Matching Items (43)
Filtering by
Description
Traditional Reinforcement Learning (RL) assumes to learn policies with respect to reward available from the environment but sometimes learning in a complex domain requires wisdom which comes from a wide range of experience. In behavior based robotics, it is observed that a complex behavior can be described by a combination of simpler behaviors. It is tempting to apply similar idea such that simpler behaviors can be combined in a meaningful way to tailor the complex combination. Such an approach would enable faster learning and modular design of behaviors. Complex behaviors can be combined with other behaviors to create even more advanced behaviors resulting in a rich set of possibilities. Similar to RL, combined behavior can keep evolving by interacting with the environment. The requirement of this method is to specify a reasonable set of simple behaviors. In this research, I present an algorithm that aims at combining behavior such that the resulting behavior has characteristics of each individual behavior. This approach has been inspired by behavior based robotics, such as the subsumption architecture and motor schema-based design. The combination algorithm outputs n weights to combine behaviors linearly. The weights are state dependent and change dynamically at every step in an episode. This idea is tested on discrete and continuous environments like OpenAI’s “Lunar Lander” and “Biped Walker”. Results are compared with related domains like Multi-objective RL, Hierarchical RL, Transfer learning, and basic RL. It is observed that the combination of behaviors is a novel way of learning which helps the agent achieve required characteristics. A combination is learned for a given state and so the agent is able to learn faster in an efficient manner compared to other similar approaches. Agent beautifully demonstrates characteristics of multiple behaviors which helps the agent to learn and adapt to the environment. Future directions are also suggested as possible extensions to this research.
ContributorsVora, Kevin Jatin (Author) / Zhang, Yu (Thesis advisor) / Yang, Yezhou (Committee member) / Praharaj, Sarbeswar (Committee member) / Arizona State University (Publisher)
Created2021
Description
Equity concerning food access has gained a lot of attention in the past decades. This problem can be seen in the dearth of supermarkets offering healthy food at reasonable prices in disadvantaged neighborhoods. Numerous studies show that the disparity in the distribution of food outlets has resulted in disparities in health outcomes. To mitigate the issue, various intervention strategies have been proposed and implemented, including introducing new supermarkets, mobile food markets, community gardens, and city farms in these neighborhoods. Among these strategies, mobile food markets have gained the attention of practitioners and policymakers for their low costs and service flexibility. Challenges remain in identifying the sites for best serving the people in need given limited resources. In this study, a new spatial optimization model is proposed to determine the best locations for mobile food markets in the City of Phoenix. The new model aims to cover the largest number of people with food access challenges while minimizing transportation costs. Compared with the existing mobile market sites, the sites provided by the new model can increase the coverage of low-food access residents with a shorter transportation distance. The new model has also been applied to help expand the service provider of the existing mobile food markets. In addition to mobile food markets, the method provided in this study can be extended to support the planning of other food outlets and food assistance services.
ContributorsLu, Junzhou (Author) / Tong, Daoqin (Thesis advisor) / Connor, Dylan (Committee member) / Kuby, Michael (Committee member) / Arizona State University (Publisher)
Created2022
Description
This dissertation consists of three chapters that investigate the rapid adoption and complex implementation of city commitments to transition to 100% renewable energy (100RE). The first paper uses a two-stage, mixed methods approach to examine 100RE commitments across the US, combining a multivariate regression of demographic, institutional, and policy factors in adoption and six interview-based state case studies to discuss implementation. Adoption of this non-binding commitment progressed rapidly for city councils around the US. Results show that many cities passed 100RE commitments with no implementation plan and minimal understanding of implementation challenges. This analysis highlights that many cities will need new institutions and administrative capacities for successful implementation of these ambitious new policies. While many cities abandoned the commitment soon after adoption, collaboration allowed cities in a few states to break through and pursue implementation, examined further in the next two studies. The second paper is a qualitative case study examining policymaking for the Utah Community Renewable Energy Act. Process tracing methods are used to identify causal factors in enacting this legislation at the state level and complementary resolutions at the local level. This Act was passed through the leadership and financial backing of major cities and committed the investor-owned utility to fulfill any city 100RE resolutions passed through 2019. Finally, the third paper is a mixed-methods, descriptive case study of the benefits of Community Choice Aggregation (CCA) in California, which many cities are using to fulfill their 100RE commitments. Cities have adopted CCAs to increase their local voice in the energy process, while fulfilling climate and energy goals. Overall, this research shows that change in the investor-owned utility electricity system is in fact possible from the city scale, though many cities will need institutional innovation to implement these policies and achieve the change they desire. While cities with greater resources are better positioned to make an impact, smaller cities can collaborate to similarly influence the energy system. Communities are interested in lowering energy costs for customers where possible, but the central motivations in these cases were the pursuit of sustainability and increasing local voice in energy decision-making.
