Siga este enlace para ver otros tipos de publicaciones sobre el tema: Autonomous vehicle testing.
Tesis sobre el tema "Autonomous vehicle testing"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 35 mejores tesis para su investigación sobre el tema "Autonomous vehicle testing".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
1
Mikesell, David Russell. "Portable automated driver for universal road vehicle dynamics testing". Columbus, Ohio : Ohio State University, 2008. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1198722243.
Texto completo
2
Kirsch, Patricia Jean. "Autonomous swarms of unmanned vehicles software control system and ground vehicle testing /". College Park, Md. : University of Maryland, 2005. http://hdl.handle.net/1903/2993.
Texto completoResumen
Thesis (M.S.) -- University of Maryland, College Park, 2005.Thesis research directed by: Dept. of Electrical and Computer Engineering. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also available in paper.
3
Hebib, Jasmina y Sofie Dam. "Vehicle Dynamic Models for Virtual Testing of Autonomous Trucks". Thesis, Linköpings universitet, Fordonssystem, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-155513.
Texto completoResumen
The simulator in a testing environment for trucks is dependent on accurate vehicle dynamic models. There are multiple models at Volvo, all developed to support the objectives of individual research. A selection of four, named Single Track model (STM), Global Simulation Platform (GSP), One-Track Model with linear slip (OTM) and Volvo Transport Model (VTM), are evaluated to examine the usage of them. Four different scenarios are therefore generated to emulate common situations in traffic. Depending on the results, the models and their corresponding limitsforusagearedescribed. Theevaluationismadebycomparingallmodelsto the best model for each scenario by measuring the normalized error distribution. It is shown that at certain thresholds, other models can perform close enough to the best model. In the end of the report, future improvements for the evaluated models and external models are suggested.
4
Nordenström, Martin. "Future certification of autonomous vehicles and the use of virtual testing methods". Thesis, KTH, Fordonsdynamik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-288717.
Texto completoResumen
One of the biggest obstacles to launching autonomous vehicles is the current legislation, which currently does not cover automation level higher than level 2. Work on developing the legal requirements takes place at UN level within WP29 (The UNECE World Forum for Harmonization of Vehicle Regulations). As a world-leading vehicle manufacturer, Scania is aspiring to pave the way for sustainable transport solutions. At Scania, well-established methodologies on certification of different systems exist, although the process of certification of autonomous driving systems needs to be developed.This master thesis investigates the current situation regarding the elaboration of regulations to cover autonomous vehicles, future certification methods related to these systems, and how this applies to Scania. Particular focus is being on the investigation of virtual certification methods. This can form the basis for various departments at Scania in their work with future autonomous systems and how to get these certified.The future certification work for autonomous vehicles will be based on a validation process based on a process called the ‘Multi-pillars approach’ / ‘Three-pillars approach’. The idea is that the autonomous vehicle should be certified based on a process where the basis for certification is made by validating and justifying its systems. This will be done through simulations and other methods to ensure that the systems are satisfactory. A less extensive work should then be done in the testing of the autonomous vehicle on the test track and in traffic, where only less demanding situations must be validated.The functional requirements of the autonomous vehicle will largely control the validation process that is carried out for the ‘Multi-pillars approach’ / ‘Three-pillars approach’. For example, the definition of ODD (Operational Design Domain) is crucial for the validation that the vehicle will undergo at a later stage.Ett av de största hindren för att lansera självkörande fordon är den nuvarande lagstiftningen som i dagsläget inte täcker automationsnivå högre än nivå 2. Arbetet med att ta fram lagkraven sker på FN nivå inom WP29 (The UNECE World Forum for Harmonization of Vehicle Regulations). Som en världsledande fordonstillverkare strävar Scania efter att bana väg för hållbara transportlösningar. På Scania finns väletablerade metoder för certifiering av olika system, men processen för certifiering av autonoma fordon måste dock utvecklas.Detta examensarbete undersöker den aktuella situationen när det gäller utformandet av regelverk för att täcka autonoma fordon, framtida certifieringsmetoder relaterade till dessa system och hur detta påverkar Scania. Särskilt fokus ligger på utredning av virtuella certifieringsmetoder. Detta kan ligga till grund för olika avdelningar på Scania i deras arbete med framtida autonoma system och hur man får dessa certifierade.Det framtida certifieringsarbetet för autonoma fordon kommer att bygga på en valideringsprocess som bygger på en process som kallas för ”Multi-pillars approach”/”Three- pillars approach”. Tanken är att fordonet ska certifieras utifrån en process där grunden till certifiering görs genom att validera och rättfärdiga sina system. Detta ska ske genom simulering och andra metoder för att säkerhetsställa att systemen är tillfredställande. Ett mindre omfattande arbete ska sedan göras i testningen av fordonen på testbana och ute i trafik, där endast mindre krävande situationer ska valideras.De funktionella kraven på fordonen kommer till stor del att styra den valideringsprocessen som görs med för ”Multi-pillars approach”/”Three-pillars approach”. Exempelvis är definierandet av ODD (Operational Design Domain) avgörande för den validering som fordonet i ett senare skede ska genomgå.
5
Volland, Kirk N. "Design, construction and testing of a prototype holonomic autonomous vehicle". Thesis, Monterey, Calif. : Naval Postgraduate School, 2007. http://bosun.nps.edu/uhtbin/hyperion-image.exe/07Dec%5FVolland.pdf.
Texto completoResumen
Thesis (M.S. in Applied Physics)--Naval Postgraduate School, December 2007.Thesis Advisor(s): Harkins, Richard. "December 2007." Description based on title screen as viewed on January 24, 2008. Includes bibliographical references (p. 189-192). Also available in print.
6
Arslan, Suat. "Testing and evaluation of the Small Autonomous Underwater Vehicle Navigation System (SANS)". Thesis, Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2000. http://handle.dtic.mil/100.2/ADA376607.
Texto completoResumen
Thesis (M.S. in Electrical Engineering) Naval Postgraduate School, March 2000.Thesis advisor(s): Yun, Xiaoping; Bachmann, Eric R. "March 2000." Includes bibliographical references (p. 93-94). Also available online.
