Deep Q-Learning Based Adaptive MAC Protocol with Collision Avoidance and Efficient Power Control for UWSNs

Journal of Marine Science and Engineering, Год журнала: 2025, Номер 13(3), С. 616 - 616

Опубликована: Март 20, 2025

Underwater wireless sensor networks (UWSNs) widely used for maritime object detection or monitoring of oceanic parameters that plays vital role prediction tsunami to life-cycle marine species by deploying nodes at random locations. However, the dynamic and unpredictable underwater environment poses significant challenges in communication, including interference, collisions, energy inefficiency. In changing make routing possible among or/and base station (BS) an adaptive receiver-initiated deep with power control collision avoidance MAC (DAWPC-MAC) protocol is proposed address The framework based on Deep Q-Learning (DQN) optimize network performance enhancing a varying locations, conserving path loss respect time depth reducing number relaying communication reliable ensuring synchronization. environment, shaped variations environmental such as temperature (T) latitude, longitude, depth, carefully considered design protocol. Sensor are enabled adaptively schedule wake-up times efficiently transmission communicate other and/or courier node data collection forwarding. DAWPC-MAC ensures energy-efficient time-sensitive transmission, improving packet delivery rati (PDR) 14%, throughput over 70%, utility more than 60% compared existing methods like TDTSPC-MAC, DC-MAC, ALOHA MAC. These enhancements significantly contribute longevity operational efficiency time-critical applications.

Язык: Английский

---

Adaptive collision avoidance strategy for USVs in perception-limited environments using dynamic priority guidance

Advanced Engineering Informatics, Год журнала: 2025, Номер 65, С. 103355 - 103355

Опубликована: Апрель 12, 2025

Язык: Английский

---

Deep multi-view information-powered vessel traffic flow prediction for intelligent transportation management

Transportation Research Part E Logistics and Transportation Review, Год журнала: 2025, Номер 197, С. 104072 - 104072

Опубликована: Март 21, 2025

Язык: Английский

---

Seafarer competency analysis: Data-driven model in restricted waters using Bayesian networks

Ocean Engineering, Год журнала: 2024, Номер 311, С. 119001 - 119001

Опубликована: Авг. 15, 2024

Despite the efforts of maritime authorities to enhance seafarer competencies through International Convention on Standards Training, Certification and Watchkeeping for Seafarers (STCW), human error remains a leading cause accidents. To thoroughly investigate impact various errors among seafarers accidents, this paper aims examine relationships between accidents using data-driven approach from perspective bridge resource management (BRM). Through analysis historical accident reports, dataset associated with is established. The least absolute shrinkage selection operator (LASSO) method employed identify critical prevention. Then, Bayesian Network (BN) model, based Tree Augmented Naive Bayes (TAN) method, constructed reveal relationship types, which are validated by sensitivity case study. results indicate that key all 'Maneuvers', 'Amend/maintain ship course', 'Decision making', 'Cognitive capacity', 'Information', 'Procedure operations', 'Situational awareness' 'Communication'. Moreover, study underscores importance leveraging lessons learned past mitigate risks ensure safe operations. findings contribute deeper understanding dynamics unveiling joint different This offers valuable insights in strengthening safety regulations.

Язык: Английский

---

A multi-stage collision avoidance model for autonomous ship based on fuzzy set theory with TL-DDQN algorithm

Ocean Engineering, Год журнала: 2024, Номер 311, С. 118912 - 118912

Опубликована: Авг. 8, 2024

Язык: Английский

---

Ship Anomalous Behavior Detection in Port Waterways Based on Text Similarity and Kernel Density Estimation

Journal of Marine Science and Engineering, Год журнала: 2024, Номер 12(6), С. 968 - 968

Опубликована: Июнь 8, 2024

The navigational safety of ships on waterways plays a crucial role in ensuring the operational efficiency ports. Ship anomalous behavior detection is an important method water traffic surveillance that can effectively identify abnormal ship behavior, such as sudden acceleration or deceleration. In order to detect potential real time, for proposed based text similarity and kernel density estimation. Under assumption known patterns entering leaving port, this behaviors violate time. Firstly, estimation applied construct pattern model trajectories used estimate values motion states. Simultaneously, semantic transformation convert trajectory into text, which ship’s pattern. Subsequently, historical data target are transformed textual trajectories, inbound outbound patterns. Furthermore, constructed real-time state motion, points exceed threshold anomaly factor marked anomalies. Finally, effectiveness validated using simulation data, results indicate accuracy more than 90% comprehensive behavior. This study, approaching from perspective port patterns, enriches methods waterways.

Язык: Английский

---

Graph-driven multi-vessel long-term trajectories prediction for route planning under complex waters

Ocean Engineering, Год журнала: 2024, Номер 313, С. 119511 - 119511

Опубликована: Окт. 23, 2024

Язык: Английский

---

Mixed traffic conditions of autonomous and human-driven ships: Assessing channel traffic capacity bounds and optimizing channel management

Ocean Engineering, Год журнала: 2024, Номер 314, С. 119734 - 119734

Опубликована: Ноя. 12, 2024

Язык: Английский

---

Declarative ship arenas under favourable conditions

Ocean Engineering, Год журнала: 2024, Номер 316, С. 119927 - 119927

Опубликована: Дек. 1, 2024

Язык: Английский

---

An improved hierarchical deep reinforcement learning algorithm for multi-intelligent vehicle lane change

Neurocomputing, Год журнала: 2024, Номер 609, С. 128482 - 128482

Опубликована: Авг. 28, 2024

Язык: Английский

---