No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Sweden has defined a national transportation objective that aim at making public transportation the most attractive option in more situations. One way of making public transportation mote attractive is to ensure that it is rapid. BRT stands for Bus Rapid Transit and is concept that ensures speed and reliability of public transportation buses through combined priority measures. However, accessibility measures for urban buses are often compromised and not prioritized politically. If the travel time saved for these measures can be estimated, then they are more likely to be implemented. Two components that have been defined for Swedish BRT were tested on a proposed BRT corridor in the city of Södertälje through microsimulation tool VISSIM. The two components that were tested was boarding and alighting through all doors and bus lanes, both median and curb side. The components were tested on a 3,2 km long corridor using the current actuated traffic signal that was currently in use. To test the measures a sensitivity analysis was carried out with three different levels of traffic and five different passenger levels for boarding and alighting. Only the impact on travel time for the bus was studied. The travel time savings from the measure boarding and alighting through all doors was 1 to 3 percent but only significant for the higher levels of boarding with an average of 12 to 15 boarding passengers per stop. The travel time savings of the curb side bus lane with actuated traffic signals was estimated to be up to 5 percent and for the median bus lane up to 11 percent in congested traffic conditions.

Machine perception is a key element for the research on autonomous driving vehicles. In particular, we focus on the problem of lane detection with a single camera. Many lane detection systems have been developed and many algorithms have been published over the years. However, while they are already commercially available to deliver lane departure warnings, their reliability is still unsatisfactory for fully autonomous scenarios. In this work, we questioned the reasons for such limitations. After examining the state of the art and the relevant literature, we identified the key methodologies adopted. We present a self-standing discussion of bird’s eye view (BEV) warping and common image preprocessing techniques, followed by gradient-based and color-based feature extraction and selection. Line fitting algorithms are then described, including least squares methods, Hough transform and random sample consensus (RANSAC). Polynomial and spline models are considered. As a result, a general processing pipeline emerged. We further analyzed each key technique by implementing it and performing experiments using data we previously collected. At the end of our evaluation, we designed and developed an overall system, finally studying its behavior. This analysis allowed us on one hand to gain insight into the reasons holding back present systems, and on the other to propose future developments in those directions. / Thesis / Master of Science (MSc)

Driving field experiment data were used to investigate lane changes in which a slow lead vehicle was present to: 1) characterize lane changes, 2) develop predictive models, 3) provide collision avoidance system (CAS) design guidelines. A total of 3,227 slow lead vehicle lane changes over 23,949 miles were completed by sixteen commuters. Two instrumented vehicles, a sedan and an SUV, were outfitted with video, sensor, and radar data systems that collected data in an unobtrusive manner. Results indicate that 37.2% of lane changes are slow lead vehicle lane changes, with a mean completion time of 6.3 s; most slow lead vehicle lane changes are leftward, rated low in urgency and severity. A stratified sample of 120 lane changes was selected to include a range of maneuvers. On the interstate, lane changes are performed less often, <i>t</i>(30) = 2.83, <i>p</i> = 0.008, with lower urgency ratings, <i>F</i>(1, 31) = 5.24, <i>p</i> = 0.05, as compared to highway lane changes, as interstates are designed for smooth flow. Drivers who usually drive sedans are more likely to make lane changes than drivers of SUVs, <i>X</i> ²⁺(1)= 99.6247, <i>p</i> < 0.0001, suggesting that driving style is maintained regardless of which experimental vehicle is driven. Turn signals are used 64% of the time but some drivers signal after the lane change starts. Of cases in which signals are not used, 70% of them are made with other vehicles nearby. Eyeglance analysis revealed that the forward view, rearview mirror, and left mirror are the most likely glance locations. There are also distinct eyeglance patterns for lane changing and baseline driving. Recommendations are to use forward view or mirror-based visual displays to indicate presence detection, and auditory displays for imminent warnings. The "vehicle + signal" logistic regression model is best overall since it takes advantage of the distance to the front and rear adjacent vehicle, forward time-to-collision (TTC), and turn signal activation. The use of additional regressors would also improve the model. Five design guidelines are included to aid in the development of CAS that are useable, safe, and integrated with other systems, given testing and development. / Ph. D.

The problem of roadway navigation and obstacle avoidance for unmanned ground vehicles has typically needed very expensive sensing to operate properly. To reduce the cost of sensing, it is proposed that an algorithm be developed that uses a single visual camera to image the roadway, determine where the lane of travel is in the image, and segment that lane. The algorithm would need to be as accurate as current lane finding algorithms as well as faster than a standard k- means segmentation across the entire image. This algorithm, named IRIS, was developed and tested on several sets of roadway images. The algorithm was tested for its accuracy and speed, and was found to be better than 86% accurate across all data sets for an optimal choice of algorithm parameters. IRIS was also found to be faster than a k-means segmentation across the entire image. IRIS was found to be adequate for fulfilling the design goals for the algorithm. IRIS is a feasible system for lane identification and segmentation, but it is not currently a viable system. More work to increase the speed of the algorithm and the accuracy of lane detection and to extend the inherent lane model to more complex road types is needed. IRIS represents a significant step forward in the single camera roadway perception field. / Master of Science