Predictive Models for ABS and TPMS based on Gaussian Naïve Bays

Abstract

The car industry is currently preoccupied with the issue of safety. The increasing number of accidents occurring around the world as a result of automobile problems is a major contributing factor to these incidents. The amount of complicated electronics that is used in vehicles is becoming more prevalent every day. A great effort has been made in evaluating vehicle features in relation to vehicle components. Through such systems, a smart architecture and complex function designs are involved. During all of this vehicle transformation and evolution, the automotive industry recognises a high demand for vehicle safety. While designing and manufacturing this system, automotive experts understand a need for a strict monitoring and feedback system for complex vehicle architecture, which can notify the end user if there is any indication of a failure ahead of time.
 In order to effectively participate in vehicle design activities, it is critical to grasp the significance of safety features. Tire system failures and braking system failures have played a large role in several recent traffic accidents. The failures of the tyre system and the braking system in the vehicle are addressed in this study. While investigating this system, it is discovered that it is supported by complex electrical systems, which include an ECU (electronic controller unit), sensors, and a wire system.
 Through the use of these technologies, censored data can be processed in a timely manner and made available for diagnostic purposes. Nevertheless, car diagnostics is needed after any vehicle failure but that does not serve the aim of maintaining vehicle safety. As a result, predictive analysis or predictive diagnostics may be a viable option for informing the driver about the health of a particular vehicle component in advance. In this study, the author discusses the concepts of vehicle prognostics for the tyre pressure monitor system and the antilock braking system, which are accomplished using a statistical method of machine learning. In today's world, machine learning is expanding in breadth, and the world is becoming more aware of its enormous potential in the field of data analytics. It is the purpose of this study to introduce methodologies by which machine learning can assist vehicle predictive analytics to attain the intended goal of vehicle safety.The author of this article discusses how Bayesian statistics may be used to produce predictions in the form of probability estimation. The prediction's outcome is thoroughly analysed.