ANT: Deadline-Aware Adaptive Emergency Navigation Strategy for Dynamic Hazardous Ship Evacuation With Wireless Sensor Networks

Efficient and safe ship evacuation strategy plays a critical role in protecting passengers' lives when ships encounter accidents. Existing wireless sensor network (WSN)-based emergency navigation methods mainly consider the dynamics of hazards and accordingly plan evacuation paths to minimize human exposure to the environmental hazards, such as fire and smoke. However, without sufficient consideration of the ship capsizing time and the impact of dynamic ship inclination on the passengers' walking speed, these methods may fail to evacuate passengers before the deadline. In this paper, we propose ANT, a deadline-aware adaptive emergency navigation strategy for dynamic hazardous ship evacuation with WSNs, which informs each passenger about a hazard-avoided evacuation path to successfully reach lifeboats within the specified deadline under all circumstances. To achieve this aim, ANT analyzes the process of ship capsizing to predict the specific limited evacuation time and the worst-case traversal delay. Next, an online evacuation path planning strategy is proposed based on a real-time adaptive routing algorithm to maximize navigation efficiency while ensuring user safety. We evaluate ANT by conducting prototype experiments and extensive simulations. The results demonstrate that ANT improves the navigation success ratio by 40% and 5%, compared with state-of-the-art emergency navigation systems, namely, medial axis-based approach and ENO-based oscillation-free emergency navigation approach, respectively.