7
Jun, Hyun Il. "The implementation and testing of a robotic arm on an autonomous vehicle". Thesis, Monterey, Calif. : Naval Postgraduate School, 2007. http://bosun.nps.edu/uhtbin/hyperion-image.exe/07Dec%5FJun.pdf.
Texto completoResumen
Thesis (M.S. in Applied Physics)--Naval Postgraduate School, December 2007.Thesis Advisor(s): Harkins, Richard. "December 2007." Description based on title screen as viewed on January 18, 2008. Includes bibliographical references (p. 35-36). Also available in print.
8
Doepke, Edward Brady. "DESIGN AND FLIGHT TESTING OF A WARPING WING FOR AUTONOMOUS FLIGHT CONTROL". UKnowledge, 2012. http://uknowledge.uky.edu/me_etds/20.
Texto completoResumen
Inflatable-wing Unmanned Aerial Vehicles (UAVs) have the ability to be packed in a fraction of their deployed volume. This makes them ideal for many deployable UAV designs, but inflatable wings can be flexible and don’t have conventional control surfaces. This thesis will investigate the use of wing warping as a means of autonomous control for inflatable wings. Due to complexities associated with manufacturing inflatable structures a new method of rapid prototyping deformable wings is used in place of inflatables to decrease cost and design-cycle time. A UAV testbed was developed and integrated with the warping wings and flown in a series of flight tests. The warping wing flew both under manual control and autopilot stabilization.
9
Mercer, Anthony Scott. "Autonomous unmanned ground vehicle for non-destructive testing of fiber reinforced polymer bridge decks". Morgantown, W. Va. : [West Virginia University Libraries], 2006. https://eidr.wvu.edu/etd/documentdata.eTD?documentid=4943.
Texto completoResumen
Thesis (M.S.)--West Virginia University, 2006.Title from document title page. Document formatted into pages; contains x, 100 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 83-86).
10
Sevcik, Keith Wayne Oh Paul Yu. "A hardware-in-the-loop testing facility for unmanned aerial vehicle sensor suites and control algorithms /". Philadelphia, Pa. : Drexel University, 2010. http://hdl.handle.net/1860/3262.
Texto completo
11
Davis, Duane T. "Design, implementation and testing of a common data model supporting autonomous vehicle compatibility and interoperability". Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2006. http://library.nps.navy.mil/uhtbin/hyperion/06Sep%5FDavis%5FPhD.pdf.
Texto completoResumen
Dissertation (PhD. in Computer Science)--Naval Postgraduate School, September 2006.Dissertation Advisor(s): Don Brutzman. "September 2006." Includes bibliographical references (p. 317-328). Also available in print.
12
Sidhu, Anmol. "Development of an Autonomous Test Driver and Strategies for Vehicle Dynamics Testing and Lateral Motion Control". The Ohio State University, 2010. http://rave.ohiolink.edu/etdc/view?acc_num=osu1269352974.
Texto completo
13
Argui, Imane. "A vision-based mixed-reality framework for testing autonomous driving systems". Electronic Thesis or Diss., Normandie, 2024. http://www.theses.fr/2024NORMIR37.
Texto completoResumen
Cette thèse explore le développement et la validation des systèmes de navigation autonome dans un environnement de réalité mixte (RM), avec pour objectif de combler l’écart entre la simulation virtuelle et les tests en conditions réelles. Les travaux mettent l’accent sur le potentiel des environnements en réalité mixte pour tester les systèmes autonomes de manière sûre, efficace et économique. La thèse est structurée en plusieurs parties, et commence par une revue des technologies de pointe dans la navigation autonome et les applications en réalité mixte. En utilisant des modèles à base de règles et des modèles d’apprentissage, des expérimentations visent à évaluer les performances des robots autonomes dans des environnements simulés, réels et de RM. Un des objectifs principaux est de réduire le « reality gap »—c’est-à-dire la différence entre les comportements observés en simulation et ceux observés dans des applications réelles—en intégrant des éléments réels avec des composants virtuels dans des environnements de RM. Cette approche permet des tests et une validation plus proche des contraintes réelles sans les risques associés aux essais physiques. Une partie importante du travail est consacrée à la mise en œuvre et au test d’une stratégie d’augmentation hors ligne visant à améliorer les capacités de perception des systèmes autonomes à l’aide des informations de profondeur. De plus, l’apprentissage par renforcement est appliqué pour évaluer son potentiel dans les environnements de RM. La thèse démontre que ces modèles peuvent apprendre efficacement à naviguer et à éviter les obstacles dans des simulations virtuelles et obtenir des résultats similaires lorsqu’ils sont transférés dans des environnements de RM, soulignant la flexibilité du cadre pour différents modèles de systèmes autonomes. À travers ces expériences, la thèse montre le potentiel des environnements de réalité mixte comme une plateforme polyvalente et robuste pour faciliter le développement des technologies de navigation autonome, offrant une approche plus sûre et plus évolutive pour la validation des modèles avant leur déploiement dans le monde réelThis thesis explores the development and validation of autonomous navigation systems within a mixed-reality (MR) framework, aiming to bridge the gap between virtual simulation and real-world testing. The research emphasizes the potential of MR environments for safely, efficiently, and cost-effectively testing autonomous systems. The thesis is structured around several chapters, beginning with a review of state-of-the-art technologies in autonomous navigation and mixed-reality applications. Through both rule-based and learning-based models, the research investigates the performance of autonomous robots within simulated, real, and MR environments. One of the core objectives is to reduce the "reality gap"—the discrepancy between behaviors observed in simulations versus real-world applications—by integrating real- world elements with virtual components in MR environments. This approach allows for more accurate testing and validation of algorithms without the risks associated with physical trials. A significant part of the work is dedicated to implementing and testing an offline augmentation strategy aimed at enhancing the perception capabilities of autonomous systems using depth information. Furthermore, reinforcement learning (RL) is applied to evaluate its potential within MR environments. The thesis demonstrates that RL models can effectively learn to navigate and avoid obstacles in virtual simulations and perform similarly well when transferred to MR environments, highlighting the framework’s flexibility for different autonomous system models. Through these experiments, the thesis establishes MR environments as a versatile and robust platform for advancing autonomous navigation technologies, offering a safer, more scalable approach to model validation before real-world deployment
14
Patil, Mayur. "Test Scenario Development Process and Software-in-the-Loop Testing for Automated Driving Systems". The Ohio State University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=osu1574794282029419.
