Machine learning is a powerful method to handle the self-driving problem. Researchers use machine learning to construct a neural network and train it to drive the car. A self-driving car is a safety-critical system. However, the neural network is not necessarily reliable. The output of a neural network can be easily influenced by many factors, such as the quality of training data and the runtime environment. Also, it takes time for the neural network to generate the output. That is, the self-driving car may not respond in time. Such weaknesses will increase the risk of accidents. In this thesis, considering the safety of self-driving cars, we apply a delay-aware shielding mechanism to the neural network to protect the self-driving car. Our approach is an improvement based on previous research on runtime safety enforcement for general cyber-physical systems that did not consider the delay to generate the output. Our approach contains two steps. The first is to use formal language to specify the safety properties of the system. The second step is to synthesize the specifications into a delay-aware enforcer called the shield, which enforces the violated output to satisfy the specifications during the whole delay. We use a lane keeping system as a small but representative case study to evaluate our approach. We utilize an end-to-end neural network as a typical implementation of such a lane keeping system. Our shield supervises those outputs of the neural network and verifies the safety properties during the whole delay period with a prediction. The shield can correct it if a violation exists. We use a 1/16 scale truck and construct a curvy lane to test our approach. We conduct the experiments both on a simulator and a real road to evaluate the performance of our proposed safety mechanism. The result shows the effectiveness of our approach. We improve the safety of a self-driving car and we will consider more comprehensive driving scenarios and safety features in the future. / Master of Science / Self-driving cars is a hot topic nowadays. Machine learning is a popular method to achieve self-driving cars. Machine learning constructs a neural network, which imitates a human driver's behavior to drive the car. However, a neural network is not necessarily reliable. Many things can mislead the neural network into making wrong decisions, such as insufficient training data or a complex driving environment. Thus, we need to guarantee the safety of self-driving cars. We are inspired to use formal language to specify the safety properties of the self-driving system. A system should always follow those specifications. Then the specifications are synthesized into an enforcer called the shield. When the system's output violates the specifications, the shield will modify the output to satisfy the specifications. Nevertheless, there is a problem with state-of-the-art research on specifications. When the specifications are synthesized into a shield, it does not consider the delay to compute the output. As a result, the specifications may not be always satisfied during the period of the delay. To solve such a problem, we propose a delay-aware shielding mechanism to continually protect the self-driving system. We use a lane keeping system as a small self-driving case study. We evaluate the effectiveness of our approach both on the simulation platform and the hardware platform. The experiments show that the safety of our self-driving car is enhanced. We intend to study more comprehensive driving scenarios and safety features in the future.

This bachelor thesis examines the drivers of secondhand shopping in the fashion industry. The thesis addresses the complexity of consumer behavior and attitudes towards secondhand shopping. Through focus group interviews with consumers, the thesis has identified a gap between positive attitudes towards secondhand shopping and actual consumption behavior. The thesis reveals that economic benefits, sustainability considerations and a desire for a unique style are motivations for consumers to buy secondhand clothing. However, concerns about quality, accessibility, convenience, and store atmosphere are factors that influence consumers' purchasing decisions. To address the identified barriers and overcome the gap between attitude and action, the thesis suggests concrete improvements proposals, such as increasing the transparency of online shopping experiences, improving the cleanliness and atmosphere of the shops, and leveraging influencer marketing to reshape the negative perception that exists of secondhand. The thesis also points out several strategies that can be implemented to convey a message of increasing the acceptance and popularity of secondhand consumption, which ultimately contributes to promoting sustainability and reducing the environmental impact of the fashion industry.

