Toward Optimal MEC-Based Collision Avoidance System for Cooperative Inland Vessels: A Federated Deep Learning Approach

Excess path loss due to obstruction by building corners at road intersections in urban and suburban areas (and the consequent loss of connectivity between vehicles or vehicles and infrastructure) limits the performance of vehicular ad hoc networks, especially those used to implement cooperative collision avoidance systems and emergency vehicle alert systems that warn drivers of oncoming traffic on cross streets. In recent years, researchers have developed simplified models of the excess path loss due to propagation around intersection corners that are useful in the simulation of vehicular ad hoc networks in suburban and urban environments. However, the geometry of actual intersections often exhibit a great deal more complexity than the simplified or idealized versions that form the basis for such existing excess path loss models. Here, we use ray-tracing simulations based upon accurate renderings of thirty-five intersections in and around urban and suburban areas of Vancouver, BC. to assess the accuracy of commonly used closed- form corner propagation models and compare the performance of assisted and unassisted cooperative collision avoidance systems.

Estimating meteorological visibility range under foggy weather conditions: A deep learning approach

Procedures and systems for air and marine separation assurance are compared. Discrete lane area navigation, pilot warning indicators and collision avoidance systems are discussed. It is recommended that a cooperative collision avoidance system be developed for marine use, and that satellite technology be applied to the precision navigation requirements of coastal confluence area, and channel traffic management. A need for lower cost equipments for general aviation is stressed.

In recent years, AIoT (Artificial Intelligence of Things) applications have received lots of attention since it exploits widely deployed IoT devices to collect data and perform an action while leveraging AI to obtain knowledge and insights. When deploying AIoT in 5G networks, Multi-access Edge Computing (MEC) is appropriate for enabling local management of IoT devices and computation of machine learning (ML) algorithms. Facing the multitude of threats in the ML data layer, IoT service layer, and 5G communications layer, MEC should enable corresponding detection and mitigation schemes to protect AIoT applications. In this article, we examine the existing security solutions located in each layer and discuss the interrelated challenges. For example, a network-traffic-based IDS solution in MEC might capture IoT malware but fail to identify a growing file-less attack. We suggest that firmware emulation in IoT endpoint should be included to provide system-level behaviors to network-based detectors in MEC so that file-less attacks can be distinguished. The potential of a backdoor attack aiming to poison data or corrupt ML models cannot be ignored in AIoT applications, and a corresponding detector should be implemented in MEC. Due to the rise of low-cost Software-Defined Radio (SDR), malicious attacks using rogue base stations (BSs) have become more popular. It implies that security protection at MEC in the communications layer is necessary. This article, therefore, proposes a novel platform, M3Inspector, where inspectors located in mobiles and AIoT machines collect information from surrounding BSs and provide them to MEC. MEC determines the rogue BS and makes a notification to users subscribing to the local service. A realistic 5G experimental platform with rogue BSs is developed. The results demonstrate that attack detection and mitigation can be implemented in the MEC paradigm to significantly improve the security protection of AIoT from the perspectives of rogue BS attacks and file-less attacks.

Operations in the Next Generation Air Transportation System (NextGen) are planned to have a reduced dependency on primary surveillance radar and use Automatic Dependent Surveillance - Broadcast (ADS-B) to provide aircraft state information to Air Traffic Control (ATC) and onboard collision detection and avoidance systems. With the ongoing integration of Unmanned Aircraft Systems (UAS) in the National Airspace System (NAS), ADS-B is expected to become the main surveillance input to UAS collision avoidance. Traffic Alert and Collision Avoidance System (TCAS) is unable to meet requirements in terms of adaptability to future operations with reduced separation margins, addition of new user groups (e.g., UAS), and support of additional surveillance inputs. Airborne Collision Avoidance System (ACAS X) is being designed to meet these requirements and to increase surveillance, alerting, and resolution performance. The RTCA has released Minimum Operational Performance Standards (MOPS) to provide system and operational performance characteristics that ADS-B must provide. This paper analyzes recorded ADS-B data from test flights that were performed in the Continental U.S. (CONUS) and verifies measured performance with specified operational characteristics in terms of coverage, accuracy, and integrity.