Texto completo
15
Alberi, Thomas James. "A Proposed Standardized Testing Procedure for Autonomous Ground Vehicles". Thesis, Virginia Tech, 2008. http://hdl.handle.net/10919/32321.
Texto completoResumen
Development of unmanned vehicles will increase as the need to save lives rises. In both military and civilian applications, humans can be taken out of the loop through the implementation of safe and intelligent autonomous vehicles. Although hardware and software development continue to play a large role in the autonomous vehicle industry, validation of these systems will always be necessary. The ability to test these vehicles thoroughly and efficiently will ensure their proper and flawless operation.
On November 3, 2007 the Defense Advanced Research Projects Agency held the Urban Challenge to drive the development of autonomous ground vehicles for military use. This event required vehicles built by teams across the world to autonomously navigate a 60 mile course in an urban environment in less than 6 hours. This thesis addresses the testing aspect of autonomous ground vehicles that exhibit the advanced behaviors necessary for operating in such an event. Specifically, the experiences of Team Victor Tango and other Urban Challenge teams are covered in detail. Testing facilities, safety measures, procedures, and validation methods utilized by these teams provide valuable information on the development of their vehicles. Combining all these aspects results in a proposed testing strategy for autonomous ground vehicles.Master of Science
16
SINI, JACOPO. "Novel Validation Techniques for Autonomous Vehicles". Doctoral thesis, Politecnico di Torino, 2021. http://hdl.handle.net/11583/2950482.
Texto completo
17
Kalidas, Karthik. "Simulation Framework for Testing Autonomous Vehicles in a School for the Blind Campus". The Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu160613302915035.
Texto completo
18
Gangadharan, Athul. "An Evaluation of Automatic Test Case Generation strategy from Requirements for Electric/Autonomous Vehicles". Thesis, Uppsala universitet, Institutionen för informatik och media, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-419583.
Texto completoResumen
Software testing is becoming more prominent within the automotive industry due to more complex systems, and functions are implemented in the vehicles. The vehicles in the future will have the functionality to manage different levels of automation, which also means that vehicles driven by humans will have more supportive functionality to increase safety and avoid accidents. These functionalities result in a massive growth in the number of test scenarios to indicate that the vehicles are safe, and this makes it impossible to continue performing the tests in the same way as it has been done until today. The new conditions require that the test scenarios and Test Cases both be generated and executed automatically. In this thesis, an investigation and evaluation are performed to analyze the Automatic Test Case Generation methods available for inputs from Natural Language Requirements in an automotive industrial context at NEVS AB. This study aims to evaluate the NAT2TEST strategy by replacing the manual method and obtain a similar or better result. A comparative analysis is performed between the manual and automated approaches for various levels of requirements. The results show that utilizing this strategy in an industrial scenario can improve efficiency if the requirements to be tested are for well-documented lower-level requirements.
19
Narasimhan, Ramakrishnan Akshra. "Design and Evaluation of Perception System Algorithms for Semi-Autonomous Vehicles". The Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu1595256912692618.
Texto completo
20
Puttige, Vishwas Ramadas Engineering & Information Technology Australian Defence Force Academy UNSW. "Neural network based adaptive control for autonomous flight of fixed wing unmanned aerial vehicles". Awarded by:University of New South Wales - Australian Defence Force Academy. Engineering & Information Technology, 2009. http://handle.unsw.edu.au/1959.4/43736.
Texto completoResumen
This thesis presents the development of small, inexpensive unmanned aerial vehicles (UAVs) to achieve autonomous fight. Fixed wing hobby model planes are modified and instrumented to form experimental platforms. Different sensors employed to collect the flight data are discussed along with their calibrations. The time constant and delay for the servo-actuators for the platform are estimated. Two different data collection and processing units based on micro-controller and PC104 architectures are developed and discussed. These units are also used to program the identification and control algorithms. Flight control of fixed wing UAVs is a challenging task due to the coupled, time-varying, nonlinear dynamic behaviour. One of the possible alternatives for the flight control system is to use the intelligent adaptive control techniques that provide online learning capability to cope with varying dynamics and disturbances. Neural network based indirect adaptive control strategy is applied for the current work. The two main components of the adaptive control technique are the identification block and the control block. Identification provides a mathematical model for the controller to adapt to varying dynamics. Neural network based identification provides a black-box identification technique wherein a suitable network provides prediction capability based upon the past inputs and outputs. Auto-regressive neural networks are employed for this to ensure good retention capabilities for the model that uses the past outputs and inputs along with the present inputs. Online and offline identification of UAV platforms are discussed based upon the flight data. Suitable modifications to the Levenberg-Marquardt training algorithm for online training are proposed. The effect of varying the different network parameters on the performance of the network are numerically tested out. A new performance index is proposed that is shown to improve the accuracy of prediction and also reduces the training time for these networks. The identification algorithms are validated both numerically and flight tested. A hardware-in-loop simulation system has been developed to test the identification and control algorithms before flight testing to identify the problems in real time implementation on the UAVs. This is developed to keep the validation process simple and a graphical user interface is provided to visualise the UAV flight during simulations. A dual neural network controller is proposed as the adaptive controller based upon the identification models. This has two neural networks collated together. One of the neural networks is trained online to adapt to changes in the dynamics. Two feedback loops are provided as part of the overall structure that is seen to improve the accuracy. Proofs for stability analysis in the form of convergence of the identifier and controller networks based on Lyapunov's technique are presented. In this analysis suitable bounds on the rate of learning for the networks are imposed. Numerical results are presented to validate the adaptive controller for single-input single-output as well as multi-input multi-output subsystems of the UAV. Real time validation results and various flight test results confirm the feasibility of the proposed adaptive technique as a reliable tool to achieve autonomous flight. The comparison of the proposed technique with a baseline gain scheduled controller both in numerical simulations as well as test flights bring out the salient adaptive feature of the proposed technique to the time-varying, nonlinear dynamics of the UAV platforms under different flying conditions.
21
Neves, Vânia de Oliveira. "Automatização do teste estrutural de software de veículos autônomos para apoio ao teste de campo". Universidade de São Paulo, 2015. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-15092015-090805/.
