Autonomous vehicles offer means to complete unsafe military operations without endangering the lives of soldiers. Such solutions have fueled many efforts towards designing autonomous, or unmanned, systems. Military and academic research efforts alike continue to focus on developing these systems. While many different autonomous vehicles have been introduced, however, such complex systems have limited drive-by-wire operability. The complete process to up-fit a vehicle to fully autonomous operation involves the design, up-fit, testing and verification of many different subsystems. The objective of this thesis is to design and model an autonomous steering system requiring little modifications to an existing steering system. It is desirable to still operate the vehicle manually as well as preserve the vehicle's visual appearance. Up-fit and implementation of the designed steering system and verification of its functionality has been documented as well. Utilization of the supplied controller and software has enabled the testing and characterization of the system. The proposed design offers a solution to a wide variety of wheeled vehicles steered via the traditional and common steering wheel method. In addition, modifications have been made to an existing simulation of an unmanned vehicle in a military testbed environment (Fort Benning). The simulation accounts for the control methodology as it has been designed and tested with, which offers the ability to analyze the dynamics of the unmanned system. / Master of Science

Autonomous vehicles (AVs) are a class of safety-critical systems that are capable of decision-making and operate with little or no human intervention. For such complex systems designed to function in diverse operational domains such as rain, snow, freeway, urban roads, etc., system safety is paramount. Management of the system's safety throughout its life-cycle, from the conceptualization stage to the end of the lifecycle, is of primary importance. We describe a revision of functional safety standard ISO 26262 to support autonomous vehicles and the underlying electronic/electrical control architecture. There is a need to modify the Automotive Safety Integrity Levels (ASILs) defined in the ISO 26262 as "Controllability", a factor in determining an ASIL, is no longer applicable; the driver is no longer in a position to control the vehicle. The vehicle has taken over the responsibility of evaluating the environment and determines its next course of action to complete its current mission. These decisions have a tremendous impact on the overall safety of the system during a hazardous event and can be the difference between a successful journey and a traffic incident. To better enable the designers of such systems, we introduce a new method to assess the functional safety and derive safety goals, which are the top level safety requirement. We present a new metric-Risk Mitigation Factor to assess the decision making capability of the vehicle and to replace controllability in the ASIL definition. The case study presented highlights the advantages of using the introduced metric in defining safety goals for the autonomous vehicle. / Master of Science / Autonomous vehicles (AVs) are changing the way we perceive mobility and transportation. AVs are soon to be a part of everyday life, from giving you a ride to the office to taking children to the dentist. All the possible benefits of AVs are attainable if the systems designed are safe for use. Safety in AVs is the primary challenge in design and development. It is crucial to incorporate the principles of safety in system design from the beginning of the inception phase to the end of the lifecycle of the vehicle. The challenges for ensuring safety in AVs are enormous, from implementing the correct operation for a system to assuring that system behavior is safe in the presence of a malfunction; the scale and complexity of the systems drive the safety requirements. In the work presented, we focus on the functional safety of the underlying electrical/ electronic architecture of the vehicle, describing a revision of the automotive functional safety standard ISO 26262 for AV development. We propose to leverage the decision-making capabilities of the vehicle to assure safety in a hazardous situation.

abstract: Autonomous Vehicles (AVs), or self-driving cars, are poised to have an enormous impact on the automotive industry and road transportation. While advances have been made towards the development of safe, competent autonomous vehicles, there has been inadequate attention to the control of autonomous vehicles in unanticipated situations, such as imminent crashes. Even if autonomous vehicles follow all safety measures, accidents are inevitable, and humans must trust autonomous vehicles to respond appropriately in such scenarios. It is not plausible to program autonomous vehicles with a set of rules to tackle every possible crash scenario. Instead, a possible approach is to align their decision-making capabilities with the moral priorities, values, and social motivations of trustworthy human drivers.Toward this end, this thesis contributes a simulation framework for collecting, analyzing, and replicating human driving behaviors in a variety of scenarios, including imminent crashes. Four driving scenarios in an urban traffic environment were designed in the CARLA driving simulator platform, in which simulated cars can either drive autonomously or be driven by a user via a steering wheel and pedals. These included three unavoidable crash scenarios, representing classic trolley-problem ethical dilemmas, and a scenario in which a car must be driven through a school zone, in order to examine driver prioritization of reaching a destination versus ensuring safety. Sample human driving data in CARLA was logged from the simulated car’s sensors, including the LiDAR, IMU and camera. In order to reproduce human driving behaviors in a simulated vehicle, it is necessary for the AV to be able to identify objects in the environment and evaluate the volume of their bounding boxes for prediction and planning. An object detection method was used that processes LiDAR point cloud data using the PointNet neural network architecture, analyzes RGB images via transfer learning using the Xception convolutional neural network architecture, and fuses the outputs of these two networks. This method was trained and tested on both the KITTI Vision Benchmark Suite dataset and a virtual dataset exclusively generated from CARLA. When applied to the KITTI dataset, the object detection method achieved an average classification accuracy of 96.72% and an average Intersection over Union (IoU) of 0.72, where the IoU metric compares predicted bounding boxes to those used for training. / Dissertation/Thesis / Masters Thesis Mechanical Engineering 2020

Autonomous vehicles (AVs) are a disruptive technology that is expected to vastly change the current transportation system. AV potential benefits in terms of safety, mobility, efficiency and other impacts types have been documented in the literature. AVs are expected to increase travel demand due to the enhanced ease of making trips and provision of mobility to people currently with travel-limiting disabilities. The potential increase in travel demand, with its attendant congestion, may probably be offset by the transportation network capacity increase due to the reduced operational headways between AVs. However, such capacity benefits can be fully realized only when AVs fully saturate the market, because operating at low headways may be unsafe for Human Driven Vehicles (HDVs). Thus, to promote AV ownership while capturing the capacity benefits of an AV-only traffic stream, the conversion of traditional lanes to AV-exclusive use is prescribed often. In the AV-exclusive lanes, the vehicles can operate at reduced headways and at higher speeds, sharply increasing throughput. However, the metric used frequently by researchers for AV-exclusive lane evaluation is the total system travel time. AV-exclusive lanes may appear to be beneficial in terms of total system travel time but may come at a cost of environmental protection and social equity, the other two elements of sustainable development. Appropriating HDV lanes for AV-exclusive use will cause congestion on HDV lanes thereby increasing their emissions. Further, the AVs benefits may be accompanied by increased cost of HDV travel, which raises questions about equity. This thesis therefore presents a sustainable AV-exclusive lane deployment strategy by formulating and solving a multicriteria bi-level optimization problem with equity-related constraints. Mathematically, the problem is described as a discrete network design problem. Recognizing the difficulty of solving this NIP hard problem, the thesis combines the active set method with heuristic conditionalities to improve computational efficiency. The thesis’s framework can be used by agencies for evaluation and decision support regarding AV-exclusive lane deployment in a manner that fosters long-term sustainability.

No description available.

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

<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>

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.

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.