Jeonghong Park

Precision navigation and mapping under bridges with an unmanned surface vehicle

Navigation relative to the surrounding physical structures and obstacles is an important capability for safe vehicle operation. This capability is particularly useful for unmanned surface vehicles (USVs) operating near large structures such as bridges, waterside buildings, towers and cranes, where global positioning system signals are restricted or unavailable due to the line-of-sight restrictions. This study proposes a high-precision navigation technique using dead-reckoning sensors and lidars, which enables building a parameterized map of artificial bridge structures and estimating the vehicle’s position relative to the parameterized map simultaneously. Also, three-dimensional reconstruction of the surrounding structures is carried out by fusing camera and lidar measurements for realistic 3D visual mapping which may facilitate automated surveys and inspection of structural safety. Field experiments using a newly developed USV system in a real-world bridge environment were performed to verify and demonstrate the performance of the proposed navigation and mapping algorithms. The field test results are presented and discussed in this paper.

Development of an Unmanned Surface Vehicle System for the 2014 Maritime RobotX Challenge

This paper addresses the development of an unmanned surface vehicle (USV) system by Team Angry-Nerds from KAIST for the inaugural Maritime RobotX Challenge competition, which was held on October 20-26, 2014, in Marina Bay, Singapore. The USV hardware was developed on a catamaran platform by integrating various system components, including propulsion, sensors, computer, power, and emergency systems. The competition comprised five mission tasks: 1) navigation and control, 2) underwater search and report, 3) automatic docking, 4) buoy search and observation, and 5) obstacle detection and avoidance. Onboard intelligence was a key factor for all of the mission tasks which needed to be performed autonomously with no human intervention. Software algorithms for vehicle autonomy were developed, and executable computer codes were implemented and integrated with the developed USV hardware system. This paper describes the development process of the USV system and its application to the competition mission tasks.

Predictive Evaluation of Ship Collision Risk Using the Concept of Probability Flow

This paper presents a semianalytical approach for evaluating the collision risk between two moving surface ships. The concept of probability flow is introduced to develop an analytically sound problem formulation, which allows for an accurate estimation of collision probability considering time-varying ship trajectory uncertainties. For efficient computation, the flow of collision probability is separated into diffusion and drift components. These two probability components are combined to obtain the instantaneous collision probability, and this instantaneous probability is integrated in time to quantify the expected risk of collision. To demonstrate the feasibility of the proposed approach, traffic simulations are performed for several representative maritime traffic scenarios and the obtained simulation results are discussed.

Autonomous collision avoidance for unmanned surface ships using onboard monocular vision

This study presents the development of vision-based techniques for autonomous collision avoidance by an unmanned surface ship using an onboard monocular camera. In order to determine the initiation of an evasive maneuver, the range and bearing measurements of each target traffic ship with respect to the observer (e.g., own ship) need to be provided for trajectory estimation. A tracking estimator is used to estimate the target ship trajectory in the framework of bearings-only tracking based on the extended Kalman filter (EKF) algorithm with a Continuous White Noise Acceleration (CWNA) model. To enhance the observability of the tracking filter, the vertical pixel distance from the water horizon to each target ship is used as a range measurement. When the estimated separation distance between the target and own ships is less than a predefined minimum separation, the own ship alters its heading angle to avoid an imminent collision following the standard marine traffic rules. Field experiment results are presented and discussed to demonstrate the feasibility and validity of the proposed approach.

Autonomous detection and tracking of a surface ship using onboard monocular vision

This study addresses autonomous detection and tracking of a surface ship using a monocular camera mounted on an unmanned surface vehicle (USV). Automatic feature extraction and tracking filter algorithms are used for vision-based detection and tracking in real-time. For target tracking, the bearing and range to the target ship with respect to the own ship are obtained by computer vision techniques. The pixel distance from the horizon to the target in camera images is used to extract the range information, and a probing maneuver procedure is designed to enhance observability of the target tracking filter. The feasibility and performance of the proposed tracking approach are validated through field experiments with two USVs.

