Hatch Detection And Scene Visualization Technology In Grab Anti-collision System Of Photoelectric Ship Unloader

Grab ship unloader is the key equipment in the unloading of dry bulk cargo terminal.Compared with continuous ship unloading equipment,such as hinged ship unloader,it has the advantages of flexibility,strong adaptability and low cost.Under the current manual unloading mode,the production efficiency and safety depend on the experience and proficiency of drivers.The probability of collision accident between grab and hull is significantly increased under the condition that the driver’s operating vision is limited and long-time high-intensity work.The handling of collision accidents is usually accompanied by shutdown and repair compensation,which greatly affects the production of the wharf.Therefore,it is very important to judge the position and posture relationship between the grab and the hull intelligently and give an alarm before the accident.Based on the problem of grab collision in actual production,this paper puts forward the application of photoelectric ship unloader in Wharf production,which has a significant effect on solving the problem of anti-collision.By installing a laser radar scanning system at the work site,the three-dimensional spatial information of the hatch being unloaded and its vicinity is captured,the hatch position is detected and extracted,and the material pile is modeled and displayed,so that the grab can safely land to the best grab point.This method can reduce the probability of collision between grab and hull,and realize remote monitoring.The main contents of this paper are as follows:（1）The characteristics and advantages of photoelectric ship unloader as the implementation means of anti-collision scheme are analyzed,and the anti-collision scheme based on photoelectric equipment and actual working conditions is introduced.Including equipment installation,scanning mode,selection of the best grab point and alarm strategy.According to the requirements of the implementation of the scheme,specific work indicators are proposed to guide the realization of hatch detection and scene visualization.（2）An effective hatch inspection scheme is proposed.It includes the methods of quickly eliminating irrelevant regions,extracting deck surface based on plane fitting,initial positioning of hatch based on low resolution image after rasterization and accurate boundary search.Measures such as rapid positioning of the target area and data dimensionality reduction are adopted to reduce the amount of data in the processing process,so that the time and accuracy indicators meet the requirements.（3）The visualization scheme of scene data is proposed,which provides an accurate and intuitive reference for the selection of the best capture point.Through the analysis and visualization of scene data,anti-collision model and stockpile model,subjective opinions are provided for the safe unloading process.Integrate the information and visual content of each part to provide a complete graphical user interface.According to the operation effect in the actual production site,the scheme can run smoothly and efficiently.This paper compares the characteristics between this scheme and the existing scheme,highlights the advantages of this scheme in real-time updating the position of cargo hatch and the status of material pile in the hold,and meets the actual needs of safe and efficient production of the wharf.