ContributorsKunkel, Leah Christine (Author) / Breetz, Hanna L (Thesis advisor) / Parker, Nathan (Committee member) / Salon, Deborah (Committee member) / Arizona State University (Publisher)
Created2022
Description
With the acceleration of urbanization in many parts of the world, transportation challenges such as traffic congestion, increasing carbon emissions, and the “first/last-mile” connectivity problems for commuter travel have arisen. Transport experts and policymakers have proposed shared transportation, such as dockless e-scooters and bike-sharing programs, to solve some of these urban transportation issues. In cities with high population densities, multimodal mobility hubs designed to integrate shared and public transportation can be implemented to achieve faster public connections and thus increase access to public transport on both access and egress sides. However, haphazard drop-offs of these dockless vehicles have led to complaints from community members and motivated the need for neighborhood-level parking areas (NLPAs). Simultaneously, concerns about the equitable distribution of transportation infrastructure have been growing and have led to the Biden Administration announcing the Justice40 Initiative which requires 40% of certain federal investments to benefit disadvantaged communities. To plan a system of NLPAs to address not only the transportation shortcomings while elevating these recent equity goals, this thesis develops a multi-objective optimal facility location model that maximizes coverage of both residential areas and transit stations while including a novel constraint to satisfy the requirements of Justice40. The model is applied to the City of Tempe, Arizona, and uses GIS data and spatial analyses of the existing public transportation stops, estimates of transit station boardings, population by census block, and locations of disadvantaged communities to optimize NLPA location. The model generates Pareto optimal tradeoff curves for different numbers of NLPAs to find the non-dominated solutions for the coverage of population nodes and boardings. The analysis solves the multi-objective model with and without the equity constraint, showing the effect of considering equity in developing a multimodal hub system, especially for disadvantaged communities. The proposed model can provide a decision support tool for transport and public authorities to plan future investments and facilitate multimodal transport.
ContributorsQuan, Hejun (Author) / Kuby, Michael (Thesis advisor) / Frazier, Amy (Thesis advisor) / Tong, Daoqin (Committee member) / Arizona State University (Publisher)
Created2022
Description
With rapid advances in technology development and public adoption, it is crucial to understand how these services will shape the future of travel depending on the extent to which people will use these services; impact the transportation and infrastructure systems such as changes in the use of transit and active modes of travel; and influence how technology developers create and update these transportation technologies to better serve people’s mobility needs. This dissertation explores how two major emerging services, namely ridehailing services and autonomous vehicles (AVs), will be used in the future when they are widely available and vastly used, and how they may impact the transportation infrastructure and societal travel patterns. The four proposed chapters use comprehensive quantitative and qualitative methods to explore the status of these technologies from theory, through robust modeling frameworks, to practice, by investigating the recent AV pilot deployments in real-world settings. In the second chapter, it was found that increased frequency of ridehailing use is significantly associated with a decrease in bus usage, suggesting that ridehailing functions more as a substitute for buses than as a complement and implying that transit agencies should explore ways to incorporate ridehailing services in their plans to enhance transit usage. Next, the third chapter showed that interest in using AVs for running errands had a positive and significant effect on AV ownership intent, even after accounting for a host of variables. The fourth chapter depicted how ridehailing experiences have a considerable effect on the willingness to ride AV-based services in both private and shared modes, suggesting that experience is crucial for future adoption of these services. Then, two recent real-world AV experiences are explored in the fifth chapter. Lessons learned from these experiments reinforced the importance of first-hand experiences in promoting AV awareness and trustworthiness, potentially leading to greater degrees of adoption. Finally, the results and discussions presented in this dissertation strengthen the body of literature on key emerging transportation technologies and inform policymakers and stakeholders to properly prepare cities and the public to welcome these technologies into our transportation system in an efficient, equitable, and complementary way.
