A navegação autônoma é um dos problemas fundamentais na área de robótica móvel. Esse problema vem sendo pesquisado nessa área por décadas e ainda apresenta um grande potencial para pesquisas científicas. A maior parte dos algoritmos e soluções desenvolvidas nessa área foi concebida para que robôs operem em ambientes estruturados. No entanto, outra questão de grande interesse para pesquisadores da área é a navegação em ambientes externos. Em ambientes não estruturado os veículos autônomos (robôs de grande porte) devem ser capazes de desviar de obstáculos, que eventualmente apareçam no caminho. Esta dissertação aborda o desenvolvimento de um sistema inteligente capaz de gerar e executar um planejamento de caminho para o estacionamento de veículos autônomos em ambientes semi-estruturados. O sistema é capaz de reconhecer vagas de estacionamento por meio de sensores instalados no veículo, gerar uma trajetória válida que o conduza até a vaga e enviar os comandos de esterçamento e aceleração que guiam o veículo pelo caminho gerado / Autonomous navigation is one of the fundamental problems in mobile robotics. This problem has been addressed for decades and still has great potential for scientific research. Most solutions and algorithms developed in this field is designed for robots that operate in structured environments. However, another issue of great interest to researchers in this area is autonomous navigation in outdoor environments. In partially structured environments autonomous vehicles (large robots) must be able to avoid obstacles that may arise along the way. This dissertation addresses the development of an intelligent system able to generate and run a path planning for parking of autonomous vehicles in semi-structured environments. The system is able to recognize parking lots using sensors installed in the vehicle, generate a valid path that leads up to the parking lot and send the steering commands and acceleration that to guide the vehicle to its goal point

A navegação autônoma é um dos problemas fundamentais na área de robótica móvel. Esse problema vem sendo pesquisado nessa área por décadas e ainda apresenta um grande potencial para pesquisas científicas. A maior parte dos algoritmos e soluções desenvolvidas nessa área foi concebida para que robôs operem em ambientes estruturados. No entanto, outra questão de grande interesse para pesquisadores da área é a navegação em ambientes externos. Em ambientes não estruturado os veículos autônomos (robôs de grande porte) devem ser capazes de desviar de obstáculos, que eventualmente apareçam no caminho. Esta dissertação aborda o desenvolvimento de um sistema inteligente capaz de gerar e executar um planejamento de caminho para o estacionamento de veículos autônomos em ambientes semi-estruturados. O sistema é capaz de reconhecer vagas de estacionamento por meio de sensores instalados no veículo, gerar uma trajetória válida que o conduza até a vaga e enviar os comandos de esterçamento e aceleração que guiam o veículo pelo caminho gerado / Autonomous navigation is one of the fundamental problems in mobile robotics. This problem has been addressed for decades and still has great potential for scientific research. Most solutions and algorithms developed in this field is designed for robots that operate in structured environments. However, another issue of great interest to researchers in this area is autonomous navigation in outdoor environments. In partially structured environments autonomous vehicles (large robots) must be able to avoid obstacles that may arise along the way. This dissertation addresses the development of an intelligent system able to generate and run a path planning for parking of autonomous vehicles in semi-structured environments. The system is able to recognize parking lots using sensors installed in the vehicle, generate a valid path that leads up to the parking lot and send the steering commands and acceleration that to guide the vehicle to its goal point

This thesis contributes to research that informs the design of autonomous vehicles (AVs). It examines interactions among various types of road users, such as pedestrians and drivers, and describes how findings can contribute to the design of AVs. The work was undertaken as part of a research internship at Nissan Research Center-Silicon Valley on the Human Understanding in Design team. Methods included video ethnography “travel-alongs” which captured the experience of travel from the point of view of drivers and pedestrians, analysis of interaction patterns taken from video of intersections, and analysis of road laws. Findings address the implications of what it will mean for AVs to exist as social entities in a world of varied road contexts, and how AVs might navigate the social act of driving on roads they share with a variety of human users. This thesis contributes to an emerging body of research and application on the subject of the AV in the world.