The demand for in-vehicle information systems (IVIS) has been increasing through the years. There are numerous systems that can be incorporated into vehicles and various ways in which the information can and should be presented to the driver. The way the information is presented to the drivers is extremely important in terms of increasing safety and decreasing driver distraction. The expected outcomes of this research included the development of human factors guidelines for the design and use of in-vehicle information systems. It was a desirable goal to identify the most suitable information presentation formats for certain tasks, since this may influence the drivers' attention and driving performance. This study focused on how the factors of interest may affect drivers' attention and driving performance while performing IVIS secondary tasks related to specific applications. This was accomplished through an on-road within-factors experiment. Sixteen participants performed secondary tasks related to three IVIS applications at two levels of difficulty. The tasks were presented using five types of displays. Data collected from video and in-vehicle sensors were statistically analyzed to determine significant effects between the factors. Driving performance, external reaction time, and perceived mental workload results were compiled into general guidelines for the design and use of IVIS. The findings of this study strongly suggest that visual displays should not be used for the presentation of IVIS. Auditory and multi-modal (i.e. both visual and auditory interface) displays are the most appropriate ways to present IVIS information. A normal speech rate is preferred over a fast speech rate. IVIS tasks should be kept as simple as possible in terms of the number of steps. From the three manipulated factors (type of display, IVIS application, and task level of difficulty), the type of display had the largest number of significant results across the dependent variables measurements. The visual display led to the worst driver performance, while auditory and multi-modal displays yielded significantly better driving performance. / Master of Science

Variable speed limits (VSL) are dynamic traffic management systems designed to increase the efficiency and safety of highways. While the macroscopic performance of VSL systems is well explored in the existing literature, there is a need to further understand the microscopic behavior of vehicles driving in VSL zones. Specifically, driver compliance to advisory VSL systems is quantified based on a driving-simulation experiment and introduced into a broader microscopic behavior model. Statistical analysis indicates that VSL compliance can be predicted based upon several VSL design parameters. The developed two-state microscopic model is calibrated to driving-simulation trajectory data. A calibrated VSL microscopic model can be utilized for new VSL control and macroscopic performance studies, adding an increased dimension of realism to simulation work. As an example, the microscopic model is implemented within VISSIM (overriding the default car-following model) and utilized for a safety-mobility performance assessment of an incident-responsive VSL control algorithm implemented in a MATLAB COM interface. Examination of the multi-objective optimization frontier reveals an inverse relationship between safety and mobility under different control algorithm parameters. Engineers are thus faced with a decision between performing multi-objective optimization and selecting a dominant VSL control objective (e.g. maximizing safety versus mobility performance). / Master of Science / Variable speed limits (VSL) are dynamic traffic management systems designed to increase the efficiency and safety of highways. While the system performance of VSL systems is well explored in previous research, there is a need to further understand the individual behavior of vehicles driving under VSL control. Specifically, driver compliance to advisory VSL systems is modelled based on a driving-simulation experiment. Low compliance equates to poor VSL performance so it is important for engineers to have the ability to predict compliance based on VSL design conditions. The compliance model is introduced into a driver behavior model that quantifies and predicts the driver decision process on VSL controlled highways. The driver behavior model parameters are set using data obtained from the driving-simulation experiment. Utilization of the developed driver behavior model will increase the accuracy of future simulation work on VSL system performance. In this study, the model is implemented within a traffic simulation software to conduct an assessment of the trade-offs between safety and mobility VSL performance for different VSL control designs. An accident is modelled in the simulation software, and VSL is utilized to respond to and alleviate the incident. Simulation results indicate an inverse relationship between safety and mobility performance – indicating that engineers must select a primary objective when selecting VSL control design parameters.

No

Driving under the influence of alcohol is one of the biggest public health problems in the United States. A number of theories exist as to why people drive after drinking, one of the dominant being that they believe that they are not legally intoxicated. The purpose of the present study is to determine if people tend to accurately estimate the number of alcoholic beverages they can consume before reaching a blood alcohol concentration (BAC) of .08 percent. Participants consisted of27 college students. Results showed a significant difference between the participants' estimated number of drinks and the actual number of drinks as determined by a Virtual Bar program. This virtual bar delivery system calculates BAC based on gender, weight, and number of drinks in a given time. Links to past research and theory are presented.

Deindividuation is a psychological phenomenon that occurs when a given environment reduces the "individuality" or identifiability of a person. These environments may cause a psychological reduction in self-consciousness, potentially leading to violations of sociocultural norms (Festinger, Pepitone, & Newcomb, 1952; Singer, Brush, & Lublin, 1965). The present research sought to empirically test deindividuation theory among automobile drivers utilizing the anonymizing factor of observation. Participants (N = 31) used a driving simulator and were either in the observed condition or an unobserved condition. Analysis of driving data did not reveal significant results, however self-report data had some interesting trends. Though limited in scope, this research begins to shed light on deindividuation of drivers and may provide a foundation for future research.