Texto completoResumen
Veículo autônomo inteligente (ou apenas veículo autônomo VA) é um tipo de sistema embarcado que integra componentes físicos (hardware) e computacionais (software). Sua principal característica é a capacidade de locomoção e de operação de modo semi ou completamente autônomo. A autonomia cresce com a capacidade de percepção e de deslocamento no ambiente, robustez e capacidade de resolver e executar tarefas lidando com as mais diversas situações (inteligência). Veículos autônomos representam um tópico de pesquisa importante e que tem impacto direto na sociedade. No entanto, à medida que esse campo avança alguns problemas secundários aparecem como, por exemplo, como saber se esses sistemas foram suficientemente testados. Uma das fases do teste de um VA é o teste de campo, em que o veículo é levado para um ambiente pouco controlado e deve executar livremente a missão para a qual foi programado. Ele é geralmente utilizado para garantir que os veículos autônomos mostrem o comportamento desejado, mas nenhuma informação sobre a estrutura do código é utilizada. Pode ocorrer que o veículo (hardware e software) passou no teste de campo, mas trechos importantes do código nunca tenham sido executados. Durante o teste de campo, os dados de entrada são coletados em logs que podem ser posteriormente analisados para avaliar os resultados do teste e para realizar outros tipos de teste offline. Esta tese apresenta um conjunto de propostas para apoiar a análise do teste de campo do ponto de vista do teste estrutural. A abordagem é composta por um modelo de classes no contexto do teste de campo, uma ferramenta que implementa esse modelo e um algoritmo genético para geração de dados de teste. Apresenta também heurísticas para reduzir o conjunto de dados contidos em um log sem diminuir substancialmente a cobertura obtida e estratégias de combinação e mutação que são usadas no algoritmo. Estudos de caso foram conduzidos para avaliar as heurísticas e estratégias e são também apresentados e discutidos.Intelligent autonomous vehicle (or just autonomous vehicle - AV) is a type of embedded system that integrates physical (hardware) and computational (software) components. Its main feature is the ability to move and operate partially or fully autonomously. Autonomy grows with the ability to perceive and move within the environment, robustness and ability to solve and perform tasks dealing with different situations (intelligence). Autonomous vehicles represent an important research topic that has a direct impact on society. However, as this field progresses some secondary problems arise, such as how to know if these systems have been sufficiently tested. One of the testing phases of an AV is the field testing, where the vehicle is taken to a controlled environment and it should execute the mission for which it was programed freely. It is generally used to ensure that autonomous vehicles show the intended behavior, but it usually does not take into consideration the code structure. The vehicle (hardware and software) could pass the field testing, but important parts of the code may never have been executed. During the field testing, the input data are collected in logs that can be further analyzed to evaluate the test results and to perform other types of offline tests. This thesis presents a set of proposals to support the analysis of field testing from the point of view of the structural testing. The approach is composed of a class model in the context of the field testing, a tool that implements this model and a genetic algorithm to generate test data. It also shows heuristics to reduce the data set contained in a log without reducing substantially the coverage obtained and combination and mutation strategies that are used in the algorithm. Case studies have been conducted to evaluate the heuristics and strategies, and are also presented and discussed.
22
Weber, Michael. "Development of a method for practical testing of camera-based advanced driver assistance systems in automotive vehicles using augmented reality". Electronic Thesis or Diss., Bourgogne Franche-Comté, 2023. http://www.theses.fr/2023UBFCA027.
Texto completoResumen
Les systèmes avancés d'aide à la conduite (ADAS) assistent le conducteur, lui offrent du confort et prennent la responsabilité d'accroître la sécurité routière. Ces systèmes complexes font l'objet d'une phase d'essai approfondie qui permet d'optimiser la qualité, la reproductibilité et les coûts. Les systèmes d'aide à la conduite de demain prendront en charge des proportions toujours plus grandes de situations de conduite dans des scénarios de plus en plus complexes et représentent un facteur clé pour la conduite autonome. Les méthodes d'essai actuelles pour les systèmes d'aide à la conduite peuvent être divisées en deux catégories : la simulation et la réalité. Le concept de base de la simulation est de bénéficier de la reproductibilité, de la flexibilité et de la réduction des coûts. Cependant, la simulation ne peut pas encore remplacer complètement les essais en conditions réelles. Les conditions physiques, telles que les conditions météorologiques, le revêtement de la route et d'autres variables, jouent un rôle crucial dans l'évaluation des essais routiers des systèmes d'aide à la conduite et ne peuvent pas être entièrement reproduites dans un environnement virtuel. Ces méthodes d'essai reposent sur des tests de conduite réels sur des sites d'essai spéciaux ainsi que dans un trafic routier réel, ce qui prend beaucoup de temps et est très coûteux. Des méthodes d'essai nouvelles et efficaces sont donc nécessaires pour ouvrir la voie aux futurs systèmes d'assistance à la conduite automobile. Une nouvelle approche, Vehicle in the Loop (VIL), déjà utilisée dans l'industrie aujourd'hui, combine les avantages de la simulation et de la réalité. L'approche retenue dans ce projet est une nouvelle méthode qui vient s'ajouter aux solutions VIL existantes. Tirant parti de la simulation et de la réalité pour tester les ADAS, ce projet présente une nouvelle approche utilisant la réalité augmentée (AR) pour tester ADAS basés sur des caméras, de manière reproductible, rentable et rapide. Une puissance informatique élevée est nécessaire pour des conditions environnementales automobiles complexes, telles que la vitesse élevée du véhicule et le nombre réduit de points d'orientation sur une piste d'essai par rapport aux applications de réalité augmentée à l'intérieur d'un bâtiment. Un modèle tridimensionnel contenant des informations précises sur le site d'essai est généré sur la base de la combinaison de la localisation et de la cartographie visuelles simultanées (vSLAM) et de la segmentation sémantique. L'utilisation d'un processus d'augmentation spécial nous permet d'enrichir la réalité avec des usagers de la route virtuels afin de présenter une preuve de concept pour de futures procédures d'essaiAdvanced Driver Assistance Systems (ADAS) support the driver, offer comfort, and take responsibility for increasing road safety. These complex systems endure an extensive testing phase resulting in optimization potential regarding quality, reproducibility, and costs. ADAS of the future will support ever-larger proportions of driving situations in increasingly complex scenarios and represent a key factor for Autonomous Driving (AD). Current testing methods for ADAS can be divided into simulation and reality. The core concept behind the simulation is to benefit from reproducibility, flexibility, and cost reduction. However, simulation cannot yet completely replace real-world tests. Physical conditions, such as weather, road surface, and other variables, play a crucial role in evaluating ADAS road tests and cannot be fully replicated in a virtual environment. These test methods rely on real driving tests on special test sites as well as in real road traffic and are very time-consuming and costly. Therefore, new and efficient test methods are required to pave the way for future ADAS. A new approach Vehicle in the Loop (VIL), which is already being used in the industry today, combines the advantages of simulation and reality. The approach in this project is a new method besides existing VIL solutions. Taking advantage of testing ADAS in simulation and reality, this project presents a new approach to using Augmented Reality (AR) to test camera-based ADAS in a reproducible, cost- and time-efficient way. High computer power is needed for complex automotive environmental conditions, such as high vehicle speed and fewer orientation points on a test track compared to AR applications inside a building. A three-dimensional model with accurate information about the test site is generated based on the combination of visual Simultaneous Localization and Mapping (vSLAM) and Semantic Segmentation. The use of a special augmentation process allows us to enrich reality with virtual road users to present a proof of concept for future test procedures
23
Bhat, Sriharsha. "Hydrobatics: Efficient and Agile Underwater Robots". Licentiate thesis, KTH, Farkostteknik och Solidmekanik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-286062.
