Quality assurance is an important part of many industrial processes that involves different methods of determining the quality of a product. One of these methods is deflectometry, a method that uses a display to show patterns and a camera to capture the reflection of these patterns through a specular object which is being measured. The goal and end result of these measurements is a height profile of the object's surface. While there are many ways of performing deflectometry through using different types of patterns and setups, this project focuses on using only a single smartphone to capture all data required for measurements. This involves showing a sequence of patterns on the smartphone's display and using its front-facing camera to capture the reflection through the specular object. The patterns chosen for this purpose are binary checkerboard patterns that are simple enough for the camera to capture in a good way and efficient enough to perform the calculations in a reasonable time frame. Using this method, the ability of a smartphone to perform deflectometric measurements was evaluated by testing on several different types of mirrors as well as on a real car body. The method can produce results that closely replicate the real world object and can calculate quantities that are used to measure the quality of car assembly, often with an accuracy of well under 1 mm. The method can handle a large variety of reflective objects at varying distances and form while only requiring some known parameters of the smartphone used in testing.

Outside of the city environment, there are many unstructured and rough environments that are challenging in vehicle navigation tasks. In these environments, vehicle vibrations caused by rough terrain can be harmful for humans. In addition, a human operator can not work around the clock. A promising solution is to use artificial intelligence to replace human operators. I test this by using the artificial intelligence technique know as reinforcement learning, with the algorithm Proximal Policy Optimization, to perform some basic locomotion tasks in a simulated environment with a simple terrain vehicle. The terrain vehicle consists of two chassis, each having two wheels attached, connected to each other with an articulation joint that can rotate to turn the vehicle. I show that a trained model can learn to operate the terrain vehicle and complete basic tasks, such as finding and following a path while avoiding obstacles. I tested robustness by evaluating performance on sloped terrains with a model trained to operate on flat ground. The results from the tests with different slopes show that, for most environments, the trained model could handle slopes up to around 7.5-10 degrees without much issue, even though it had no way of detecting the slope. This tells us that the models can perform their tasks quite well even when disturbances are introduced, as long as these disturbances doesn't require them to significantly change their behaviors.

Thermal cameras have historically been of interest mainly for military applications. Increasing image quality and resolution combined with decreasing price and size during recent years have, however, opened up new application areas. They are now widely used for civilian applications, e.g., within industry, to search for missing persons, in automotive safety, as well as for medical applications. Thermal cameras are useful as soon as it is possible to measure a temperature difference. Compared to cameras operating in the visual spectrum, they are advantageous due to their ability to see in total darkness, robustness to illumination variations, and less intrusion on privacy. This thesis addresses the problem of detection and tracking in thermal infrared imagery. Visual detection and tracking of objects in video are research areas that have been and currently are subject to extensive research. Indications oftheir popularity are recent benchmarks such as the annual Visual Object Tracking (VOT) challenges, the Object Tracking Benchmarks, the series of workshops on Performance Evaluation of Tracking and Surveillance (PETS), and the workshops on Change Detection. Benchmark results indicate that detection and tracking are still challenging problems. A common belief is that detection and tracking in thermal infrared imagery is identical to detection and tracking in grayscale visual imagery. This thesis argues that the preceding allegation is not true. The characteristics of thermal infrared radiation and imagery pose certain challenges to image analysis algorithms. The thesis describes these characteristics and challenges as well as presents evaluation results confirming the hypothesis. Detection and tracking are often treated as two separate problems. However, some tracking methods, e.g. template-based tracking methods, base their tracking on repeated specific detections. They learn a model of the object that is adaptively updated. That is, detection and tracking are performed jointly. The thesis includes a template-based tracking method designed specifically for thermal infrared imagery, describes a thermal infrared dataset for evaluation of template-based tracking methods, and provides an overview of the first challenge on short-term,single-object tracking in thermal infrared video. Finally, two applications employing detection and tracking methods are presented.

