stress) model, and IWRAP (International Association of Marine Aids to Navigation and Lighthouse Authorities Waterway Risk Assessment Program) MKII, and this was compared with areas with frequent alarms based on the 7-day ship traffic data (V-Pass data) of the actual waterway. The waterway with a high risk coincided with the waterway with frequent alarms, whereas the waterway with a relatively low risk, despite the presence of frequent alarms, was found to be a waterway with actual traffic congestion. In addition, to reflect the actual traffic situation, the developed collision prevention algorithm was applied to the cases extracted from the V-Pass data. As a result, it was confirmed that the alarm was gradually generated based on the distance to the dangerous situation without generating a false alarm. The developed algorithm is expected to be utilized as a generic technology as a collision detection technology, which is one of the core technologies for MASS development. -
dc.description.abstract 본 연구의 목적은 자율운항선박의 기반기술을 마련하기 위한 소형 선박의 충돌 예방 알고리즘을 개발하고 개발된 알고리즘의 성능과 타당성을 검증하는 것이다. 효율적인 충돌 예방 알고리즘을 개발하기 위하여 두 선박간의 거리, DCPA (Distance at the Closest Point of Approach), TCPA (Time to the Closest Point of Approach)와 해상 교통 위험도 평가 모델인 PARK (Potential Assessment of Risk) model을 이용하였으며, WAVE (Wireless Access in Vehicular Environments) 통신을 이용하였다. 충돌 예방 알고리즘은 선박 감지, 위험 선박 식별, 지속적인 모니터링, 경보 발생의 4단계로 개발했다. 3마일 이내의 선박을 탐지하고 DCPA와 TCPA를 이용하여 위험 선박을 식별한다. 또한 PARK Model 위험도를 기반으로 지속적으로 모니터링하고 거리에 따라 4단계 경보 신호를 발생시킨다. 개발된 충돌 예방 알고리즘의 경보작동을 검증하기 위해 시뮬레이션 실험과 시나리오 기반의 실선 실험을 진행했다. 그 결과, 소형 선박이 위험한 상황으로 조우하는 경우에 개발된 알고리즘에 의하여 단계별 경보가 발생하고, 위험한 상황이 해제되면 경보도 해제되는 것을 확인했다. 또한 소형 선박의 침로 변화에 따라 DCPA, TCPA가 빈번하게 변화함에도 비교적 연속적으로 경보를 발생하고, 위험 상황에서는 경보를 지속시키는 것을 확인했다. 또한 개발된 충돌 예방 알고리즘을 실해역에 적용하여 그 타당성을 검토하고자 했다. 먼저 실해역의 7일간 선박운항데이터 (V-Pass data)를 기반으로 PARK Model, ES (Environment stress) model, IWRAP (International Association of Marine Aids to Navigation and Lighthouse Authorities Waterway Risk Assessment Program) MkII를 이용하여 해상 위험성을 평가하고, 이를 경보다발구역과 비교했다. 위험이 높게 나타난 해역과 경보다발구역이 일치하였으며, 경보가 많이 발생하였으나 비교적 위험이 낮게 평가된 구역은 실제 교통이 혼잡한 해역인 것을 확인했다. 또한 실제의 교통상황을 반영하기 위하여 실해역 선박운항데이터에서 추출한 사례에 개발된 충돌 예방 알고리즘을 적용했다. 그 결과 오경보 없이 위험한 상황까지의 거리에 따라 점진적으로 발생하는 것을 확인했다. 개발된 알고리즘은 자율운항선박 개발의 핵심 기술 중 하나인 충돌 감지 기술로 자율운항선박 개발 시에 기반기술로 활용될 수 있을 것으로 기대된다. -
dc.description.tableofcontents 1. Introduction 1 1.1 Background and Purpose 1 1.2 Scope and Method 4 2. MASS technology and previous study 7 2.1 MASS Research Trends 7 2.2 Previous study on the collision prevention algorithm 11 2.2.1 Comparison of maneuvering performance of large and small vessels 11 2.2.2 Previous study on collision prevention algorithms for large vessels 14 2.2.3 Previous study on collision prevention algorithms for small vessels 16 2.3 Semi-conclusions 18 3. Development of a collision prevention algorithm for small vessels 20 3.1 Application of WAVE communication to develop collision prevention algorithms 20 3.1.1 Overview of WAVE communication 20 3.1.2 Application of WAVE communication to the sea 21 3.2 Overview of the development of collision prevention algorithms 24 3.3 Development of Collision Prevention Algorithms 27 3.3.1 Setting of vessel detection range 27 3.3.2 Primary criteria (DCPA, TCPA) 29 3.3.3 Secondary criteria (PARK model) 34 3.3.4 Alarm occurrence range 41 3.3.5 Final algorithm 44 4. Verification of the collision prevention algorithm through empirical experiments 47 4.1 Verification through simulation experiments 47 4.1.1 Overview of the simulation experiment 47 4.1.2 Result of simulation experiment 48 4.2 Verification through real ship experiments 51 4.2.1 1:1 real ship experiment 51 4.2.2 1:2 real ship experiment 80 4.3 Semi-conclusions 104 5. Evaluation of feasibility through application of waterways 106 5.1 Data interpolation for waterway application 106 5.2 Evaluation based on maritime traffic risk assessment model 115 5.2.1 Overview of maritime traffic risk assessment model 115 5.2.2 Chart-based comparison 118 5.2.3 Leg-based comparison 124 5.3 Evaluation through application of cases to waterways 130 5.3.1 Overview of evaluation through application of cases in waterways 130 5.3.2 Evaluation results from the application of cases in waterways 130 5.4 Utilization of MASS generic technology 139 6. Conclusion 141 -
dc.format.extent 169 -
dc.language eng -
dc.publisher 한국해양대학교 해양과학기술전문대학원 -
dc.rights 한국해양대학교 논문은 저작권에 의해 보호받습니다. -
dc.title A Study on the Development of Collision Prevention Algorithm for Small Vessels Preparing the Generic Technology for MASS -
dc.title.alternative 자율운항선박 기반기술마련을 위한 소형 선박의 충돌 예방 알고리즘 개발에 관한 연구 -
dc.type Dissertation -
dc.date.awarded 2022. 2 -
dc.embargo.liftdate 2022-03-08 -
dc.contributor.alternativeName Myoung-ki, Lee -
dc.contributor.department 해양과학기술전문대학원 해양과학기술융합학과 -
dc.contributor.affiliation 한국해양대학교 해양과학기술전문대학원 해양과학기술융합학과 -
dc.description.degree Doctor -
dc.identifier.bibliographicCitation [1]이명기, “A Study on the Development of Collision Prevention Algorithm for Small Vessels Preparing the Generic Technology for MASS,” 한국해양대학교 해양과학기술전문대학원, 2022. -
dc.subject.keyword 자율운항선박 -
dc.subject.keyword 소형 선박 -
dc.subject.keyword 충돌 예방 알고리즘 -
dc.subject.keyword 해상 위험 평가 모델 -
dc.identifier.holdings 000000001979▲200000002763▲200000603174▲ -
Appears in Collections:
기타 > 기타
Files in This Item:
There are no files associated with this item.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse