Background of the plan

With the development and utilization of water transportation, economic exchanges between the north and south of the Yangtze River have become increasingly frequent. The demand for bridges in cities along the river has been increasing year by year, and large-span bridges are constantly being built. Although this has eased the pressure of land transportation and promoted regional economic and social development, a large number of bridges have become obstacles to ship navigation. The contradiction between bridge construction and safe navigation is becoming increasingly prominent, and the risk of ship bridge collisions is constantly increasing. Ship bridge collision accidents occur from time to time.

In recent years, the navigation of ships in bridge areas has been facing potential risks of ship bridge collisions. How to take effective measures to reduce the probability of ship bridge collisions, control and manage collision accidents in bridge areas, and protect the safety of bridges and passing ships has received widespread attention and high importance at home and abroad.

Recently, with the release of the "Three Year Action Plan for Hidden Danger Management of Ship Collision Bridges", the urgent issue of bridge safety and health has ushered in the best governance opportunity. Based on the current construction situation and future development needs, building an active collision avoidance system for bridges is in line with development needs and should be implemented as soon as possible.

Technical pain points

Pain point 1 of target extraction: low automation level of marine radar
Pain point 2 of target extraction: Marine radar requires manual monitoring
Pain point 3 of target extraction: The impact of AIS passivity and limitations
Pain point 4 of target extraction: The phenomenon of ships sailing on shore due to AIS trajectories being point traces

Scheme Introduction

Solution for Bridge Anti-collision Active Warning System is an intelligent warning system that integrates multiple information for high-precision tracking, dynamic warning, collision avoidance guidance, and visual fusion. Through the process of automatic alarm, automatic intervention, unmanned operation, and manual review intervention, it achieves automatic warning of ultra-high, yaw, and overspeed risks. In the warning state, VHF automatic alarm call is activated, and in the case of continuous warning state, sound and light alarm automatic warning is activated. The monitoring center platform automatically generates warning information and pushes it to remote terminals and mobile apps. During the warning process, manual review and disposal can be carried out. In case of endangering the safety of the bridge, the emergency response process can be initiated manually to command and implement safety precautions. This system can detect and track ships in monitored waters in real-time 24/7, plot their motion trajectories, determine their threat level, and command optoelectronic devices to track and identify targets, capture and collect evidence, providing a good solution for active collision warning and alert of bridges. The system has a wide coverage area, a large number of monitoring targets, fast monitoring speed, and a large number of collaborative cameras. It is not affected by rainy days, haze, or night time, and has a high degree of automation, reliability, and efficiency. It can effectively reduce the workload of security personnel on duty.

system architecture

major function

1、 System Composition and Functions: (Multi-source Information Fusion) The system adopts a one-station-one-center architecture. 

2、 System composition and function: The warning area provides 4x distance and time redundancy

The warning area setting model is to monitor the upstream and downstream 1000 meters, divided into three supervision areas. The range of 1000 to 700 meters is the warning area, the range of 700 meters is the warning area, and the range of 400 meters is the alarm area. The prompt area is used to detect ships with excessive height, deviation, and overspeed in advance; The warning area is used for automatic target classification and recognition, threat assessment, and warning of ships with high altitude, yaw, and overspeed. At the same time, it automatically generates green navigation guidance lines for crossing bridges; The alarm area is used for emergency reporting and initiating disposal functions.

3、 System composition and functionality: adaptive technology and functionality

Advantages of the Solution

• The focus of this Solution is to complete the full monitoring of bridge crossing vessels in the water area of the bridge, and to complete early warning based on collected data; The difficulty lies in ensuring the accuracy of early warning and determining potential risks, especially for ship loss of control behavior, which is difficult to accurately identify in a short period of time. By using high-precision sensors and algorithm design, this difficulty can be overcome to the greatest extent possible.
• By using phased array radar signal processing algorithms and unique adaptive threshold design, and integrating the comprehensive processing capability of ship data information, an intelligent, real-time, and convenient target automatic detection, tracking, identification, and early warning function is provided for ship supervision in checkpoint waters; The system can detect and track ships in real-time 24/7, plot their motion trajectories, determine their threat level, and command optoelectronic equipment to track and identify targets, capture and collect evidence; Not affected by weather conditions such as light, night, haze, rain, and fog, it has a high degree of automation, reliability, and efficiency, which can effectively reduce the workload of security personnel.
• By using video deep learning detection to obtain the basic information of passing ships (ship name, ship type, captain, heading, etc.), the basic information of the ships is overlaid onto the current video frame. The overlaid image of the basic information of the ships will always follow the movement of the ships in the camera frame, with illustrations and text; Through the fusion, discrimination, and analysis of video recognition information and AIS information, non AIS ships, pseudo AIS ships, etc. are identified. Through deep learning and artificial intelligence technology, ships are detected, located, classified, listed, and named, providing an intelligent, real-time, and convenient function for electronic declaration and supervision of ships.