Texto completoResumen
The term hydrobatics refers to the agile maneuvering of underwater vehicles. Hydrobatic capabilities in autonomous underwater vehicles (AUVs) can enable increased maneuverability without a sacrifice in efficiency and speed. This means innovative robot designs and new use case scenarios are possible. Benefits and technical challenges related to hydrobatic AUVs are explored in this thesis. The dissertation contributes to new knowledge in simulation, control and field applications, and provides a structured approach to realize hydrobatic capabilities in real world impact areas. Three impact areas are considered - environmental monitoring, ocean production and security. A combination of agility in maneuvering and efficiency in performance is crucial for successful AUV applications. To achieve such performance, two technical challenges must be solved. First, these AUVs have fewer control inputs than degrees of freedom, which leads to the challenge of underactuation. The challenge is described in detail and solution strategies that use optimal control and model predictive control (MPC) are highlighted. Second, the flow around an AUV during hydrobatic maneuvers transitions from laminar to turbulent flow at high angles of attack. This renders flight dynamics modelling difficult. A full 0-360 degree envelope flight dynamics model is therefore derived, which combines a multi-fidelity hydrodynamic database with a generalized component-buildup approach. Such a model enables real-time (or near real-time) simulations of hydrobatic maneuvers including loops, helices and tight turns. Next, a cyber-physical system (CPS) is presented -- it safely transforms capabilities derived in simulation to real-world use cases in the impact areas described. The simulator environment is closely integrated with the robotic system, enabling pre-validation of controllers and software before hardware deployment. The small and hydrobatic SAM AUV (developed in-house at KTH as part of the Swedish Maritime Robotics Center) is used as a test platform. The CPS concept is validated by using the SAM AUV for the search and detection of a submerged target in field operating conditions. Current research focuses on further exploring underactuated control and motion planning. This includes development of real-time nonlinear MPC implementations running on AUV hardware, as well as intelligent control through feedback motion planning, system identification and reinforcement learning. Such strategies can enable real-time robust and adaptive control of underactuated systems. These ideas will be applied to demonstrate new capabilities in the three impact areas.Termen hydrobatik avser förmåga att utföra avancerade manövrer med undervattensfarkoster. Syftet är att, med bibehållen fart och räckvidd, utvigda den operationella förmågan i manövrering, vilket möjliggör helt nya användningsområden för maximering av kostnadseffektivitet. I denna avhandling undersöks fördelar och tekniska utmaningar relaterade till hydrobatik som tillämpas på undervattensrobotar, vanligen kallade autonoma undervattensfarkoster (AUV). Avhandlingen bidrar till ny kunskap i simulering, reglering samt tillämpning i experiment av dessa robotar genom en strukturerad metod för att realisera hydrobatisk förmåga i realistiska scenarier. Tre nyttoområden beaktas - miljöövervakning, havsproduktion och säkerhet. Inom dessa nyttoområden har ett antal scenarios identifierats där en kombination av smidighet i manövrerbarhet samt effektivitet i prestanda är avgörande för robotens förmåga att utföra sin uppgift. För att åstadkomma detta måste två viktiga tekniska utmaningar lösas. För det första har dessa AUVer färre styrytor/trustrar än frihetsgrader, vilket leder till utmaningen med underaktuering. Utmaningen beskrivs i detalj och lösningsstrategier som använder optimal kontroll och modellprediktiv kontroll belyses. För det andra är flödet runt en AUV som genomför hydrobatiska manövrar komplext med övergång från laminär till stark turbulent flöde vid höga anfallsvinklar. Detta gör flygdynamikmodellering svår. En full 0-360 graders flygdynamikmodell härleds därför, vilken kombinerar en multi-tillförlitlighets hydrodynamisk databas med en generaliserad strategi för komponentvis-superpositionering av laster. Detta möjliggör prediktering av hydrobatiska manövrar som t.ex. utförande av looping, roll, spiraler och väldigt snäva svängradier i realtids- eller nära realtids-simuleringar. I nästa steg presenteras ett cyber-fysikaliskt system (CPS) – där funktionalitet som härrör från simuleringar kan överföras till de verkliga användningsområdena på ett effektivt och säkert sätt. Simulatormiljön är nära integrerad i robot-miljön, vilket möjliggör förvalidering av reglerstrategier och mjukvara innan hårdvaruimplementering. En egenutvecklad hydrobatisk AUV (SAM) används som testplattform. CPS-konceptet valideras med hjälp av SAM i ett realistiskt sceanrio genom att utföra ett sökuppdrag av ett nedsänkt föremål under fältförhållanden. Resultaten av arbetet i denna licentiatavhandling kommer att användas i den fortsatta forskningen som fokuserar på att ytterligare undersöka och utveckla ny metodik för reglering av underaktuerade AUVer. Detta inkluderar utveckling av realtidskapabla ickelinjära MPC-implementeringar som körs ombord, samt AI-baserade reglerstrategier genom ruttplaneringsåterkoppling, autonom systemidentifiering och förstärkningsinlärning. Sådan utveckling kommer att tillämpas för att visa nya möjligheter inom de tre nyttoområdena.
