Neural Net for Preventive Diagnostics System of Technical State of Vehicles in an Intelligent Transport System

Abstract

One of the main challenges facing European experts is organizing a dynamically functioning and efficient transportation sector. Efforts in this regard have been focused mainly on projects aimed at developing intelligent automotive transportation systems (IATS), which integrate information and communication technologies (ICT) into transport infrastructures and vehicles (Car-to-Car, Network on Wheels, FleetNet, COM2REACT, CARTALK2000, SAFE TUNNEL, CVIS, GST, WILLWARN, etc.). This work is multifaceted and contingent upon specific objectives. One of the most significant problems in developing and implementing new transport systems is striking the right economic balance between upgrading existing infrastructure and introducing innovative technologies, as embodied by the concept of the so-called Intelligent Automotive Cooperative Transport System (IACTS), which considers interactions both between vehicles themselves and between vehicles and communication infrastructure. In this case, urban transport management encompasses real-time monitoring of road conditions, along with implementing controls or influencing traffic flows based on gathered data to alleviate congestion, enhance safety, efficiency, eco-friendliness, etc. For these purposes, neural networks, characterized by rapid information processing and decision-making capabilities, are widely employed. Specifically, they can be utilized for predictive analyses of vehicle malfunctions, forming the foundation for relevant services. The goal of this study is to design a neural network for a preventive diagnostic system targeting the technical status of vehicles within an IACTS framework, thereby mitigating the impacts of vehicular breakdowns during road operations.