Most ship collisions are caused by human error, making them unpreventable as long as humans operate a ship. Autonomous navigation technology is expected to prevent these accidents. Collision avoidance is a key issue in the realization of autonomous ships. To date, rule- and artificial intelligence (AI)-based collision avoidance algorithms have been studied. However, most of them assumed a simple environment in which other ships maintain their speed and course. During actual navigation in congested waters, safe and efficient collision avoidance are achieved through cooperative maneuvers between ships that obey the COLREGs. To realize cooperative maneuvers of ships, multi-agent deep reinforcement learning was applied to learn collision avoidance maneuvers considering the motions of other ships and COLREGs. Then, the developed multi-agent AI was evaluated with typical one-on-one encounters. The multi-agent AI performed cooperative collision avoidance maneuvers in head-on and crossing situations. Furthermore, the influence of reward setting regarding the degree of compliance with COLREGs on the cooperative collision avoidance was investigated. These results demonstrated that the multi-agent AI was promising for realizing cooperative collision avoidance maneuvers similar to those performed by humans.

This survey paper explores the application of machine learning (ML) techniques in the domain of workplace injuries, focusing on three key areas: risk prediction, return to work (RTW) strategies, and demographic analysis. Through an extensive review of literature from January 2015 to July 2024, this paper examines the latest advancements in ML-driven approaches to workplace safety and identifies important research gaps. This paper highlights how classical ML techniques, such as ensemble models and decision trees, have become essential tools for identifying workplace injury risks, enabling more accurate interventions. It emphasizes the importance of leveraging ML in personalized RTW programs, which use data-driven insights to improve recovery outcomes and reduce economic demands. In the context of demographic analysis, this paper explores how ML algorithms can uncover disparities in injury rates across various age groups, industries, and occupations, underscoring the need for targeted safety measures. Moreover, research gaps are identified, particularly regarding the emerging potential of advanced ML techniques, such as deep learning and large language models (LLMs), for analyzing structured and unstructured safety data, methods that have not yet been widely applied in workplace injury research. As such, future research should apply recent advances in ML, integrating these approaches with comprehensive and accessible datasets to enhance the prediction and prevention of workplace injuries, provide more detailed analytics and insights, and improve safety protocols across all industries. This comprehensive survey is an invaluable resource for researchers and practitioners leveraging ML to address complex challenges in workplace safety.

Background Peripheral Arterial Disease(PAD) is a prevalent cardiovascular condition affecting millions worldwide, and the advent of machine learning(ML) techniques offers promising opportunities to enhance risk prediction and personalise patient management in this complex disease. This systematic review and meta-analysis aimed to evaluate the performance of ML models in predicting adverse outcomes in PAD patients compared to traditional statistical approaches. Methods A comprehensive search of major databases(MEDLINE,EMBASE,CINAHL,Cochrane CENTRAL) was conducted for studies published between 2000 and 2024. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the PROBAST tool. The quality of included studies was also assessed using criteria outlined by Qiao et al. A modified Hierarchical Summary Receiver Operating Characteristic(HSROC) analysis was performed to compare the predictive performance of various ML models and traditional regression methods. Results Thirteen studies met the inclusion criteria. Gradient boosted models demonstrated the highest predictive performance with a diagnostic odds ratio(DOR) of 36.593(95% CI:24.44-54.79), sensitivity of 0.853, and specificity of 0.863. The best performing ML models for outcomes within 2 years, showed a DOR of 34.169(95% CI:22.840-51.117). Traditional regression models consistently underperformed compared to ML approaches, with the lowest DOR of 3.326(95% CI:1.814-6.097). The quality assessment revealed a mix of methodological rigor, with 54% of studies rated as low risk of bias and 46% as unclear. Conclusion Advanced ML techniques demonstrate superior predictive power for adverse outcomes in PAD patients compared to traditional regression methods.