ContributorsMagassy, Tassio Bezerra (Author) / Pendyala, Ram M (Thesis advisor) / Khoeini, Sara (Committee member) / Polzin, Steven E (Committee member) / Salon, Deborah (Committee member) / Arizona State University (Publisher)
Created2023
Description
Transportation infrastructure facilitates humans in moving themselves and material goods, and thereby supports the functioning of human society. Transportation planners, engineers, and decision makers in the 20th century largely excluded local stakeholders from planning processes; the resultant built environment has perpetuated inequity and social division. Transportation system planning has often been conducted in specialized departments with little interdisciplinary collaboration. Integration of diverse perspectives and ontologies throughout transportation planning processes can produce robust, resilient, equitable, and sustainable transportation systems. Geodesign is a framework for planning the built environment that necessarily involves voices from multiple perspectives including local stakeholders, design professionals, geographic scientists, and information technology coordinators. Geodesign uses geographic information systems to create designs that reflect stakeholder needs, values, and priorities while addressing the study area’s geographic context. Geodesign has been used primarily for land use planning and has only addressed transportation planning concerns in relation to land use.This dissertation consists of an introduction, three projects that apply the geodesign framework to transportation planning and a concluding chapter. The introduction details the rationale for this research. The first project is a systematic review of geodesign projects that address transportation systems. The review seeks to identify epistemological alignment between the geodesign framework and participatory transportation planning. The results demonstrate that geodesign comports with transportation planners’ existing practices and uses of planning support systems. The combination of geodesign and transportation planning methods for stakeholder engagement could produce a synergistic framework for transportation infrastructure planning. The second project applies geodesign to locating refueling stations for hydrogen fuel cell vehicles around Hartford, Connecticut. Network designs generated by workshop participants were compared to networks generated by optimization models. The third project applies geodesign to locating sites for micromobility hubs in Tempe, Arizona, via short-form workshop series format. Participants considered the format conducive to collaborative public participatory design. These three projects demonstrate the suitability of the geodesign framework for node-based transportation facility planning via communicative rationality. The conclusion summarizes these three projects and highlights the reproducibility of the geodesign method for node-based transportation facility location planning in other study areas.
ContributorsLopez Jaramillo, Oscar (Author) / Kuby, Michael (Thesis advisor) / Wentz, Elizabeth (Committee member) / Ruddell, Darren (Committee member) / Arizona State University (Publisher)
Created2023
Description
Recently the domestic aviation industry has been influenced by rapidly growing ultra low-cost carriers (ULCCs). The pattern of airport markets served by ULCCs is incongruous with legacy carriers and low-cost airlines alike. Existing literature, however, is limited for North American ULCCs: research has only recently begun to identify them separately from mainstream low-cost carriers. This study sought to understand the market factors that influence ULCC service decisions. The relationship between ULCC operations and airport market factors was analyzed using three methods: mapping 2019 flight data for four ULCCs combined, two regression analyses to evaluate variables, and three case studies examining distinct scenarios through interviews with airport managers. Enplanement data were assembled for every domestic airport offering scheduled service in 2019. Independent variables were collected for each Part 139 airport. The first model estimated an ordinary least squares regression model to analyze ULCC enplanements. The second model estimated a binary logistic equation for presence of ULCC service. Case studies for Bellingham, Waco, and Lincoln were selected using compelling airport factors and relevant ULCC experience. Maps of ULCC enplanements revealed concentrations of operations on the East Coast. Both regression analyses showed strong relationships between population and non-ULCC enplanements (two measures of airport market size) and ULCC operations. A significant relationship also existed between tourism and enplanements. In the logit model, distance and competition variables were associated with ULCC presence. Case studies emphasized the importance of airport fees and competition in ULCC preferences, although aeronautical costs were generally not significant in the regressions.
