Enhancing Scan Matching Algorithms via Genetic Programming for Supporting Big Moving Objects Tracking and Analysis in Emerging Environments

Big moving objects arise as a novel class of big data objects in emerging environments. Here, the main problems are the following: (i) tracking, which represents the baseline operation for a plethora of higher-level functionalities, such as detection, classification, and so forth; (ii) analysis, which meaningfully marries with big data analytics scenarios. In line with these goals, in this paper we propose a novel family of scan matching algorithms based on registration, which are enhanced by using a genetic pre-alignment phase based on a novel metrics, fist, and, second, performing a finer alignment using a deterministic approach. Our experimental assessment and analysis confirms the benefits deriving from the proposed novel family of such algorithms.