<div>The goal of this research is to develop a method of control for a team of unmanned aerial and ground robots that is resilient, robust, and scalable given both complete and incomplete information of the environment. The method presented in this paper integrates approximate and optimal methods of path planning integrated with a market-based task allocation strategy. Further work presents a solution to unmanned ground vehicle path planning within the developed mission planning system framework under incomplete information. Deep reinforcement learning is proposed to solve movement through unknown terrain environment. The final demonstration for Advantage-Actor Critic deep reinforcement learning elicits successful implementation of the proposed model.</div>

Transport is one of the basic needs of a functioning society. Unfortunately, transport also pollutes our cities and release greenhouses gases. Driverless technology is a technology predicted to disrupt the future transport system, and perhaps change how we travel from private cars to shared vehicles. This study focuses on the aspect of privately owned versus shared driverless vehicles, to create more knowledge of how the future transport system will look. A utility-based demand model is used to find the demand for private and shared transport when driverless vehicles are available. The utility of different transport options is estimated by looking at earlier studies about the performance of driverless cars, driverless buses and shared driverless taxis, which is used as input for the utility model. The results indicate that driverless technology will not be a catalyst that makes transport go from private to shared. While driverless buses can improve public transport, and shared driverless taxis outcompete current taxis, driverless technology will also improve private vehicles. The results in this study imply that the sustainability improvements earlier reports have predicted with a high use of shared driverless transportation might not materialise unless efforts are done to increase use of shared transportation. / Transport är ett av de grundläggande behoven för ett välfungerande samhälle. På samma gång släpper transporter ut både växthusgaser och skadliga partiklar. Självkörande teknik är något som förväntas revolutionera framtidens transportsystem, förhoppningen är att de ska förändra hur folk reser från privata bilar till delade transporter. Denna studie fokuserar på den förhoppningen. Kommer framtidens transporter ske i privata självkörande fordon eller delade självkörande fordon och vad i sin tur betyder det för framtidens transportsystem? Med en nyttobaserad efterfråge- och bilinnehavsmodell modelleras efterfrågan av självkörande delade taxis, självkörande bussar och självkörande privatbilar. Resultaten indikerar att självkörande teknik inte nödvändigtvis kommer vara en katalysator som får människor att sluta äga och använda privatbilar. Självkörande bussar kan göra kollektivtrafiken bättre, och självkörande delade taxibilar kommer troligtvis användas mer än dagens taxis. Men självkörande privatbilar kommer också ha många fördelar, och de som äger dem kommer dessutom troligtvis köra längre sträckor än dagens bilister. Resultatet av denna rapport indikerar därför att de stora förväntningarna som finns på självkörande teknik gällande delade transporter kan vara felaktiga, om inte andra åtgärder också görs för att öka delning. Att delningen inte ökar gör också att de hållbarhetsförbättringar som vissa tidigare rapporter förutspått inte nödvändigtvis kommer ske.