Pose (position and orientation) tracking in room-scaled environments is an enabling technique for many applications. Today, virtual reality (vr) and augmented reality (ar) are two examples of such applications, receiving high interest both from the public and the research community. Accurate pose tracking of the vr or ar equipment, often a camera or a headset, or of different body parts is crucial to trick the human brain and make the virtual experience realistic. Pose tracking in room-scaled environments is also needed for reference tracking and metrology. This thesis focuses on an application to metrology. In this application, photometric models of a photo studio are needed to perform realistic scene reconstruction and image synthesis. Pose tracking of a dedicated sensor enables creation of these photometric models. The demands on the tracking system used in this application is high. It must be able to provide sub-centimeter and sub-degree accuracy and at same time be easy to move and install in new photo studios. The focus of this thesis is to investigate and develop methods for a pose tracking system that satisfies the requirements of the intended metrology application. The Bayesian filtering framework is suggested because of its firm theoretical foundation in informatics and because it enables straightforward fusion of measurements from several sensors. Sensor fusion is in this thesis seen as a way to exploit complementary characteristics of different sensors to increase tracking accuracy and robustness. Four different types of measurements are considered; inertialmeasurements, images from a camera, range (time-of-flight) measurements from ultra wide band (uwb) radio signals, and range and velocity measurements from echoes of transmitted acoustic signals. A simulation study and a study of the Cramér-Rao lower filtering bound (crlb) show that an inertial-camera system has the potential to reach the required tracking accuracy. It is however assumed that known fiducial markers, that can be detected and recognized in images, are deployed in the environment. The study shows that many markers are required. This makes the solution more of a stationary solution and the mobility requirement is not fulfilled. A simultaneous localization and mapping (slam) solution, where naturally occurring features are used instead of known markers, are suggested solve this problem. Evaluation using real data shows that the provided inertial-camera slam filter suffers from drift but that support from uwb range measurements eliminates this drift. The slam solution is then only dependent on knowing the position of very few stationary uwb transmitters compared to a large number of known fiducial markers. As a last step, to increase the accuracy of the slam filter, it is investigated if and how range measurements can be complemented with velocity measurement obtained as a result of the Doppler effect. Especially, focus is put on analyzing the correlation between the range and velocity measurements and the implications this correlation has for filtering. The investigation is done in a theoretical study of reflected known signals (compare with radar and sonar) where the crlb is used as an analyzing tool. The theory is validated on real data from acoustic echoes in an indoor environment.

The goal of this thesis is to use fringe-pattern phase analysis to calibrate the distortion of a camera lens. The benefit of using this method is that the distortion can be calculated using data from each individual pixel and the methodology does not need any model. The phase used to calibrate the images is calculated in two different ways, either utilizing the monogenic signal or through fringe-pattern phase analysis. The calibration approaches were also validated through different methods. Primarily by utilizing the Hough transform and calibrating simulated distortion. The thesis also introduces a validation approach utilizing the phase orientation calculated through the monogenic signal. The thesis also implements different approaches such as flat field correction to limit the impact of the image sensor noise to mitigate the phase noise. It is also investigated which fringe-pattern frequencies are best suited for calibration through comparative analysis. The comparative analysis identified problems with too high and low frequencies of the fringe-patterns when calibrating using fringe-pattern phase analysis.

Denna rapport är ett kandidatexamensarbete vid Kungliga Tekniska Högskolan och ärämnat att fungera som en handbok för autonoma fordon, detta genom att arbetet är indelati tre huvudområden; det autonoma fordonets uppbyggnad, det autonoma fordonetshistoria samt det autonoma fordonets roll i det framtida samhället. Syftet med arbetet äratt ge en övergripande helhetsbild av dessa tre områden. Arbetet har avgränsats till attendast fokusera på personbilsindustrin, och tillvägagångssättet har varitinformationsinsamling från vetenskapliga källor. Ett autonomt fordon använder sig av ett flertal olika sensorer och system för att kunnaläsa av sin omgivning och därefter fatta beslut som resulterar i konkreta handlingar. Deviktigaste systemen innefattar lidar, radar, kameror, ultraljudssensorer, GPS, IMU samtkodare. Den största delen av utvecklingen av autonoma fordon har skett under de senastehundra åren. Startskottet kan anses vara Francis Houdinas något primitiva bil “LinriccanWonder” år 1925. Idag, cirka hundra år senare, har flera stora fordonstillverkare valt attlägga ett visst fokus på att utveckla dessa betydligt mer avancerade fordon. Fördelarnamed dessa fordon är många, och en integration av dessa på vägarna kan bland annat ledatill ett bättre trafikflöde, säkrare vägar, och en ökad grad av självständighet inom ett flertalgrupper i samhället. Dock finns det ett flertal utmaningar att ta itu med innan dessa fordonkan bli en möjlighet, vilka innefattar frågor kring bland annat cybersäkerhet, integritetoch ansvar. Frågan kring ansvar är möjligtvis den största flaskhalsen i utvecklingen ochhandlar om vem som bär ansvaret om ett autonomt fordon skulle infinna sig i en olycka,vilket om det läggs på fordonstillverkaren möjligtvis kan stjälpa utvecklingen av dessafordon.

