With the development of digital technology during the last decade and the improvement of Geographic Information Systems (GIS), it has become easier for various scientific fields to predict and extrapolate data. Various organisations and institutions continuously develop algorithms and software to assist with specific challenges in multiple fields of science. These technologies and principles have also been effectively applied in the soil science field of pedology. Traditional soil mapping, although effective, is time consuming, arduous and expensive. It is thus important to develop methods whereby the soil forms of an area can be identified faster while providing accurate information to the reader/ user. Conservation areas, such as Phinda Private Game Reserve (Phinda), which covers a large area (greater than 28 000 ha) can benefit from a soil map. The importance of a proper soil map has a great many uses in conservation, but not every organisation or individual can afford it. This is where digital soil mapping (DSM) or Predictive Soil Mapping (PSM) comes into its own. Substantial research and development have been done in the form of methodology and software systems for DSM although it has not been effectively applied to conservation management. By applying these techniques, accurate and interactive soil maps were developed without the burdensome expenses or dangers associated with traditional soil observations in a conservation area. The application of DSM and the use of the soil land inference model (SoLIM) at Phinda resulted in maps based on the Fey soil-form classification as well as a soil sensitivity index (SSI). The SSI was developed based on the various soil forms present at Phinda and the factors that determine its sensitivity to various types of degradation. These digital maps indicated accuracies of 71% (Fey classification) and 72% for the SSI. The kappa values indicated a substantial agreement (0.63) for the Fey classification map and a moderate agreement (0.57) for the SSI map. The SSI was then combined with the predator sightings and the location of infrastructure and commercial lodges to derive the agreement of activities, game drives, which includes off-road driving (ORD) on sensitive soils. As erosion is a concerning problem, predominantly caused by human activities within Phinda, it was necessary to use the SSI map as a base of comparison. This digitally produced soil map will be presented to the conservation management at Phinda whereby planning can be conducted, literally, from the ground up. Proper planning will thus prevent a loss of soil and consequently a loss of biodiversity. All the information was then combined to developed recommendations for Phinda as to improve the overall road network by upgrading, removing and rehabilitating certain roads and provide advice concerning ORD. These decisions, in turn, prevent long-term soil and biodiversity loss while still providing clients with a true African bush experience. / College of Agriculture and Environmental Sciences

Indiana University-Purdue University Indianapolis (IUPUI) / Drinking and driving is increasingly becoming a detrimental behavior, especially amongst college-aged students in the U.S. and other countries. Additionally, research shows that college-age students in the U.S. are more likely to drink and drive, than college-age students in Germany. Fishbein and Ajzen’s Theory of Reasoned Action asserts that subjective norms and attitudes signify behavioral intentions. In order to test the TRA and understand the drinking and driving differences and similarities in the U.S. and Germany, focus groups of German and American college-age students were conducted to discuss subjective norms and attitudes surrounding drinking and driving behaviors, followed up by an electronic pilot study survey regarding same. The data collected illustrated that college-age drinking and driving is occurs more frequently in the U.S., and that American and German students differ in their attitudes and subjective norms surrounding drinking and driving. Future research would benefit the continued use and circulation of the electronic surveys for larger cross-cultural samples of college-age students to more effectively and quantitatively assess actual drinking and driving behaviors as it relates to subjective norms and attitudes, as suggested in the TRA.

Indiana University-Purdue University Indianapolis (IUPUI) / Self-driving is simply the capacity of a vehicle to drive itself without human intervention. To accomplish this, the vehicle utilizes mechanical and electronic parts, sensors, actuators and an AI computer. The on-board PC runs advanced programming, which permits the vehicle to see and comprehend its current circumstance dependent on sensor input, limit itself in that climate and plan the ideal course from point A to point B. Independent driving is not an easy task, and to create self-sufficient driving arrangements is an exceptionally significant ability in the present programming designing field. ROS is a robust and versatile communication middle ware (framework) tailored and widely used for robotics applications. This thesis work intends to show how ROS could be used to create independent driving programming by investigating self-governing driving issues, looking at existing arrangements and building up a model vehicle utilizing ROS. The main focus of this thesis is to develop and implement a one-tenth scale of an autonomous RACECAR equipped with Jetson Nano board as the on-board computer, PCA9685 as PWM driver, sensors, and a ROS based software architecture. Finally, by following the methods presented in this thesis, it is conceivable to build an autonomous RACECAR that runs on ROS. By following the means portrayed in this theory of work, it is conceivable to build up a self-governing vehicle.

