Abstract

In this paper, we propose a solution to reduce the stress level of the driver, minimize fuel consumption and improve safety. The system analyzes the driving style and the driver's workload during the trip while driving. If it discovers an area where the stress increases and the driving style is not appropriate from the point of view of energy efficiency and safety for a particular driver, the location of this area is saved in a shared database. On the other hand, the implemented solution warns a particular user when approaching a region where the driving is difficult (high fuel consumption and stress) using the shared database based on previous recorded knowledge of similar drivers in that area. In this case, the proposal provides an optimal deceleration profile if the vehicle speed is not adequate. Therefore, he or she may adjust the vehicle speed with both a positive impact on the driver workload and fuel consumption. The Data Envelopment Analysis algorithm is used to estimate the efficiency of driving and the driver's workload in in each area. We employ this method because there is no preconceived form on the data in order to calculate the efficiency and stress level. A validation experiment has been conducted using both a driving simulator and a real environment with 12 participants who made 168 driving tests. The system reduced the slowdowns (38%), heart rate (4.70%), and fuel consumption (12.41%) in the real environment. The proposed solution is implemented on Android mobile devices and does not require the installation of infrastructure on the road. It can be installed on any model of vehicle. The research leading to these results has received funding from the “HERMES-SMART DRIVER” project TIN2013- 46801-C4-2-R within the Spanish "Plan Nacional de I+D+I" under the Spanish Ministerio de Economía y Competitividad and from the Spanish Ministerio de Economía y Competitividad funded projects (co-financed by the Fondo Europeo de Desarrollo Regional (FEDER)) IRENE (PT-2012-1036-370000), COMINN (IPT-2012-0883-430000) and REMEDISS (IPT-2012-0882-430000) within the INNPACTO program. Document type: Part of book or chapter of book

Abstract

The increase in the number of deaths due to ground traffic accidents is a global problem. In such context, the development of new vehicular technologies is considered an alternative to improve road safety. Within the field of new vehicle technologies, it is possible to find driver assistance systems. These systems interact in an active or passive way with the driver, reducing their workload by presenting information about their surroundings, which may imply the safer direction of a land vehicle. Taking into account that one of the main reasons for road traffic fatalities in the world is the lane change in a road, hereby we created a cooperative driver assistance system for the lane change, arising from the combination of a kinematic model and a probabilistic graphical one. By combining these two models, we try to improve the response in the assistance of the system, given the direct dependence of the system with a human. Due to the latter, the response of such systems cannot be deterministic in nature. One of the motivations to use probabilistic graphical models is the flexibility of this machines learning technique in modeling the problem addressed in this thesis. In addition to this contribution of applying a specific probabilistic graphical model in our assistance system, other contributions can be found in this thesis, including the development of a Driving simulation platform with a reconfigurable structure. The ability to reconfigure the structure of the driving simulator platform was of great importance for the development and evaluation of the assistance system hereby proposed in each of its stages. In addition, the decision to model a cooperative approach in our assistance system is due to the great potential of the vehicular communications with respect to improving transport safety and efficiency. The moderate cost that is being projected in vehicular communications is another relevant fact. Finally, the description and application of our assistance system model can be considered as a possibility in the area for the development of an application that needs a close response to the reality, based on the uncertainties present in the problem under consideration. O aumento no número de mortes por causa de acidentes de tráfego terrestre é um problema global. No âmbito dessa problemática, o desenvolvimento de novas tecnologias veiculares é considerado uma alternativa para melhorar a segurança viária. Dentro do domínio das novas tecnologias veiculares, é possível encontrar sistemas de assistência ao motorista. Esses sistemas interagem de maneira ativa ou passiva com o motorista, conseguindo reduzir sua carga de trabalho, apresentando informações de seu entorno, o que pode implicar uma direção mais segura de um veículo terrestre. Levando em consideração que um dos principais motivos de mortes por acidentes de tráfego terrestre no mundo é a mudança de faixa em uma pista, neste trabalho, realizamos a tarefa de criar um sistema cooperativo de assistência ao motorista para a mudança de faixa, a partir da combinação de um modelo cinemático e de um modelo gráfico probabilístico. Mediante a combinação destes dois modelos, procuramos melhorar a resposta na assistência por parte do sistema, dada a dependência direta que o sistema tem dos humanos. Por essa última razão, a resposta deste tipo de sistemas não pode ser determinística por natureza. Uma das motivações para utilizar os modelos gráficos probabilísticos é a flexibilidade da técnica de machine learning em modelar o problema abordado nesta tese. Além dessa contribuição de aplicar um modelo gráfico probabilístico específico em nosso sistema de assistência, outras contribuições podem ser encontradas nesta tese, incluindo o desenvolvimento de uma plataforma de simulação para a condução, com uma estrutura reconfigurável. A capacidade de poder reconfigurar a estrutura da plataforma de simulação foi de grande importância para o desenvolvimento e avaliação do sistema de assistência proposto nesta tese, em cada uma de suas etapas. Além disso, a decisão de modelar um enfoque cooperativo, em nosso sistema de assistência, se deve ao grande potencial que tem as comunicações veiculares com respeito à melhora da segurança e da eficiência do transporte. O custo moderado que está sendo projetado nas comunicações veiculares é outro fato relevante. A descrição e aplicação de nosso modelo final podem ser considerados mais uma possibilidade na área para o desenvolvimento de uma aplicação, que precise de uma resposta próxima da realidade, a partir das incertezas presentes no problema considerado.