The reliance on fossil fuels needs to be fizzled out from every part of modern society. Construction equipment is one industry in which the transition away from fossil fuels is underway in favor of electric vehicles. The transition to electric vehicles is affecting the entire system of construction sites, including enhanced feasibility and incentive for increased autonomy. This thesis aims at identifying system-level design implications stemming from the electric transition of construction equipment. By using biomimicry and drawing inspiration from nature, the goal is to produce a functioning product that fulfills the requirements created based on the needs of the users. Within this thesis, Design Thinking is used to identify the needs in theconstruction equipment sector which are then used as the input for an iterative process of reverse biomimetics. The resulting concepts are tested and evaluated in several design sprints, the findings from which result in the final solution. Aided by the Design Thinking methodology human-centered approach, the solution is optimized for the users. A total of 25 needs were identified within eight different categories. A total of seven design sprints were conducted with the findings accumulating in a sensor cleaning system for autonomous vehicles inspired by the pendulum-like oscillating motion of mammals shedding precipitation. Biomimicry is identified as a promising tool to aid in the creative process by introducing novel perspectives and approaches to the problem space. The evolving development and use of autonomous constructionequipment results in the need for clear and unhindered sensor systems which the solution developed in this thesis provides. / Användandet av förbränningsmotorer behöver forslas ut från varje del av det moderna samhället. Konstruktionsindustrin är i starten av övergången till elektriska maskiner. Det är en markant övergång som påverkar varje del av byggarbetsplatsen, i synnerhet motivet till att använda autonoma maskiner som lösning på den mindre energidensitet eldrivna anläggningsmaskiner innebär. Den här uppsatsen undersöker vilka design implikationer som berörs från den elektriska övergången på ett systemperspektiv kring använd- nignen av anläggningsmaskiner, och hur de ska lösas. Som stöd för att lösa de implikationerna används biomikry. Genom att använda biomikry sökes den lösningen som bäst uppfyller de design implika- tionerna som uppstått och behoven som utforskatts. Inom den här uppsattsen används Design Thinking for att identifiera de behoven som uppstått inom industrin. De fynden blir basen för an- vändandet av en specifik del av biomimikry, nämnligen motsatt biomi- tik. De koncepten som utvecklas testas och evalueras i flera så kallade Design sprints, där fynden från flera Design sprints utformar slut- lösningen. Med stöd från människofokuset av Design Thinking blir slutlösningen direkt optimerad för användaren. Totalt 25 behov identifierades inom åtta kategorier. Totalt sju Design sprints utfördes där de induviduella resultaten utformade en slutlös- ning i form av ett sensortvättsystem för autonoma fordon, vilket tagit inspiration från den svängande rörelsen som används av däggdjur för att få bort vatten. Lösningen innebär ohindrad funcktunalitet av sen- sorerna på autonoma anläggningsmaskiner oberoende av väder eller klimat. Biomimikry har identifierats som ett lovande verktyg för att öka den kreativa processen genom att introducera nytänkande perspektiv där problemet adresseras på ett annorlunda sätt. Det utökade behovet för autonoma anläggningsmaskiner har resulterat i ett ohindrade sensor rengöringssystem vilket är specifikt det den här uppsatsen behandlar.