Indiana University-Purdue University Indianapolis (IUPUI) / The current technologies employed for autonomous driving provide tremendous performance and results, but the technology itself is far from mature and relatively expensive. Some of the most commonly used components for autonomous driving include LiDAR, cameras, radar, and ultrasonic sensors. Sensors like such are usually high-priced and often require a tremendous amount of computational power in order to process the gathered data. Many car manufacturers consider cameras to be a low-cost alternative to some other costly sensors, but camera based sensors alone are prone to fatal perception errors. In many cases, adverse weather and night-time conditions hinder the performance of some vision based sensors. In order for a sensor to be a reliable source of data, the difference between actual data values and measured or perceived values should be as low as possible. Lowering the number of sensors used provides more economic freedom to invest in the reliability of the components used. This thesis provides an alternative approach to the current autonomous driving methodologies by utilizing acoustic signatures of moving objects. This approach makes use of a microphone array to collect and process acoustic signatures captured for simultaneous localization and mapping (SLAM). Rather than using numerous sensors to gather information about the surroundings that are beyond the reach of the user, this method investigates the benefits of considering the sound waves of different objects around the host vehicle for SLAM. The components used in this model are cost-efficient and generate data that is easy to process without requiring high processing power. The results prove that there are benefits in pursuing this approach in terms of cost efficiency and low computational power. The functionality of the model is demonstrated using MATLAB for data collection and testing.

With the advancement of autonomous driving research, 3D detection based on LiDAR point cloud has gradually become one of the top research topics in the field of artificial intelligence. Compared with RGB cameras, LiDAR point cloud can provide depth information, while RGB images can provide denser resolution. Features from LiDAR and cameras are considered to be complementary. However, due to the sparsity of the LiDAR point clouds, a dense and accurate RGB/3D projective relationship is difficult to establish especially for distant scene points. Recent works try to solve this problem by designing a network that learns missing points or dense point density distribution to compensate for the sparsity of the LiDAR point cloud. During the master’s exploration, we consider addressing this problem from two aspects. The first is to design a GAN(Generative Adversarial Network)-based module to reconstruct point clouds, and the second is to apply regional point cloud enhancement based on motion maps. In the first aspect, we propose to use an imagine-and-locate process, called UYI. The objective of this module is to improve the point cloud quality and is independent of the detection stage used for inference. We accomplish this task through a GAN-based cross-modality module that uses image as input to infer a dense LiDAR shape. In another aspect, inspired by the attention mechanism of human eyes, we use motion maps to perform random augmentation on point clouds in a targeted manner named motion map-assisted enhancement, MAE. Boosted by our UYI and MAE module, our experiments show a significant performance improvement in all tested baseline models. In fact, benefiting from the plug-and-play characteristics of our module, we were able to push the performance of the existing state-of-the-art model to a new height. Our method not only has made great progress in the detection performance of vehicle objects but also achieved an even bigger leap forward in the pedestrian category. In future research, we will continue to explore the feasibility of spatio-temporal correlation methods in 3D detection, and 3D detection related to motion information extraction could be a promising direction.