The landscape of vehicular transportation is quickly shifting as emerging technologies continue to increase in intelligence and complexity. From the introduction of Intelligent Transportation Systems (ITS) to the quickly developing field of Connected and Autonomous Vehicles (CAVs), the transportation industry is experiencing a shift in focus. A move to more autonomous and intelligent transportation systems brings with it a promise of increased equity, efficiency, and safety. However, one aspect that is overlooked in this shift is cybersecurity. As intelligent systems and vehicles have been introduced, a large amount of research has been conducted showing vulnerabilities in them. With a new connected transportation system emerging, a multidisciplinary approach will be required to develop a cyber-resilient network. Ensuring protection against cyberattacks and developing a system that can handle their consequences is a key objective moving forward. The first step to developing this system is understanding how different cyberattacks can negatively impact the operations of the transportation system. This research aimed to quantify the safety and efficiency impacts of an attack on the transportation network. To do so, a simulation was developed using Veins software to model a network of intelligent intersections in an urban environment. Vehicles communicated with Road-Side Units (RSUs) to make intersection reservations – effectively simulating CAV vehicle network. Denial of Service (DoS) and Man in the Middle (MITM) attacks were simulated by dropping and delaying vehicle's intersection reservation requests, respectively. Attacks were modeled with varying degrees of severity by changing the number of infected RSUs in the system and their attack success rates. Data analysis showed that severe attacks, either from a DoS or MITM attack, can have significant impact on the transportation network's operations. The worst-case scenario for each introduced an over 20% increase in delay per vehicle. The simulation showed also that increasing the number of compromised RSUs directly related to decreased safety and operational efficiency. Successful attacks also produced a high level of variance in their impact. One other key finding was that a single compromised RSU had very limited impact on the transportation network. These findings highlight the importance of developing security and resilience in a connected vehicle environment. Building a network that can respond to an initial attack and prevent an attack's dissemination through the network is crucial in limiting the negative effects of the attack. If proper resilience planning is not implemented for the next generation of transportation, adversaries could cause great harm to safety and efficiency with relative ease. The next generation of vehicular transportation must be able to withstand cyberattacks to function. Understanding their impact is a key first step for engineers and planners on the long road to ensuring a secure transportation network. / Master of Science / The landscape of transportation is quickly shifting as transportation technologies continue to increase in intelligence and complexity. The transportation industry is shifting its focus to Connected and Autonomous Vehicles (CAVs). The move to more autonomous and intelligent transportation systems brings with it a promise of increased transportation equity, efficiency, and safety. However, one aspect that is often overlooked in this shift is cybersecurity. As intelligent systems and vehicles have been introduced, a large amount of research has been conducted showing cyber vulnerabilities in them. With a new connected transportation system emerging, a multidisciplinary approach will be required to prevent and handle attacks. Ensuring protection against cyberattacks is a key objective moving forward. The first step to developing this system is understanding how different cyberattacks can negatively impact the operations of the transportation system. This research aimed to measure the safety and efficiency impacts of an attack on the transportation network. To do so, a simulation was developed to model an intelligent urban road network. Vehicles made reservations at each intersection they passed – effectively simulating an autonomous vehicle network. Denial of Service (DoS) and Man in the Middle (MITM) attacks were simulated by dropping, and delaying vehicle's intersection reservation requests, respectively. These cyberattacks were modeled with varying degrees of severity to test the different impacts on the transportation network. Analysis showed that severe attacks can have significant impact on the transportation network's operations. The worst-case scenario for each attack introduced an over 20% increase in delay per vehicle. The simulation showed also that increasing the number of attacked intersections directly related to decreased safety and operational efficiency. Successful attacks also produced a high level of variance in their impact. One other key finding was that a single compromised RSU had very limited impact on the transportation network. These findings highlight the importance of developing security and resilience in a connected vehicle environment. Building a transportation network that can respond to an initial attack and prevent it from impacting the entire network is crucial in limiting the negative effects of the attack. If proper resilience planning is not implemented for CAVs, hackers could cause great harm to safety and efficiency with relative ease. The next generation of vehicular transportation must be able to withstand cyberattacks to function. Understanding their impact is a key first step for engineers and planners on the long road to ensuring a secure transportation network.

One key to more effective cooperative interaction in a multi-robot team is the ability to understand the behavior and intent of other robots. Observed teammate action sequences can be learned to perform trajectory recognition which can be used to determine their current task. Previously, we have applied behavior histograms, hidden Markov models (HMMs), and conditional random fields (CRFs) to perform trajectory recognition as an approach to task monitoring in the absence of commu- nication. To demonstrate trajectory recognition of various autonomous vehicles, we used trajectory-based techniques for model generation and trajectory discrimination in experiments using actual data. In addition to recognition of trajectories, we in- troduced strategies, based on the honeybee’s waggle dance, in which cooperating autonomous teammates could leverage recognition during periods of communication loss. While the recognition methods were able to discriminate between the standard trajectories performed in a typical survey mission, there were inaccuracies and delays in identifying new trajectories after a transition had occurred. Inaccuracies in recog- nition lead to inefficiencies as cooperating teammates acted on incorrect data. We then introduce the Trajectory Adaptation for Recognition (TAR) framework which seeks to directly address difficulties in recognizing the trajectories of autonomous vehicles by modifying the trajectories they follow to perform them. Optimization techniques are used to modify the trajectories to increase the accuracy of recognition while also improving task objectives and maintaining vehicle dynamics. Experiments are performed which demonstrate that using trajectories optimized in this manner lead to improved recognition accuracy.

