Subject Area: CONTROL SYSTEM ENGINEERING
Over the years the maritime industries have faced many challenges of insecurity, theft, piracy and maritime terrorism among other problems which have often threatened the lives and peace of onboard vessels and hence demand immediate solutions. To solve this, an artificial intelligence-based system has been proposed using a support vector machine, convolutional neural network, fuzzy logic, genetic algorithms and artificial neural network among others, this study has recorded great success in the recognition of vessels and trajectory approximations but there is still need for a system which can predict the position of an online vessel and also recognize the vessel with a high rate of accuracy. This was proposed to be addressed in this research using a machine learning approach to develop an improved model to optimize the performance of the Automatic Identification Systems (AIS).The result achieved 89.55% success in correct prediction of vessel latitude position and also 87.1% for longitude prediction which gives a percentage deviation of 10.45% for latitude and 12.9% for longitude which signifies that a good vessel prediction success has been recorded.