SMaRC
24
RIDOLFI, ALESSANDRO. "Modelling, control and testing of underwater mobile robots". Doctoral thesis, 2014. http://hdl.handle.net/2158/854497.
Texto completo
25
Pagliai, Marco. "Design and testing of innovative thrusters and their integration in the design of a reconfigurable underwater vehicle". Doctoral thesis, 2019. http://hdl.handle.net/2158/1154277.
Texto completoResumen
Nowadays, Unmanned Underwater Vehicles (UUVs) are increasingly used in underwater operations (e.g. exploration, monitoring, maintenance), because they make safe working under the water. Since their use is growing, it is extremely important to ensure better maneuverability and lower power consumption to improve the performances of this kind of vehicles. The work carried out in the last three years at the Mechatronics and Dynamic Modelling Laboratory (MDM Lab) of the Department of Industrial Engineering of University of Florence (UNIFI DIEF), focused exactly on these problematics. More in details, the study presented below concerned both an in-depth analysis of current commercial and research UUVs, with the aim to define the specifications and the design of a new type of underwater vehicle, and the study of commercial low cost motors, with the purpose of improving their performance thus the one of the mobile robots they are mounted on.
In the introduction a short summary of the UUVs history is reported and all the vehicle of UNIFI DIEF are presented. In the first chapter, instead, a deep analysis of the state of art of underwater vehicles is reported. This study has been divided into AUVs (Autonomous Underwater Vehicles), ROVs (Remotely Operated Vehicle) and Gliders features, taking into account especially their propulsion systems. In addition, based on also the study of the state of art, the conceptual design of an innovative vehicle capable of changing its shape and its thrusters layout is synthetized. In the following chapter (chapter 2), the design of the innovative vehicle is described, in order to verify the feasibility of the vehicle itself exploiting current technology and to estimate its performances and the budget required for its realization. In chapter 3, a low cost commercial thruster is presented. With these thrusters all the vehicles of UNIFI DIEF have been equipped; for this reason, the third chapter reports a deep experimental study to define the relation between thruster parameters and the input command. This study has been very important because it allows improving the thruster parameters estimation and consequently the control performances. In addition, in this third chapter the new vehicle designed by UNIFI DIEF with the collaboration of MDM Team s.r.l. (an University of Florence Spin Off) and named ZENO is presented. In chapter 4, it is shown how the thruster studied in the previous chapter has been upgraded and hardware modifications have been made. More in details, in this chapter it is explained how a Hall effect sensor has been integrated in the thruster in order to measure the rotor speed. In addition, it is shown how a low cost closed loop control system has been implemented for the thruster previously described and the results obtained on dedicated test rig are given. Finally, in the conclusion the results obtained are summarized, highlighting how the designed innovative vehicle is feasible and has the performances required by its hypothetical application. In addition, it is summarized how the studied thrusters have improved the UNIFI DIEF vehicles performances.
26
Curtis, Timothy. "The design, construction, outfitting,and preliminary testing of the C-SCOUT autonomous underwater vehicle (AUV) /". 2001.
Buscar texto completo
27
Weng, Ming-Hung y 翁銘宏. "An analysis of the relationship between autonomous vehicle on roads testing and patent for competitors". Thesis, 2019. http://ndltd.ncl.edu.tw/handle/5bgsur.
Texto completoResumen
碩士國立臺灣科技大學
專利研究所
107
This study based on the disengagement reports by the California Department of Motor Vehicles to find the ranking of the autonomous vehicle testing on the public roads. Meanwhile, the rate of disengagement depends on the number of disengagements per thousand mileage which can judge the performance of the autonomous vehicle. According to the global patent searching system of the Intellectual Property Office of the Ministry of Economic Affairs, Taiwan, to search the keyword "Autonomous and Vehicle" for full-text search and analysis, which analyze the count of the patent and portfolio of the patent , the main inventor ,which can find out the relation between the international patent classification and the rate of the disengagement to illustrate the trend of autonomous vehicle field in California, US. This study found that the number of patents grow up with the companies’ scale. The main inventor come from similar background to autonomous vehicle field and universities. It should be considered that The inventor may choose to start-up his own business or jump-hopping to competitors’ companies. But the huge capital is big problem to startup or leave the research team by different career plan. For the auto manufacturers and IT companies, the number of international patent classification risen may lead to disengagement rate declined.
28
Chun-LinWu y 吳俊霖. "Design and Testing of Adaptive LED Front Lighting System Using Dual Inertia Sensor for Autonomous Underwater Vehicle". Thesis, 2016. http://ndltd.ncl.edu.tw/handle/16185207549100004467.
Texto completoResumen
碩士國立成功大學
系統及船舶機電工程學系
104
This proposal presents an adaptive vehicle rotate angle system using dual micro-inertial sensor, which can apply to underwater environment. The system consists of vehicle rotate angle estimation module and PZT driving module to form adaptive front LED lighting system. Driving module is composed of two core components: symmetrical piezoelectric element and trussed preload structure. The design of the symmetrical piezoelectric element is utilizing the finite element analysis to simulate its motion characteristics, in order to get the fixed point and the driving point. The piezoelectric actuator of the preload structure design improves the general linear preload track type into torque track type, to further simplify the linear preload track into rotation track and introduce the truss structure component concept to ameliorate preload skew problem, and design a triangular truss-type preload mechanism, and then to integrate with SPE as a driving module. Finally realizing the active dynamic feedback sensing concept and enhance AUV or ROV in underwater exploration technology capability.
29
Fanelli, Francesco. "Development and Testing of Navigation Algorithms for Autonomous Underwater Vehicles". Doctoral thesis, 2018. http://hdl.handle.net/2158/1125920.