The accurate detection and assessment of stress play a pivotal role in enhancing individual well-being and healthcare outcomes. Traditional methods of stress detection often grapple with limitations in accuracy and scalability. With the advent of machine learning (ML), the potential to revolutionize stress detection has emerged. This paper presents a comprehensive study on the application of ML algorithms for stress detection, with a focus on physiological and behavioral data analysis. Central to our approach is the integration of SHapley Additive exPlanations (SHAP), an Explainable Artificial Intelligence (XAI) technique, to interpret ML models. SHAP provides a novel lens to understand the impact of individual features in the complex decision-making processes of ML models, thereby enhancing the transparency and reliability of stress predictions. We demonstrate how SHAP not only aids in elucidating model decisions but also contributes to refining the models for greater accuracy. Our results highlight the effectiveness of ML in detecting stress and the pivotal role of XAI in making these models more interpretable and trustworthy. This study underscores the synergy between advanced ML techniques and XAI, paving the way for more nuanced and reliable stress detection methodologies that are essential in diverse settings, from healthcare to workplace environments.

Both ground-based ATC (Air Traffic Control) and airborne CAS (Collision Avoidance Systems) are needed now and in the future to cope with the problem of collision aloft. They should be considered mutually complementary. The limited form of the CAS known as a PWI (Proximity Warning Indicator) is believed not to be sufficient for safety. The two essential elements of a CAS are 1) an indication of the effective maneuver to secure adequate miss distance, and 2) effective means for discovering and assessing the original risk. Miss distance produced by a maneuver never exceeds maneuver displacement. The ratio of the two may be low unless escape time is of the order of 30 seconds or more. In nearly all cases increasing escape time is more effective to insure adequate miss distance than is increasing maneuver severity. The preferred form of CAS would be self-sufficient, but this form appears not to be feasible for basic physical reasons. Cooperative CAS presents many problems, but is believed to have sufficient promise of success to justify careful study and substantial development effort.

Time/frequency technology provides a reliable aircraft collision avoidance system (CAS) that can operate in either synchronous or asynchronous modes. Precision time-ordered techniques of CAS provide both range and range-rate measurements in a one-way sense to all aircraft as well as ground stations within range of transmitted microwave signals. The cooperative system utilizes exact frequency references coupled with precise synchronization: control of frequency to 1 part in 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">8</sup> and time to less than 1 µs. In addition to performing specific functions of protecting aircraft, the time/frequency CAS provides a means for wide dissemination of submicrosecond timing. Flying clocks, which are an integral part of the airborne CAS, have been providing transcontinental and intercontinental transfer of time since 1964. CAS ground stations can serve as depositories of time and frequency derived from flying clocks, satellites, Loran-C and Omega navigation systems, or television transfer referenced to national and international time/frequency standards. Airborne relay of CAS time provides a ready means for local transfer of time to users that are not within line of sight to a ground station. Thus time and frequency can be disseminated for applications other than collision avoidance. In the process of developing the technology and originating CAS, new systems for communication, position determination, navigation, and vehicle surveillance evolved. Such systems can be structured to operate at high data rates, provide improved accuracy and use less RF spectrum than existing techniques.

This study focuses on the security of 5G mobile network major technology called Multi-Access Edge Computing (MEC) and its susceptibility to distributed denial of service (DDoS) attacks. The goal of the research is to address the effects of DDoS attacks and implement effective mitigation techniques. Several supervised Machine Learning (ML) techniques, which include Random Forest (RF), Decision Tree, Naïve Bayes, K-Nearest Neighbour, Logistics Regression, and Blending/Stack Model, are evaluated using multiple performance metrics such as accuracy, detection/recall, F1-Measure, Matthew’s correlation coefficient, Receiver Operating Characteristic, and Area Under Receiver Operating Characteristic. According to literature, ML algorithms achieve the best performance in mitigating DDoS attacks, therefore, they can be optimized to enhance their effectiveness. The research provides an overview of the existing mitigation schemes in the MEC and proposes a DDoS mitigation scheme. The findings show that hybrid models outperformed traditional ML models. Among the mitigation techniques evaluated, RF proved to be the most effective in mitigating DDoS attacks in MEC.