Abstract

In the field of modern automotive engineering, many researchers are focusing on the development of advanced vehicle control systems such as autonomous vehicle systems and Advanced Driver Assistance Systems (ADAS). Furthermore, Driver Assistance Systems (DAS) such as cruise control, Anti-Lock Braking Systems (ABS), and Electronic Stability Control (ESC) have become widely popular in the automotive industry. Therefore, vehicle control research attracts attention from both academia and industry, and has been an active area of vehicle research for over 30 years, resulting in impressive DAS contributions. Although current vehicle control systems have improved vehicle safety and performance, there is room for improvement for dealing with various situations. The objective of the research is to develop a predictive vehicle control system for improving vehicle safety and performance for autonomous vehicles and ADAS. In order to improve the vehicle control system, the proposed system utilizes information about the upcoming local driving environment such as terrain roughness, elevation grade, bank angle, curvature, and friction. The local driving environment is measured in advance with a terrain measurement system to provide terrain data. Furthermore, in order to obtain the information about road conditions that cannot be measured in advance, this work begins by analyzing the response measurements of a preceding vehicle. The response measurements of a preceding vehicle are acquired through Vehicle-to-Vehicle (V2V) or Vehicle-to-Infrastructure (V2I) communication. The identification method analyzes the response measurements of a preceding vehicle to estimate road data. The estimated road data or the pre-measured road data is used as the upcoming driving environment information for the developed vehicle control system. The metric that objectively quantifies vehicle performance, the Performance Margin, is developed to accomplish the control objectives in an efficient manner. The metric is used as a control reference input and continuously estimated to predict current and future vehicle performance. Next, the predictive control algorithm is developed based on the upcoming driving environment and the performance metric. The developed system predicts future vehicle dynamics states using the upcoming driving environment and the Performance Margin. If the algorithm detects the risks of future vehicle dynamics, the control system intervenes between the driver's input commands based on estimated future vehicle states. The developed control system maintains vehicle handling capabilities based on the results of the prediction by regulating the metric into an acceptable range. By these processes, the developed control system ensures that the vehicle maintains stability consistently, and improves vehicle performance for the near future even if there are undesirable and unexpected driving circumstances. To implement and evaluate the integrated systems of this work, the real-time driving simulator, which uses precise real-world driving environment data, has been developed for advanced high computational vehicle control systems. The developed vehicle control system is implemented in the driving simulator, and the results show that the proposed system is a clear improvement on autonomous vehicle systems and ADAS. Ph. D.