Texto completoResumen
As of today, autonomous underwater navigation can still be considered a challenging task; the determination of self-localization techniques for autonomous underwater vehicles is an open research topic, and many efforts are undertaken by researchers and companies to improve existing algorithms or to develop new solutions to increase the level of accuracy achievable with current vehicle technologies. In this framework, the research activity carried out during the Ph.D. period concentrated on the study of pose estimation algorithms for mobile robots, with special focus given to the underwater field. Starting from current solutions identified within the state of the art, the work was conducted in parallel on the topics of attitude and position estimation. A nonlinear attitude observer employing inertial and magnetic field data and suitable for use in the underwater field was derived; the estimated attitude constitutes an input for an UKF-based position estimator exploiting position, depth, and velocity measurements. Furthermore, the possibility of including the real-time estimation of sea currents within the developed estimators, relying only on already available measurements, was investigated. The performance of the resulting solutions was evaluated by means of simulations exploiting real navigation data or during suitable experimental test
campaigns which allowed to assess their effectiveness in a real-world scenario; the obtained results were satisfying, indicating that the derived algorithms may constitute a valid alternative to existing pose estimation strategies commonly adopted in the underwater field.
30
Zhao, Side. "Advanced control of autonomous underwater vehicles". Thesis, 2004. http://proquest.umi.com/pqdweb?index=0&did=775181771&SrchMode=1&sid=2&Fmt=2&VInst=PROD&VType=PQD&RQT=309&VName=PQD&TS=1233887056&clientId=23440.
Texto completo
31
OWENS, BRETT. "Concept Design and Testing of a GPS-less System for Autonomous Shovel-Truck Spotting". Thesis, 2013. http://hdl.handle.net/1974/7772.
Texto completoResumen
Haul truck drivers frequently have difficulties spotting beside shovels. This is typically a combination of reduced visibility and poor mining conditions. Based on first-hand data collected from the Goldstrike Open Pit, it was learned that, on average, 9% of all spotting actions required corrective movements to facilitate loading. This thesis investigates an automated solution to haul truck spotting that does not rely on the use of the satellite global positioning system (GPS), since GPS can perform unreliably. This thesis proposes that if spotting was automated, a significant decrease in cycle times could result.
Using conventional algorithms and techniques from the field of mobile robotics, vehicle pose estimation and control algorithms were designed to enable autonomous shovel-truck spotting. The developed algorithms were verified by using both simulation and field testing with real hardware. Tests were performed in analog conditions on an automation-ready Kubota RTV 900 utility truck. When initiated from a representative pose, the RTV successfully spotted to the desired location (within 1 m) in 95% of the conducted trials. The results demonstrate that the proposed approach is a strong candidate for an auto-spot system.Thesis (Master, Mining Engineering) -- Queen's University, 2013-01-28 09:49:20.584
32
Franchi, Matteo. "2D Forward Looking SONAR in Navigation Aiding: Development and Testing of Strategies for Autonomous Underwater Vehicles". Doctoral thesis, 2020. http://hdl.handle.net/2158/1183685.
Texto completoResumen
This work collects the results of the research activity on marine robotics carried out at the Mechatronics and Dynamic Modeling Laboratory (MDM Lab) of the Department of Industrial Engineering of the University of Florence (UNIFI DIEF) during the years 2014-2017. Reliable navigation systems are fundamental for Autonomous Underwater Vehicles (AUVs) to perform complex tasks and missions. It is well known that the Global Positioning System (GPS) cannot be employed in underwater scenarios; thus, during missions below the sea’s surface the real-time position is usually obtained with expensive sensors, such as the Doppler Velocity Log (DVL), integrated within a navigation filter such as an Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), or Dead Reckoning (DR) strategies. The main goal of this work is to develop and test a framework able to integrate a Forward-Looking SONAR (FLS), by means of linear speed estimations, into an underwater navigation system.
On the one hand, the proposed solution can work together with a standard navigation sensors set (comprising, for example, a DVL), and thus leading to a greater number of linear speed measurements. On the other hand, employing an FLS to aid navigation could potentially outline other advantages. Using an augmented set of devices able to provide navigation information represents an intrinsic boost in redundancy; DVL-denied scenarios, such as very close to the seafloor or other surfaces or when a substantial number of gaseous bubbles is present, could be managed. Indeed, as opposed to the DVL, the FLS possesses much more beams that are spread into a broader area, thus improving reliability.
DVL failings in the presence of bubbles are well-documented in the current literature and have been experienced during several tests at sea performed by UNIFI DIEF.
Conversely, the presence of bubbles, which can be noticed within FLS images as strong return echoes spots, is usually tolerable and not capable of jeopardizing FLS operations.
Moreover, although bigger AUVs enable the use of more sophisticated instrumentation and can carry a heavy payload, smaller AUVs are constrained to limited payload carrying capabilities. Hence, in addition to constituting a valuable research interest, multitasking onboard sensors represent a solution that offers compactness and avoids the use of some instruments.
Besides this, to better the dynamic modeling of the AUV, a light-weight online estimator for the longitudinal dynamics and a more realistic propulsion model are developed. Lastly, an Adaptive Unscented Kalman Filter (AUKF)-based navigation solution is proposed.
Offline validation, through the use of navigation data obtained during sea trials undertaken in La Spezia (Italy) at the NATO STO Centre for Maritime Research and Experimentation (CMRE), is presented. Afterward the results of real autonomous underwater missions performed in La Spezia (Italy) within the activities of the SEALab, the joint research laboratory between the Naval Experimentation and Support Center (Centro di Supporto e Sperimentazione Navale) (CSSN) of the Italian Navy and the Interuniversity Center of Integrated Systems for the Marine Environment (ISME), and at Vulcano Island, Messina (Italy) are reported.
33
"Search-based Test Generation for Automated Driving Systems: From Perception to Control Logic". Doctoral diss., 2019. http://hdl.handle.net/2286/R.I.53484.
Texto completoResumen
abstract: Automated driving systems are in an intensive research and development stage, and the companies developing these systems are targeting to deploy them on public roads in a very near future. Guaranteeing safe operation of these systems is crucial as they are planned to carry passengers and share the road with other vehicles and pedestrians. Yet, there is no agreed-upon approach on how and in what detail those systems should be tested. Different organizations have different testing approaches, and one common approach is to combine simulation-based testing with real-world driving.