The present thesis explores how value can be created by fully automated vehicles as a transformational innovation. To do this, the value Framework by den Ouden (2012) is used to examine the user to research value creation from a sociological and psychological perspective. Different groups of consumers are interviewed, current premium vehicle drivers, current members of car sharing and CNDs, as well as experts. The study was carried out in Germany and Sweden, including one expert from the United States. Resulting from the study, autonomous vehicles are expected to create value from different perspectives and for different consumer groups, for example by allowing the driver to spend time on something else while travelling with the car. Furthermore, risks of automated vehicles in a value perspective are examined, such as safety issues and increased vehicle miles travelled. Concluding, this research adds up to den Ouden’s (2012) framework by adding the interconnectedness of different value perspective and applying it on the example of automated vehicles.

Liability laws currently in use rely on a fault-based system that focuses on a causal connection between driver actions and the resulting road accident. The role of the driver is set to reduce with the emergence of autonomous vehicles, so how will liability adapt to meet the needs of an autonomous world? The paper discusses possible frameworks of liability that could be implemented in the future, and accentuates the importance of the causal aspects of the current framework in the new system.

<div>In recent years, there has been considerable growth in the adoption and technology development of electric vehicles (EV), autonomous vehicles (AV), and ride sharing (RS). These technologies have the potential to improve transportation sustainability. Many studies have evaluated the environmental impacts of these technologies but the existing literature has three major gaps: (1) the adoption of these three technologies need to be evaluated considering their impact on each other, (2) many existing models do not evaluate systems on a common ground, and (3) the heterogeneous preferences of riders towards these emerging technologies are not fully incorporated. To address these gaps, this work studies and quantifies the environmental and efficiency gains that can be gained through these emerging transportation technologies by developing a Parameterized Preference-based Shared Autonomous Electric Vehicle (PP-SAEV) agent-based model. The model is then applied to a case study of New York City (NYC) taxis to evaluate the system performance with increasing AV, EV, and RS adoption.</div><div><br></div><div>The outputs from the PP-SAEV model show that replacing taxi cabs in NYC with AVs along with RS potentially can reduce CO\textsubscript{2} emissions by 866 metric Tones per day and increase average vehicle occupancy from 1.2 to 3 persons in vehicles with passenger seating capacity of 4. A prediction model based on the PP-SAEV output recommends that 6000 vehicles are needed to maintain the current level of service with 100\% AV and RS adoption using capacity 4 taxis. Taxi fleets with capacity 4 with high RS and low AV adoption are also found to have the least CO\textsubscript{2} emissions. Because the heterogeneous sharing preferences of riders have shown as the major limiting factor to ride sharing, these heterogeneous sharing preferences are further modelled. The results show that high service levels are achieved when all the riders are open to sharing, and the maximum service level is reached when 30\% of riders will only accept shared rides and 70\% of the riders are either indifferent to sharing or prefer to use ride sharing over riding alone. Additionally, the service level and waiting time of riders that are inflexible (will accept only shared or non-shared rides) are greatly impacted by varying mix of riders with different sharing preference. Finally, an optimization model was built to site charging stations in a system with continually increasing EV adoption. Using the best charging station locations, transforming a fleet of autonomous or traditional vehicles to electric vehicles does not significantly change the system service level. The results show that increasing the EV adoption in fleets with 100\% RS and AV adoption reduced the daily CO\textsubscript{2} emissions by about 861 Tones and transforming a fleet of traditional taxi cabs to electric taxi cabs reduced the daily CO\textsubscript{2} emissions by 1100 Tones.</div><div><br></div><div>In summary, this dissertation evaluates the potential growth of autonomous vehicles, ride sharing, and electric vehicles in systems where riders may have heterogeneous sharing preferences, from a system performance`s perspective and assesses the environmental impacts. The developed model and the insights gained from this study can inform policy makers to develop sustainable transportation systems incorporating the emerging transportation technologies.</div>