A Novel Clustering-Based Anomaly Detection Approach Using a Game Engine for Road Safety

Despite the current high technological progress, the problem of transport safety still remains a relevant topic for research and new developments. In recent years, the implementation and use of machine learning to improve transport safety as additional measures to traditional safety methods has been increasingly discussed. This article tries to bring novelty and continue to develop the topic of transport safety through the combined use of game engines and machine learning. The literature review will consider some cases on the traditional use of machine learning, as well as existing studies with a focus on game engines to improve transport safety. The practical part of the methodology will present developments in the creation, training and testing of machine learning models on road scenarios created on the Unity game engine. In conclusion, the results of the work done and further recommendations for improvement will be analyzed and discussed.