&lt;div class="htmlview paragraph"&gt;Since 1955 the aviation community has been investigating the feasibility of an airborne collision avoidance system. These studies have included analytical and engineering design efforts as well as flight simulation tests and actual flight tests. The majority of the effort expended during these investigations has been devoted to the engineering aspects of the systems. Of equal importance in meeting the overall objective of collision avoidance is the transmission of system information to the crew in a form that can be used effectively in the operational environment. The Boeing Commercial Airplane Company fulfilled a contract for the FAA, using simulation to help determine flight deck display requirements and operational procedures for the traffic alert and collision avoidance system (TCAS II) in commercial transport aircraft.&lt;/div&gt; &lt;div class="htmlview paragraph"&gt;Using the experience gained from these studies, this paper describes the type of efforts required to develop pilot factor characteristics for future collision avoidance systems it addresses man-machine interface requirements to ensure that the system is compatible with the overall aircraft system. It presents some of the functional requirements that the display and control components of the system must meet to accomplish separation assurance. Some of the subjects that are discussed include: a) the impact of the collision avoidance system utilization philosophy on the display requirements; b) the importance of designing the collision avoidance display system so that it can be integrated into the flight deck environment; c) techniques for incorporating realistic human performance estimates into the system performance assumptions; d) the effect on system performance of using natural response patterns; e) techniques that can be used to generate pilot response time commensurate with the situational requirements; f) the role of simulation in the design process; and g) consideration for certification during the development of a collision avoidance system.&lt;/div&gt;

This research develops a performance evaluation model of humans in avoiding ship collision situations compared to the AI ship collision avoidance system (CAS). A human, the remote operator (RO) of Maritime Autonomous Surface Ships (MASS), ought to make compact collision avoidance (CA) decisions to secure safety and efficiency during remote operations. Hence, evaluating RO’s CA performance is important; however, research on developing evaluation methods concentrates merely on evaluating trainees based on instructor’s guidelines, while artificial intelligence (AI) makes CA decisions through parameter-based calculation. Therefore, this research proposes a CA performance evaluation model for RO of MASS based on CAS intervening of RO trainee’s simulation training in ship collision avoidance. In each time interval, the RO’s decisions showed divergent behaviors under given situational conditions compared to the CAS’s behaviors in equivalent situations. Findings denote the benefits of CAS intervention methods and RO performance evaluation model based on the proposed performance features.

Mobile ad hoc networks (MANETs), which are a promising method for the intelligent transportation system, include vehicular ad hoc networks (VANETs) (ITS). Developing reliable and strong cooperative collision avoidance (CCA) strategy to mitigate the growing number of road fatalities each year is one of the main difficulties facing vehicular ad hoc networks (VANETs).A proper and successful routing method aids in the successful expansion of vehicular ad hoc networks. This study explains the architecture, interface layers, safety features, and implementation of a novel priority-based direction-aware collision avoidance system (P-DVCA). It distinguishes our study in the collision area of VANETs by accounting for realistic bi-directional traffic. The scheme begins with the development of dynamic clusters, which is difficult because of the bi-directional diverse traffic and the need to avoid collisions within and between clusters. The target node is sent an early warning message that includes the safe speed and the likelihood of a collision in order to notify it of an impending danger. To determine the least expensive, shortest one with the fewest hops between the source and the endpoint. A crucial problem with VANETs is the transmission of data from a source node to the base station. Cross-layer issues must be solved for a robust and stable collision avoidance programme to function properly in a VANET communication architecture. The results of the simulation show that the suggested scheme significantly outperforms CCM and C-RACCA in terms of cluster stability, fewer collisions, low latency, and low communication overhead. According to the findings, P-DVCA offers stable clustering, minimises network congestion, and lowers communication overhead and latency.