主题：Underwater Target detection and Tracking
时间：2024年10月14下午14：30-16：40
地点：2教南228会议室
主持人：曹九稳教授
报告人简介：
Jee Woong Choi was born in Seoul, Korea, in 1970. He received his B.S. degree in Earth and Marine Science from Hanyang University, ERICA, Korea, in 1996, and M.S. and Ph.D. degrees in Underwater Acoustics from Hanyang University, ERICA, Korea, in 1998 and 2002, respectively. Since 2007, he has been a professor in the department of Marine Science and Convergence Engineering at Hanyang University, ERICA, Korea. From 2002 to 2007, he was an associate researcher at the Applied Physics Laboratory in University of Washington (APL-UW), Seattle, USA. From 2013 to 2014, he was a visiting professor at the Applied Physics Laboratory in University of Washington (APL-UW), Seattle, USA. Since 2022, He has been a Director of LINC3.0 (Leaders in IndustryUniversity Cooperation) Project Group of Hanyang University ERICA, Korea and since 2023, He has been a Vice-president for Education of Hanyang University ERICA, Korea. In 2011, he was a Chair of Organizing Committee of Pacific Rim Underwater Acoustics Conference 2011 (PRUAC 2011). Since 2018, he has been a 2nd Vice-Chairman of Western Pacific Commission for Acoustics (WESPAC). Since 2022, he has been a Vice-president of The Acoustical Society of Korea Also, he is a Fellow of the Acoustical Society of America Since 2020.
报告题目：Underwater Acoustics and its application to Target detection & Tracking
摘要：In this presentation, we will introduce the property of underwater acoustic propagation for the purpose of target detection and tracking and introduce various noises that exist underwater. Next, we introduce a cutting-edge technique for range estimation, which is called array invariant, using acoustic data. The ROTMA results, which apply this range estimation technique to real ocean experimental data, will be discussed in the presentation. Lastly, machine learning approaches for range estimation of underwater targets will be introduced.
Taek Lyul Song 教授
Tae
k Lyul Song was born in Andong, South Korea, in 1952. He received the B.S. degree in precision Nuclear Engineering from Seoul National University, Seoul, South Korea, in 1974, and the M.S. and Ph.D. degrees in mechanical engineering from university of Texas at Austin, Texas, America, in 1981 and 1983, respectively. From 1984 to 1995, he was a Principal Research Engineer in the Agency for Defense Development (ADD). Since 1995, he has been a professor in the Division of Electrical Engineering, Hanyang University ERICA, Ansan, South Korea. From 2018 to 2021, He has been a chair professor in school of automation, Hangzhou Dianzi University.
报告题目：Underwater Target Tracking in Passive SONAR System
摘要：The marine environment can impede accurate detection due to various noises. These combined impediments lower the signal-to-noise ratio (SNR) of the bearing measurement and increase the likelihood of undetected target intervals over entire signal processing time. When performing Target Motion Analysis (TMA) by using these bearing measurements, noises can cause discontinuities in the target trajectory, thereby degrade TMA performance. In this presentation, we introduce a technique for estimating the bearing measurement trajectory on a Bearing Time Records (BTR) diagram in low SNR environments. The proposed method called the integrated highest probability data association - amplitude information (IHPDA-AI) algorithm applies probabilistic distributions of signal amplitudes of the target and clutter to evaluate the data association probability calculation unlike the existing HPDA (Highest Probability Data Association) algorithm. Passive acoustics signals generated by a target used for oceanic environment analysis cause a lot of problems due to multiple paths and Doppler effects. Multiple detections caused by multiple paths need to be identified to avoid a mismatch between the true paths and the path models for target tracking. In this presentation, the MP-IPDA algorithm which was applied to multi-path (MP) tracking for Over-The-Horizon Radar is modified for target motion analysis (TMA) in a shallow oceanic environment where the speed of sound in almost constant. The MP-IPDA algorithm is further modified to an alternative algorithm called MP-IPDA* to reduce computational burden for enhancing efficiency. The MP- IPDA* algorithm is more effective as the number of multiple paths increases. A simulation study shows that the MP-IPDA* algorithm in quite effective for TMA even in the environment with very low target detection probability.