A Novel Vectorization Tracking Algorithm for Maritime Emission Monitoring Assisted with E-Nose Enabled Unmanned Aerial Vehicle

Monitoring the gas emission of maritime vessels is very challenging as the fields are almost inaccessible and the dispersion is very susceptible to the surrounding environment. In ultra-large-scale maritime scenes, general emission detection and tracking frameworks based on infotaxis will most likely fall into local turbulence traps. In this work, a novel vectorization tracking algorithm assisted with E-nose enabled unmanned aerial vehicle (UAV) is proposed for efficient monitoring the vessel emission. Firstly, a new prediction model using AIS data is developed to estimate globally vessel emission. Then, real-time measurements by the onboard E-nose are fused with the emission prediction in the vectorization framework. Based on the proposed tracking algorithm, the UAV’s flight velocity and heading can be adjusted towards the vessel emissions accurately and efficiently. Another contribution of the paper is that a series of real-world maritime experiments with a developed were carried out in order to evaluate the algorithm. The acquired results indicated that the proposed tracking algorithm and the developed E-nose enabled UAV are very efficient for such a maritime environment.