Modern maritime operations rely on diverse network components, increasing cybersecurity risks. While security solutions like Suricata generate extensive alert logs, ships often operate without dedicated personnel, requiring general crew members to review and respond alerts. This challenge is exacerbated when vessels are at sea, delaying threat mitigation due limited external support. We propose an Adaptive Threat Intelligence Response Recommendation System (ATIRS), a small language model...

In this study, we propose an autonomous underwater vehicle (AUV)-based multi-directional scanning method of objects using forward scan sonar (FSS). Recently, a was developed that can generate 3-D point cloud object FSS. However, the data comprised sparse and noisy characteristics made it difficult for recognition. Another limitation absence back side surface information object. These limitations degraded results reconstruction. We strategy to improve reconstruction where AUV determines...

Based on the paradigm shift in modern warfare, ground forces conduct pilot operations of wireless unmanned maneuvering systems, such as tactical drones, form manned and cooperative tactics after deploying relevant systems battlefield. However, security considerations for are limited to scope using only existing end-to-end encryption public key authentication modules, no defense strategy actively respond specialized cyber-electronic warfare threats has been officially established. To...

Light detection and ranging (LiDAR) sensors can create high-quality scans of an environment. However, LiDAR point clouds are affected by harsh weather conditions since airborne particles easily detected. In literature, conventional filtering artificial intelligence (AI) methods have been used to detect, remove, particles. this paper, a convolutional neural network (CNN) model was classify dust through voxel-based approach. The CNN compared several methods, where the results show that filter...

This paper focuses on the real-time obstacle avoidance and safe navigation of autonomous ground vehicles (AGVs). It introduces Selective MPC-PF-PSO algorithm, which includes model predictive control (MPC), Artificial Potential Fields (APFs), particle swarm optimization (PSO). approach involves defining multiple sets coefficients for adaptability to surrounding environment. The simulation results demonstrate that algorithm is appropriate generating paths. was implemented ROS platform using...

In this paper, we propose an underwater object scanning scheme using AUV and imaging sonar. The 3D point cloud generation method sonar has vulnerabilities such as the undesired slope of front surface, sparsity data, absence side back data. To solve these problems, present a multi-view instead single scan. efficiently perform multiple scans without excessive time-consuming, unit vector next path was selected by maximizing area beam reflection orthogonality with previous paths. verify validity...

We proposed an autonomous trajectory backtracking method using forward-looking imaging sonar. Additionally, we also dealt with a Fourier-based sonar image processing for the backtracking. suggested algorithm to estimate translational shifts and rotation angle between two images. By feeding back estimated data, AUV can compensate drift error of dead-reckoning. To verify these algorithms, used field data obtained by hovering-type `Cyclops'. verified accuracy tolerance performing experiments.

An imaging sonar loses information of elevation angle while a mapping process. To overcome this limitation, the motion autonomous underwater vehicle (AUV) can be used to obtain 3D using sonar. In paper, we propose two-stage strategy for accurately generating maps based on an It consists searching and scanning stage. stage, multi-directional is performed object. process point cloud data obtained by scanning, polygonal approximation method. This method reduces uncertainty extracting...

A method for automatic detection or recognition using high-frequency forward-looking imaging sonar is proposed. The includes segmentation of an underwater object's echo shape its acoustic shadow shape. For accuracy segmentation, some constraints were assumed on the ground and sonar's pose. separated was compared to simulated reference shapes know orientation identity by matching that use context concept. An experiment conducted in a wave tank validated qualitatively. can be applied AUV's...

The large test data volume and power consumption are major problems in testing system-on-a-chip (SoC) which is a key component of today's embedded system. To reduce the application time from an automatic equipment (ATE), new compression technique proposed this paper. Don't-cares pre-computed cube set assigned to consumption. Then, fully specified low-power transformed improve efficiency by neighboring bit-wise exclusive-or technique. Finally, converted compressed time.

UAM (Urban Air Mobility) is future aerial mobility for passengers and cargo, usually based on eVTOLs (electric vertical take-off landing aircrafts). communications a key enabling technology of systems services, to ensure efficient safe navigation provide network access passengers. Traditional terrestrial networks, however, are not reliable fulfill the requirements since they have been designed support airborne mobile users in three-dimensional space. In meantime, 6G considered as good match...

This poster presents an efficient secure scan design based on a fake key to protect secret from scan-based side channel attack. technique targeted for SoC embedding Advanced Encryption Standard (AES) core can be adopted without requiring any modification the functional body of IP core, thus overheads area, timing, and power are negligible while preserving compatibility with IEEE1149.1 standard.

This paper proposes a method to recognize underwater target objects and estimate their yaw angle using an imaging sonar. First, light sonar simulator generated template images of the from various viewing angles. Next, generative adversarial network predicted semantic map by segmenting real image for reliable recognition. Then, matching identifies object its angle. We verified proposed installing in indoor water tank. The can provide relative pose information sensing platforms which is useful...

This paper proposes a method to calibrate axis misalignment of underwater electric field sensors (EFS) for accurate signal measurements robots. An EFS measured known and the values were compared their theoretical approximate sensor. The approximation was optimized by using Levenberg's optimization algorithm. algorithm searched that minimized difference between values. After optimization, with could be calibrated. proposed tested in computer simulations. results show fields. calibration...

We describe the developed hovering-type AUV called `Cyclops' and sea trial for ocean environment mission. The system was optimally designed mission; high maneuverability, versatile sensor configuration, modular type HW/SW system, unique structure multi-sensor management control of data flow. conducted to estimate performance system. In mission, garbage found by high-resolution imaging sonar in a turbid environment. task, various such as chemical were collected along with trajectory Cyclops.

최근 몇 년 동안 무인비행체는 다양한 산업 분야에서 널리 사용되고 있다. 그러나, 무인비행체에 대한 의존도가 급격하게 높아짐에 따라 무인비행체의 보안과 안전에 우려가 커지고 현재 제어권을 탈취하거나 웹 애플리케이 션에서 무인비행체와 통신할 수 있는 권한을 탈취하는 등 취약점들이 공개되고 하지만, 관 련된 연구가 많이 부족한 실정이다. 따라서 본 논문에서는 실제 환경과 유사한 HITL 시뮬레이션 환경에서 패킷 데이터를 수집하여 데이터가 정상 데이터인지 비정상 판단하는 연구를 진행하였다. 또한, 모 델링 과정에서 Computation Cost를 줄이고 데이터 해석의 용이성을 높이는 방법과 데이터만을 학습하여 를 탐지하는 기계 학습 기반 이상 탐지 모델 및 최적화된 하이퍼 파라미터값을 제안한다.

Moving-target-defense (MTD) fundamentally avoids an illegal initial compromise by asymmetrically increasing the uncertainty as attack surface of observable defender changes depending on spatial-temporal mutations. However, existing naive MTD studies were conducted focusing only wired network And these cases have also been no formal research wireless aircraft domains with attributes that are extremely unfavorable to embedded system operations, such hostility, mobility, and dependency....