One of the expectations from fully-automated vehicles is never to cause an accident. However, an automated vehicle may not be able to avoid all collisions, e.g., the collisions caused by other road occupants. Hence, it is important for the system designers to understand the boundary case scenarios where an autonomous vehicle can no longer avoid a collision. Besides safety, there are other expectations from automated vehicles such as comfortable driving and minimal fuel consumption. All safety and functional expectations from an automated driving system should be captured with a set of system requirements. It is challenging to create requirements that are unambiguous and usable for the design, testing, and evaluation of automated driving systems. Another challenge is to define useful metrics for assessing the testing quality because in general, it is impossible to test every possible scenario.
The goal of this dissertation is to formalize the theory for testing automated vehicles. Various methods for automatic test generation for automated-driving systems in simulation environments are presented and compared. The contributions presented in this dissertation include (i) new metrics that can be used to discover the boundary cases between safe and unsafe driving conditions, (ii) a new approach that combines combinatorial testing and optimization-guided test generation methods, (iii) approaches that utilize global optimization methods and random exploration to generate critical vehicle and pedestrian trajectories for testing purposes, (iv) a publicly-available simulation-based automated vehicle testing framework that enables application of the existing testing approaches in the literature, including the new approaches presented in this dissertation.Dissertation/Thesis
Doctoral Dissertation Computer Engineering 2019
34
"DeepCrashTest: Translating Dashcam Videos to Virtual Tests forAutomated Driving Systems". Master's thesis, 2019. http://hdl.handle.net/2286/R.I.55630.
Texto completoResumen
abstract: The autonomous vehicle technology has come a long way, but currently, there are no companies that are able to offer fully autonomous ride in any conditions, on any road without any human supervision. These systems should be extensively trained and validated to guarantee safe human transportation. Any small errors in the system functionality may lead to fatal accidents and may endanger human lives. Deep learning methods are widely used for environment perception and prediction of hazardous situations. These techniques require huge amount of training data with both normal and abnormal samples to enable the vehicle to avoid a dangerous situation.
The goal of this thesis is to generate simulations from real-world tricky collision scenarios for training and testing autonomous vehicles. Dashcam crash videos from the internet can now be utilized to extract valuable collision data and recreate the crash scenarios in a simulator. The problem of extracting 3D vehicle trajectories from videos recorded by an unknown monocular camera source is solved using a modular approach. The framework is divided into two stages: (a) extracting meaningful adversarial trajectories from short crash videos, and (b) developing methods to automatically process and simulate the vehicle trajectories on a vehicle simulator.Dissertation/Thesis
Video Demonstration
Masters Thesis Computer Science 2019
35
Tian, Yuchi. "Detect and Repair Errors for DNN-based Software". Thesis, 2021. https://doi.org/10.7916/d8-mfnq-3754.
Texto completoResumen
Nowadays, deep neural networks based software have been widely applied in many areas including safety-critical areas such as traffic control, medical diagnosis and malware detection, etc. However, the software engineering techniques, which are supposed to guarantee the functionality, safety as well as fairness, are not well studied. For example, some serious crashes of DNN based autonomous cars have been reported. These crashes could have been avoided if these DNN based software were well tested. Traditional software testing, debugging or repairing techniques do not work well on DNN based software because there is no control flow, data flow or AST(Abstract Syntax Tree) in deep neural networks. Proposing software engineering techniques targeted on DNN based software are imperative. In this thesis, we first introduced the development of SE(Software Engineering) for AI(Artificial Intelligence) area and how our works have influenced the advancement of this new area. Then we summarized related works and some important concepts in SE for AI area. Finally, we discussed four important works of ours.
Our first project DeepTest is one of the first few papers proposing systematic software testing techniques for DNN based software. We proposed neuron coverage guided image synthesis techniques for DNN based autonomous cars and leveraged domain specific metamorphic relation to generate oracle for new generated test cases to automatically test DNN based software. We applied DeepTest to testing three top performing self-driving car models in Udacity self-driving car challenge and our tool has identified thousands of erroneous behaviors that may lead to potential fatal crash.
In DeepTest project, we found that the natural variation such as spatial transformations or rain/fog effects have led to problematic corner cases for DNN based self-driving cars. In the follow-up project DeepRobust, we studied per-point robustness of deep neural network under natural variation. We found that for a DNN model, some specific weak points are more likely to cause erroneous outputs than others under natural variation. We proposed a white-box approach and a black-box approach to identify these weak data points. We implemented and evaluated our approaches on 9 DNN based image classifiers and 3 DNN based self-driving car models. Our approaches can successfully detect weak points with good precision and recall for both DNN based image classifiers and self-driving cars.
Most of existing works in SE for AI area including our DeepTest and DeepRobust focus on instance-wise errors, which are single inputs that result in a DNN model's erroneous outputs. Different from instance-wise errors, group-level errors reflect a DNN model's weak performance on differentiating among certain classes or inconsistent performance across classes. This type of errors is very concerning since it has been found to be related to many real-world notorious errors without malicious attackers. In our third project DeepInspect, we first introduced the group-level errors for DNN based software and categorized them into confusion errors and bias errors based on real-world reports. Then we proposed neuron coverage based distance metric to detect group-level errors for DNN based software without requiring labels. We applied DeepInspect to testing 8 pretrained DNN models trained in 6 popular image classification datasets, including three adversarial trained models. We showed that DeepInspect can successfully detect group-level violations for both single-label and multi-label classification models with high precision.
As a follow-up and more challenging research project, we proposed five WR(weighted regularization) techniques to repair group-level errors for DNN based software. These five different weighted regularization techniques function at different stages of retraining or inference of DNNs including input phase, layer phase, loss phase and output phase. We compared and evaluated these five different WR techniques in both single-label and multi-label classifications including five combinations of four DNN architectures on four datasets. We showed that WR can effectively fix confusion and bias errors and these methods all have their pros, cons and applicable scenario.
All our four projects discussed in this thesis have solved important problems in ensuring the functionality, safety as well as fairness for DNN based software and had significant influence in the advancement of SE for AI area.