The demand for vessel real-time tracking and automated collision avoidance technologies has grown in unison with the development of new edge data processing systems that allow for extraordinary volumes of data to be processed at the collection point. This work focuses on designing and implementing an edge computing framework for real-time vessel tracking, emphasizing improving marine safety and operational efficiency. The proposed system architecture provides local processing capabilities for making high-priority autonomous collision avoidance decisions, thus achieving greater efficiency in bandwidth and latency usage. Its design includes advanced sensors, machine learning predictive analytics, and cloud-agnostic edge data processing units. The proposed approach facilitates advanced real-time monitoring of vessels for automatic alerting of onboard systems and control centers, movement patterns, and collision risk assessment. The study also addresses scalability, data security, and network reliability issues, proposing solutions with robust fault-tolerant communication protocols, adaptive data processing, and strong coordination between the edge and cloud. The study demonstrates high maritime operational efficiency can be attained with edge computing by providing marker-based ship monitoring and collision avoidance systems that are scalable, reliable, and designed for low-latency response in dynamic environments.