A study was performed to evaluate the effectiveness, efficiency, and strategy associated with three navigation methods: memorized route, conventional paper map, and a moving-map navigational display (the navigator). Thirty-two driver-subjects of both genders, and wide ranges of age (18-73) and driving experience (2,000 to 40,000 miles per year) navigated along public roadways for this research using a specially instrumented automobile. A variety of different roadway conditions were also used for this research including limited access four-lane highways, two-lane state routes, and city streets. In addition, the research was conducted under conditions of both light and moderate traffic densities. Measures taken include eye movement, navigation effectiveness, and driving performance measures. Results showed that the paper map took longer to study at the beginning of a run than the navigator. Even with this handicap, the total time taken when using the paper map was not significantly different from the time taken to use the navigator. Also, there were no differences in the directness or quality of routes selected when using either the paper map or the navigator to navigate. These findings were a result of the strategies adopted in the use of the various methods of navigation. During the initial study phase the paper map was essentially used to plan the entire route from start to finish. After the initial phase, the map was used only as an occasional reference. In contrast, effective use of the navigator could only be accomplished by repetitively glancing at the display to acquire important information as it was updated and presented. As a result, subjects spent more driving time glancing to the navigator than the paper map, and it substantially drew the subjects' gaze away from the driving task relative to the norm established in the memorized route condition, as well as in comparison to the paper map. Still, driving performance did not greatly change as a function of navigation method indicating that the additional visual attentional demand associated with the navigator was drawn primarily from spare driver resources. It is also very likely that the novelty of the navigator was responsible for some portion of the glance time spent on it. / Ph. D.

As part of the national effort to reduce drunk driving, servers of alcohol are being held liable for their alcohol-serving behavior with increasing frequency. In response to these drunk driving prevention efforts, the Training for Intervention' Procedures by Servers of Alcohol (TIPS) program was developed to teach servers skills thought necessary for acceptable standards of practice for serving alcoholic beverages. The current study evaluated the potential of this program to aid in decreasing alcoholimpaired driving. Subjects were 17 waiters, waitresses, and bartenders who were employed at two bars. After the baseline period, the servers completed successfully the 6-hour TIPS training course. Research assistants posing as regular patrons (i.e., 'pseudopatrons') visited two target bars throughout the course of the study. These pseudopatrons set the occasion for server intervention to occur by attempting to drink six alcoholic beverages in two hours. Naturalistic data were collected by having a partner with a hidden microphone record all interactions between the server and pseudopatron. The partner also noted any signs of intoxication exhibited by the pseudopatron. The results revealed that servers who had received TIPS training initiated more interventions than untrained personnel. Moreover, pseudopatrons served by TIPS trained personnel had significantly lower blood-alcohol levels and exhibited fewer signs of intoxication than those served by untrained servers. These results suggest that, if implemented on a large scale, the TIPS program has the potential to reduce drunk driving by helping to decrease the exit blood-alcohol levels of bar patrons. The need to investigate the maintenance of server intervention behavior is discussed and other suggestions for future research are presented. / Ph. D.

Work title: The response of the visual system when driving a car in real conditions and in a simulator during a parallel motor task. Aims: To implement a laboratory measurement in a driving simulator and in real conditions during a parallel motor task. To track the response of the visual system during the task and to determine potential differences in the response of the visual system and in the speed in the simulator and in the real car environment. Methods: The laboratory measurement was realized in a driving simulator OCTAVIA II. The measurement in real conditions was realized in Škoda Octavia as well. The eye-tracking method was used to track the drivers` eyes. Results: In the simulator the more experienced driver managed to do the parallel motor task within a shorter period of time than the less experienced driver, lower frequency of looks at the radio during the tasks with the more experienced driver was not proved. There was not an evidence of more balanced speed during the drive with the more experienced driver in the simulator. On the average the driver spent shorter time doing the tasks in the real environment than in the simulator and the total number of looks at the radio was lower. The speed of the drive in the real environment was lower than the speed in the simulator. Key words:...