Towards improved road traffic safety: A modelling and IoT integration approach

Abstract

Every year, the lives of millions of people are cut short due to road traffic crashes. Millions of people suffer non-fatal injuries, leading to lifetime disability. Traffic accidents are caused by many factors including reckless driving, inappropriate speeding, drunkenness, violation of traffic lights, and dangerous manoeuvres. Speed has been one of the leading causes and a fundamental risk factor for road traffic injury problems. Excessive speed is a crucial risk factor for road traffic collisions, deaths, and injuries. Cars become more difficult to manoeuvre at higher speeds, especially on curves or where evasive action is necessary. The forces experienced by the human body in a collision also increase as the speed increases. In low- and middle-income (LMI) countries, this proportion is likely high, given the higher proportion of deaths among vulnerable road users. On Rwanda’s roads, speeding is among the significant causes of accidents. Several approaches have emerged to support road safety towards reducing fatalities and severe impairment outcomes. Therefore, speed and management are at the core of a safe road system. There is a need to look for all ways of curbing the accident rate caused by inappropriate speeding by preventing people from being exposed to risk. Besides this, there is a need to know the details of the cars, like current location, and speed to gain details when an accident occurs. The Global Positioning System (GPS) advancement and the Internet of Things (IoT) have opened new systems to limit a maximum speed to a safe speed with no effect on the regular operation of vehicles. The whole process will be accomplished based on GPS technology and sensing technology to maintain the appropriate speed. This thesis aims to evaluate how to reduce inappropriate speed caused by road accidents towards a safer traffic system. The thesis comprises three scientific papers that summarize this work’s main contributions. Hence, the focus of this research is an architectural framework for vehicle monitoring, designing an intelligent electronic speed governing system for the maximum speed of the vehicle to be automatically limited to the local speed limit. The system inside the car will be able to control the vehicle’s current speed to the required maximum speed in a particular speed zone. In Paper I, we investigated the readiness for intelligent transportation systems (ITS), existing technologies, and transportation applications in Rwanda as the majority are not familiar with the concept. An IoT based conceptual framework was proposed to improve the transportation system. In Paper II, structural speed control and IoT based online monitoring system was developed to monitor vehicle data continuously. Multiple linear regression (MLR) and random forest (RF) models were evaluated to find the best model to estimate the required voltage to be supplied to the motors. Based on the coefficient of determination, the RF performs better than the MLR. In Paper III, we proposed a multi-server queueing model for traffic signal optimization to strengthen the sustainability of urban mobility systems. A numerical algorithm was successfully developed and discussed to compute steadystate performance measures.