An Intelligent Data Driven Prediction Framework for Automatic Speed Controller in Automotive Safeness Using Federated Learning

Abstract

Transportation is a vital part for the mankind. We need and rely on the newlinetransportation for commuting to work, schools, business trips, holidays and even hospital newlinecommuting. But, how far the safety is relied in the transportation. According to the World newlineHealth Organization (WHO) it has been statistically recorded that, the second major cause newlineof death is due to road accidents and that too specifically Over speeding. newlineAccording to the National Informatics Centre (NIC), it had been stated that newlineaccidents are the major prone of fatalities, permanent disabilities and severe injuries when it newlinehappens due to over speeding. Over speeding become a common factor by not only the newlinevehicles approaching over speed, in fact refers to the exceeding established speed thresholds. newlineThe consequences of over speeding extend beyond mere statistics which leads to focus on newlinetransportation safety. newlineCurrent safety measures predominantly cater to high-end and expensive vehicles newlinewith over speeding alert or static speed limiter, leaving a wide array of other transportation newlinemodes, including two-wheelers and public transport which is not into consideration. newlineMoreover, existing systems often rely on static information, limiting their efficiency to map newlineinputs or basic speed alert generators. All those alerts are usually violated by road users. newlineExisting solution of autonomous vehicles, which is basically expensive and even the road newlineinfrastructure was not met out in the developed nations too. Countries like us with high newlinepopulation, where vehicle density is higher than the road infrastructure still faces, more newlineimpact of all such vehicles exceeding the desired recommended speed newline