This study investigated effects of pilot-hole diameter, screw penetration depth depth, embedded screw orientation, magnitude of pushing-down force and turning speed on screw driving torques in particleboard (PB) and predicted screw driving torques for two PB clamped together. In addition, effects of material mechanical properties such as shear and tensile strength on screw direct withdrawal loads in PB were investigated and a model to predict screw direct withdrawal load capacity was developed. Furthermore, effects of screw driving torques on clamping pressure between two particleboards were investigated. It was concluded in general that stripping torques (STT) per thread for driving screws into tested PB materials had significantly higher means than seating torque (SET) per thread. The SET and STT per thread for screws driven into PB materials with 0.75-inch penetration depth was higher than 0.50-inch depth when no pilot-hole or pilot-holes with diameter of 1/8 in were drilled. Statistical analyses on screw driving torques for two PB materials clamped together, illustrated that the SET and STT tended to increase when the pilot-hole diameter decreased from 11/64 to 1/16 in. The prediction of the SET and STT values of driving screws through a face into an edge of PB materials can be predicted by the connection of two separate PB materials between a single face specimen with countersink in the surface and a single edge specimen. In addition, a face into a face PB connection, the prediction of both SET and STT can be predicted by connection of two separate face specimens. Results on shear and internal bond (IB) strength demonstrated that there was a clear trend that the shear and IB strengths were higher in the surface of the particleboards than the other levels. In the case of screw direct withdrawal (SDW), the results illustrated that the SDW values of particleboard types did not differ from each other when the pilot-hole diameters of 1/8 and 9/64 in used to drive the screw into the material.

An optimal automotive seat backrest angle has not been identified, and currently, no universal method for analyzing sitting discomfort exists. The purposes of this study were to: (1) identify an optimum seat backrest angle or range of angles based on objective and subjective discomfort measures, and (2) evaluate existing methods for analyzing sitting discomfort data. Eight participants (4 male 4 female) completed three, two hour test sessions in a driving simulator. Results showed that subjective and objective measures were moderately correlated. The 120° seat backrest angle (measured from horizontal) resulted in less discomfort than the 105° and 135° seat backrest angles. Time weighted subjective discomfort ratings were the most effective subjective measure of sitting discomfort. Results also indicated that participants were able to identify discomfort differences for few body regions.

No description available.

Pedestrian safety is among one of the largest concerns in the transportation profession. Many treatments have been developed and implemented to improve pedestrian safety. This current research focuses on the efficiency of in-pavement warning lights systems and involves multiple objectives. The primary objective is to evaluate the yielding rates and crosswalk usage of existing and proposed in-pavement lights systems with comparisons including before and after data through a case study approach. A secondary objective is to evaluate where drivers are looking when they approach in-pavement lights systems and develop a model to evaluate their behavior. The research described herein formulated these objectives into two research hypotheses and used statistical evaluation methodologies to provide quantitative and/or qualitative responses to the developed hypotheses. Data on pedestrian and driver behavior in the field, and the interaction between, them was collected using video camera technology in the Amherst, Massachusetts area. Data regarding drivers scan patterns during the approach to a crosswalk with in-pavement warning light system was collected using a driving simulator and an eye tracker. In total, 1,949 non-staged pedestrians and 606 staged pedestrians were observed crossing at the seven crosswalk locations in the field experiment and a total of 32 drivers participated in 576 crosswalk scenarios in the driving simulator evaluation. The field evaluation resulted in increased yielding rates and crosswalk usage after installation of in-pavement warning lights, while driving simulator evaluation resulted in drivers not becoming accustomed to scanning for lights instead of a pedestrian. Recommendations include installation of in-pavement warning lights at traditional, midblock crosswalks and continued exploration of all crosswalks in the driving simulator evaluation.

This research studies travelers' route choice behavior in a driving simulator with real-time information en-route. We investigate whether travelers plan strategically for real-time information en-route or simply select a fixed path from origin to destination at the beginning of a trip, and whether network complexity and a parallel driving task affect subjects' strategic thinking ability. In this study, strategic thinking refers to a traveler's route choice decision taking into account future diversion possibilities downstream enabled by information at the diversion node. All of the subjects in this study participated in driving-simulator-based tests while half of the subjects participated in additional PC-based tests. Three types of maps were used. The first type required a one-time choice at the beginning of a trip to test the traveler's risk attitude. The other two types offered route choices both at the beginning of and during a trip to test the traveler's strategic thinking. The study shows that a significant portion of route choice decisions are strategic in a realistic driving simulator environment. Furthermore, different network complexities impose different cognitive demands on a subject and affect his/her strategic thinking ability. A subject tends to be more strategic in a simple network. Lastly, a parallel driving task does not significantly affect a subject's strategic thinking ability. This seemingly counterintuitive conclusion might be caused by the simplicity of the tested network.