Development of a Hover-capable AUV System for In-water Visual Inspection via Image Mosaicking

Recently, UUVs (unmanned underwater vehicles) have increasingly been applied in various science and engineering applications. In-water inspection, which used to be performed by human divers, is a potential application for UUVs. In particular, the operational safety and performance of in-water inspection missions can be greatly improved by using an underwater robotic vehicle. The capabilities of hovering maneuvers and automatic image mosaicking are essential for autonomous underwater visual inspection. This paper presents the development of a hover-capable autonomous underwater vehicle system for autonomous in-water inspection, which includes both a hardware platform and operational software algorithms. Some results from an experiment in a model basin are presented to demonstrate the feasibility of the developed system and algorithms. Received 9 November 2015, revised 13 May 2016, accepted 24 June 2016 Corresponding author Jinwhan Kim: +82-042-350-1519, jinwhan@kaist.ac.kr ◯c 2016, The Korean Society of Ocean Engineers It is noted that this paper is revised edition based on proceedings of KSOE 2015 in Daejeon. 1. 서 론 해양 플랜트, 댐, 선체 등과 같은 해양 구조물들의 관리 및 유 지보수를 위해서는 수중 육안 검사가 필요하다. 이러한 수중 검 사는 일반적으로 유인 잠수부에 의해 수행되는데, 고난이도 고 위험 작업의 특성상 비용 대비 효율이 크게 떨어지며 안정성의 문제로 충분하고 세밀한 검사가 어려워 검사 결과의 정확성과 신뢰도에 명확한 한계가 존재한다. 최근 해양 로봇공학 분야가 지속적으로 발전함에 따라 임무 특성에 맞는 여러 형태의 무인 수중 운동체들이 활용되고 있다. 특히 ROV(Remotely-operated vehicle)는 수중 검사 및 탐사를 수행하기 위해 가장 널리 활용되어 온 수중 운동체의 한 형태 이나, 지상의 조작 콘솔에 연결된 유선 작업의 특성상 그 운용 효율과 범위의 제약이 크다. 특히 제한된 시계 조건에서 ROV 가 운용될 경우, 전송되는 영상을 바탕으로 운용자의 시각에 전 적으로 의존해야 하므로 전반적인 작업 효율이 운용자의 숙련 도에 크게 좌우되며 시간에 따른 피로 증가로 집중도와 효율이 떨어지기 때문에 장시간 운용이 어렵다. 한편, 일반적인 어뢰형의 AUV(Autonomous underwater vehicle) 는 그 외형 상 저항 및 추진에 유리하고 에너지 효율이 높아 운 용 반경이 넓다는 장점이 있으나 운동성의 제약으로 제자리 유 영이나 임의 방향으로 자유로운 이동이 불가능하다. 이러한 이 유로, 해양 물리량 측정 등 단순한 운용 외에 복잡한 임무 수행 에는 적합하지 못하다. 본 연구에서는 기존의 유인 작업 또는 ROV를 사용한 수중 구조 물 검사 임무의 한계를 극복하고자 제한된 운동성을 갖는 기존의 어뢰형 AUV가 아닌 제자리에서 유영이 가능한 H-AUV (Hover-capable AUV) 시스템의 하드웨어 플랫폼 및 운용 알고리 즘의 개발을 목표로 한다. 2장에서는 H-AUV 플랫폼의 하드웨어 설계 및 구현에 대한 내용을 기술하고, 3장에서 카메라 영상 기반

Three-dimensional reconstruction of bridge structures above the waterline with an unmanned surface vehicle

This study addresses three-dimensional (3D) reconstruction of bridge structures over water by fusing sensor measurements from inertial sensors, cameras and lidars mounted on an unmanned surface vehicle (USV). While the accurate navigation capability is strongly required for successful 3D reconstruction, global positioning system (GPS) signals which are essential for accurate navigation are severely deteriorated near the bridge structures or almost completely blocked underneath the bridge decks. In this study, a parameterized feature map is introduced by augmenting the map state with the geometric parameters of the detected bridge piers, and relative navigation is performed with respect to this map in the framework of simultaneous localization and mapping (SLAM). This parameterized SLAM approach allows for high-precision navigation and mapping with no need of GPS fixes. The feasibility of the proposed algorithm was demonstrated through field experiments.