Research And Implementation Of Monitoring, Analysis And Actively-enabled System On Internet Of Vehicle Safety Events

Internet of vehicles(IoV)is one of the most promising and fastest growing industries as of today.With the growing maturity of big data computing and the popularization of 5G network,IoV will become increasingly reliable and innovative.In terms of exploring IoV,improvements on a technical level and new ideas dedicated to provide services to users are both very valuable.Safety is a significant issue in relation to the development of Internet of vehicles,especially in networking information security and traditional car accidents caused by driver negligence.Through research and investigation of various IoV systems at home and abroad,Internet of vehicles is accompanied by various safety events from driver behaviors that jeopardize safety to networking information security events.This thesis discusses how to deal with these safety events in the IoV industry.We developed a system to provide real-time monitoring and analysis of all kinds of safety events through using the advantages of big data to analyze those events and delegate plans to improve the overall safety of Internet of vehicles.Monitoring,analysis and actively-enabled system on Internet of Vehicle safety events is a proposed solution for dealing with safety events,which include the following aspects:1.Classification algorithm to depict safety events.Based on the study of the existing international standards,statistics on various safety events in IoV systems,and the theory of fuzzy mathematics,we proposed a unified classification algorithm for all types of safety events and modified the algorithm to produce results that conformed to the ISO-26262 safety rating standards.2.Life cycle management of safety events.Using Flink flow computing framework and Kafka message middleware to detect and analyze the safety events in the IoV platform,the system will give an accurate prediction regarding whether a safety event occurred on the system or not.If a safety event is detected,the system will immediately call the classification algorithm to start preliminary classification.Then,according to the event’s developing situation,the system will increase or decrease the safety level of the event.While the event exists on the system,all relevant information will be recorded and saved to the disk at the end of the event to prepare for subsequent big data analysis.3.Actively-enabled module based on pre-plan.When classified safety events are sent to the actively-enabled module,the module will initiate a pre-designed plan.The system itself will not solve these safety events.Instead,it will inform the other modules on the IoV platform by sending a message that enables recipient modules to solve those safety events.4.Statistical analysis of big data.The large amount of raw data generated in the process of dealing with safety events can be used for big data analysis.The data is calculated by off-line big data computing to get the most frequent events,user big data portraits,etc.,in order to obtain more information conducive to service IoV system and further optimize its platform architecture.