The development and application of artificial intelligence (AI) for military purposes are increasing rapidly in many parts of the world. Military powers are driving programs aimed at the advantages that AI can generate. Simultaneously, ethical questions arise concerning autonomous military systems. This study aims to provide clarity on how future Swedish officers with different backgrounds within the profession relate to the ethical issues that accompany the use of autonomous weapon systems. In this study, the respondents are presented with two fictitious scenarios based on the principles of distinction and proportionality, describing ethically problematic attacks that affect civilians. In each scenario, respondents are asked to take a stance on attacks carried out with different degrees of autonomy. The results of the study show that future officers consider the ethical defensibility of an attack to decrease as the degree of autonomy in the weapon system used increases.

This thesis explores the application of Contrastive Language-Image Pre-Training (CLIP), a vision-language model, in an automated video surveillance system for anomaly detection. The ability of CLIP to perform zero-shot learning, coupled with its robustness against minor image alterations due to its lack of reliance on pixel-level image analysis, makes it a suitable candidate for this application. The study investigates the performance of CLIP in tandem with various anomaly detection algorithms within a visual surveillance system. A custom dataset was created for video anomaly detection, encompassing two distinct views and two varying levels of anomaly difficulty. One view offers a more zoomed-in perspective, while the other provides a wider perspective. This was conducted to evaluate the capacity of CLIP to manage objects that occupy either a larger or smaller portion of the entire scene. Several different anomaly detection methods were tested with varying levels of supervision, including unsupervised, one-class classification, and weakly- supervised algorithms, which were compared against each other. To create better separation between the CLIP embeddings, a metric learning model was trained and then used to transform the CLIP embeddings to a new embedding space. The study found that CLIP performs effectively when anomalies take up a larger part of the image, such as in the zoomed-in view where some of the One- Class-Classification (OCC) and weakly supervised methods demonstrated superior performance. When anomalies take up a significantly smaller part of the image in the wider view, CLIP has difficulty distinguishing anomalies from normal scenes even using the transformed CLIP embeddings. For the wider view the results showed on better performance for the OCC and weakly supervised methods.

Grasping novel objects is challenging because of incomplete object data and because of uncertainties inherent in real world applications. To robustly perform grasps on previously unseen objects, feedback from touch is essential. In our research, we study how information from touch sensors can be used to improve grasping novel objects. Since it is not trivial to extract relevant object properties and deduce appropriate actions from touch sensing, we employ machine learning techniques to learn suitable behaviors. We have shown that grasp stability estimation based on touch can be improved by including an approximate notion of object shape. Further we have devised a method to guide local grasp adaptations based on our stability estimation method. Grasp corrections are found by simulating tactile data for grasps in the vicinity of the current grasp. We present several experiments to demonstrate the applicability of our methods. The thesis is concluded by discussing our results and suggesting potential topics for further research. / Att greppa nya föremål är utmanande, både eftersom roboten inte har fullständig information om objekten och på grund av den inneboende osäkerheten i verkliga tillämpningar. Återkoppling från känselsensorer är viktigt för att kunna greppa föremål som inte påträffats tidigare. I vår forskning så studerar vi hur information från känselsensorer kan användas för att förbättra greppandet av nya föremål. Eftersom det är svårt att extrahera relevanta egenskaper om föremål och härleda lämpliga åtgärder, baserat på känselsensorer, så har vi använt maskininlärning för att lära roboten lämpliga beteenden. Vi har visat att uppskattningar av stabiliteten av ett grepp baserat på känselsensorer kan förbättras genom att även använda en grov approximation av föremålets form. Vi har även konstruerat en metod som vägleder lokala justeringar av grepp, baserat på vår metod som uppskattar stabiliteten av ett grepp. Dess justeringar hittas genom att simulera känselsensordata för grepp i närheten av det nuvarande greppet. Vi presenterar flera experiment som demonstrerar tillämpbarheten av våra metoder. Avhandlingen avslutas med en diskussion om våra resultat och förslag på möjliga ämnen för fortsatt forskning. / <p>QC 20170510</p>