Abstract

This thesis investigated susceptibility of the human brain to auditory signals under different (driving) conditions and the ability of humans to act on alerts in specific contexts In Part I, three experiments were conducted to answer the main question: “How susceptible are humans to auditory stimuli?” To answer this question, we measured electroencephalogram (EEG) Event-Related Potential (ERP) responses in an oddball paradigm in (automated) driving tasks and cognitive load inducing tasks. The main finding was that automated driving and cognitive load inducing tasks can both reduce susceptibility to auditory alerts. Given this observation that auditory susceptibility is sometimes reduced and, more generally, to explore ways to support accurate human behavior while driving, in Part II, we explored interventions that might alleviate the associated potential problems. In the first experiment, we introduced early warnings, or pre-alerts, to allow drivers to anticipate a later acute imperative alert for an upcoming transition of control from the car to the human driver. The results showed that pre-alerts can support safer transitions of control. In the second experiment, we tested an intervention where visual in-car warnings were presented to indicate an upcoming, necessary lane change. Results showed that, on average, drivers were able to change lanes in time. However, at the same time, there are occasionally some drivers that did not respond in a timely manner.

Abstract

New driver support systems are developed and introduced to the market at increasing speed. In conditions of traffic congestion drivers may be supported by a "Congestion Assistant", a system that combines the features of a Congestion Warning System (acoustic warning and gas pedal counterforce) and a Stop & Go system (automatic gas and brake pedal during congestion). To gain understanding of the effects of driving with a Congestion Assistant on drivers, mental workload of drivers was registered under different conditions as well as acceptance of the system. Mental workload was measured by means of physiological registrations, i.e. heart rate, a secondary task and with the aid of subjective scaling techniques. Acceptance was measured with an acceptance scale. The study was carried out in an advanced driving simulator. Driving with the Congestion Assistant while in congestion potentially leads to decreased driver mental workload, whereas just before congestion starts, i.e. developing just noticeable, the system may add to the workload of the driver. Acceptance is generally high after experiencing the system, though not in all respects. (c) 2008 Elsevier Ltd. All rights reserved.

Abstract

utomated truck platooning is getting an increasing interest for its potentially beneficial effects on fuel consumption, driver workload, traffic flow efficiency, and safety. Nevertheless, one major challenge lies in the safe and comfortable transitions of control from the automated system back to the human drivers, especially when they have been inattentive during highly automated driving. In this study, we investigated truck drivers’ take-over response times after a system initiated request to take back control in non-critical truck platooning scenarios. 22 professional truck drivers participated in the truck driving simulator experiment and everyone was instructed to drive under three task conditions during highly automated driving: Driver monitoring condition (drivers were instructed to monitor the surroundings), Driver not-monitoring condition (drivers were provided with a hand-held tablet and were asked to use this), and Eyes-closed condition (drivers were not allowed to open their eyes). The total take-over response time was divided into the perception response time and the movement response time by manual video annotation. Results showed significantly longer total take-over times with high variability in both Driver not-monitoring and Eyes-closed conditions compared to the Driver monitoring condition. Hand movement response time was found to be the dominant component of the total take-over time, being influenced by the motoric manoeuvres to resume physical readiness before taking over control (e.g., putting away the hand-held tablet, or adjusting seating position). These results suggest the importance of a personalized driver readiness predictor as an input parameter for a safe and comfortable transition of control. © 2019 Elsevier Ltd

Abstract

In-vehicle systems that assist the driver with his driving task are developed and introduced to the market at increasing rate. Drivers may be supported during congested traffic conditions by a so-called Congestion Assistant consisting of a mix of informing, assisting and controlling functions. This paper describes the impacts of the Congestion Assistant on the driver in terms of driving behaviour and acceptance. Thirty-seven participants took part in a driving simulator study. The observed driving behaviour showed promising improvements in traffic safety when approaching the traffic jam. Moreover, positive effects of the system on traffic efficiency can be expected in the jam. The participants stated to appreciate the Congestion Assistant, although not all functions were equally rated. To increase the performance and acceptance of the total system, some refinements were suggested. Keywords: Congestion Assistant; Driving simulator; Integrated system; Driving behaviour; Acceptance

Abstract

Co-operative driving with speed adaptation functionality has great potential to improve traffic-throughput, traffic-safety, and environmental-impact on heavily used traffic-infrastructures. A driving-simulator study was performed to investigate the driver behaviour with respect to such driver-support systems (Zero-, Advisory-, Intervention-, and Controlling). This paper describes the results of one specific scenario, a cut-in scenario. The results show a small reaction time, which was smaller than the response-time required for stabilizing the manoeuvre. Subjective measures show that the experienced-effort for the Controlling system was the smallest, that the satisfaction for the Controlling system was the highest and that the usefulness of all systems was positive.