ContributorsTaplin, Drew (Author) / Kuby, Michael (Author) / Salon, Deborah (Author) / King, David A. (Author)
Created2023-01-31
Description
Transportation network connectivity has been linked to positive urban outcomes, including increased rates of active transportation, reduced reliance on automobiles and other social and economic benefits. While many stakeholders in greenfield development processes have emphasized the positive benefits of connectivity and connectivity has increased in many U.S. metros in the past two decades, many street networks remain fragmented and local connectivity remains far below that of historic patterns. This paper explores barriers to and influences on connectivity outcomes in new community construction in the Phoenix metropolitan area, employing mixed qualitative and quantitative methods. Interviews were conducted with members from various stakeholder groups in the subdivision development process. Case studies were developed with space syntax and network analysis measurements to illustrate the influence of variables and stakeholders on the planning process. Participants illustrated a complex political and economic reality surrounding the concept of connectivity, with site conditions and development market dynamics playing the clearest roles in shaping connectivity. The result is subdivisions are achieving moderate levels of connectivity and improving from historic patterns of dendricity but remain entrenched in planning paradigms built around self-contained sites and the policy and market limitations for robust connectivity beyond individual developments.
ContributorsSchumerth, Noah John (Author) / King, David A (Thesis advisor) / Kuby, Michael (Committee member) / Lindsay, Lucas (Committee member) / Arizona State University (Publisher)
Created2022
Description
Public transportation is considered a solution to congestion and a tool for reducing greenhouse gas emissions. It is becoming popular even in cities with the harshest climate conditions as these cities grow rapidly and are trying to provide sustainable alternatives for their vehicle-oriented communities. A lot must be taken into consideration whendesigning transit systems to reduce riders' vulnerability to heat in cities with high temperatures averaging 40°C during the summer and humidity levels reaching 90 percent. Using transit systems in Dubai, United Arab Emirates, and Phoenix Metropolitan, United States, as case studies, this paper focuses on both qualitative and quantitative research methods to observe the built environment around public transit stations and measure the temperatures and humidity levels to compare with the experienced temperatures and the built environment observations. The results show that the design of transit stations and the public realm significantly impacts a rider's experience. The findings show that passive cooling, shading, and vegetation as the best practices in the two case studies. Both transit systems have certain elements that work efficiently and other elements that need improvement to provide a better rider experience. Identifying these best practices helps develop recommendations for the future of designing transit systems in desert cities worldwide.
ContributorsAlbastaki, Mohamed (Author) / King, David (Thesis advisor) / Salon, Deborah (Committee member) / Kelley, Jason (Committee member) / Arizona State University (Publisher)
Created2022
Description
Much of modern urban planning in the United States is concerned with making cities more walkable. However, this is occurring as the urban landscape of the U.S. is altered radically by changes in crime patterns after the summer of 2020. This paper seeks to find out what the relationship is between walkability and crime in major U.S. cities after 2020. Using multiple linear regressions at the city and neighborhood scale, walkability is found to be a significant, positive predictor of 2019 violent crime rate, 2020 violent crime rate, 2020 property crime rate, and 2020 total crime rate at the city level. It was found to be a positive, but not significant predictor at the neighborhood level. Walkability has no protective influence against crime/rising crime, and it appears that as crime gets worse it tends to get worse in the cities that are more walkable, but other variables such as African American population are better determinants of crime. Urban planners should seek to increase walkability while also finding a way to mitigate potential exposure to crime.
ContributorsWeisman, Holden (Author) / Kelley, Jason (Thesis director) / Salon, Deborah (Committee member) / Barrett, The Honors College (Contributor) / School of Geographical Sciences and Urban Planning (Contributor)
Created2023-05