Simulation Based Evaluation of Advanced Driver Assistance Systems

Abstract

Road transportation is an essential element of mobility in most countries and we can observe an increasing demand for both goods and passenger traffic. There are however important societal and economical problems related to road transportation in terms of congestions, traffic safety and environmental effects. During the last decades vehicles have increasingly been equipped with different types of Advanced Driver Assistance Systems (ADAS). These systems can to some extent compensate for human behaviour and errors that cause congestions, accidents and air pollution. Most studies conducted to evaluate ADAS have focused on ADAS impacts on the driver or on the vehicle. Since an ADAS might influence not only driving behaviour and vehicle dynamics, but also the interaction between equipped and non-equipped vehicles, it is also important to consider the resulting effect on the traffic system. A reliable and realistic evaluation approach needs to include estimations of drivers' decisions in different traffic situations with respect to the ADAS functionality and how such decisions affect the traffic system as a whole.The overall aim of the thesis is to develop a simulation based evaluation framework for investigations of impacts of different types of cruise controllers on the traffic system. The objective is also to apply the framework to evaluate a fuel minimizing cruise controller for trucks, the Look Ahead Cruise Control (LACC). The framework developed consists of a combination of a microscopic traffic simulation model, and a vehicle and ADAS simulation model. When applied for a specific ADAS, as for example the LACC, the framework needs to be complemented with a driver model that captures the changes in driving behaviour due to the system of interest. In this thesis a driver model for LACC equipped trucks was developed based on results from a driving simulator experiment, a field operational test, and a focus group study. Simulation experiments were carried out to observe the LACC impacts on the traffic system with respect to penetration rate, traffic density, and variation in the desired speed. Environmental effects were estimated using emission calculations.The results show that an increase in traffic flow influences LACC-trucks more than CC-trucks with respect to fuel consumption and emissions. However, the results also show lower fuel consumption and emissions for LACC-trucks despite increased traffic flow. Increased penetration rate of LACC-trucks does not show any negative effect on traffic efficiency. iv v Acknowledgments

A Multi-Level Framework for Traffic Safety Assessment under Automated Driving Functionalities: the Need and Outline of a Holistic Approach

“…However, validated models of human and vehicle behavior implemented in simulation can provide insights into the likely impacts on traffic safety. An interesting example found in literature is a simulation-based evaluation framework developed for investigations of impacts of different types of cruise controllers on the traffic system (11).…”

A New Indonesian Traffic Obstacle Dataset and Performance Evaluation of YOLOv4 for ADAS

“…YOLO [11], YOLOv2 [26], YOLOv3 [27], YOLOv4 [28], SSD [10], RetinaNet [29], have begun to be widely applied in the world of computer vision. CNN-based deep learning methods are used explicitly for object detection and classification systems, including for intelligent transport systems, such as autonomous driving or self-driving cars [30], traffic monitoring [31,32], and advanced driver assistance systems [33,34].…”

Modelling Eco-Driving Support System for Microscopic Traffic Simulation

“…Drivers were for example found to deactivate the ACC in congested traffic or when overtaking. A similar approach was also used in [32,37] to model truck driver's de-and reactivation of a fuel minimising cruise controller for trucks. Real truck drivers were for example found to sometimes deactivate the fuel minimising cruise controllers during overtakings when the cruise controller decreased the speed to save fuel.…”

“…in 30]), or by explicitly using separate models of the support system and the interaction of the driver with the support system [as e.g. in 32,37]. While the former may be sufficient for investigations of the effects of driver support systems for existing traffic situations, it is not appropriate for use in investigating future traffic scenarios where both penetration and compliance rates are likely to differ from the existing scenario.…”