Augmented Reality är ett brett fält med många olika användningsområden. Det hargjorts flera studier om hur AR kan hjälpa oss att lösa olika arbetsuppgifter på ett effektivare sätt samt att det ingjuter självsäkerhet hos användaren som utför en uppgiftmed hjälp av AR. AR har även letat sig in i tillverkningsindustrin som gör det möjligtför ingenjörer att följa en arbetsprocess med hjälp av AR.Kretskort som är en del av ett system som gör det möjligt för oss att exempelvis ha tillgång till datorer och mobiltelefoner och behöver ibland felsökas. Elektronikingenjörerbehöver ibland identifiera testpunkter och komponenter i den krets på kretskortet somska felsökas. Detta kan underlättas med AR. I denna studie kommer ett experimentatt genomförs för att se om kretskort kan lämpa sig väl som ett Marker-based target.Marker-based target är en bild som möjliggör objektspårning för exempelvis en Head-mounted display, i detta fall en HoloLens. Här efter kommer Marker-based target attbenämnas som mönsterbaserad bild.Experimentet kommer att genomföras med en testgrupp om 10 personer där samtliga kommer att få testa att använda en HoloLens i syftet att lokalisera komponentereller testpunkter på ett kretskort. Detta kommer sedan att jämföras med en traditionell metod genom att identifiera samma komponenter eller testpunkter med en PDF.Testpersonerna kommer även att få svara på en enkät som tittar på om denna metodlämpar sig för att identifiera komponenter med hjälp av AR. En applikation kommeratt skapas som möjliggör att implementera metoden och att testpersonerna kommerkunna använda ett gränssnitt i lokaliseringsprocessen. Resultatet visar att med HoloLensen i jämförelse med traditionella metoden är nästan likvärdig tidsmässigt för attlokalisera komponenter men att vanan av användandet av HoloLensen hade inflytandepå resultatet. / Augmented Reality is an broad field with many different areas of applications. Therehave been several studies on how AR can help us solve various tasks in a more efficientway and that it instills self-confidence in the user who performs a task with the help of AR. AR has also found its way into the manufacturing industry, which enables engineers to follow a work process with the help of AR. For example circuit boards that arepart of a system that makes it possible for us to have access to computers and mobilephones, does sometimes need to undergo troubleshooting process by engineers duringmanufacture.Electronics engineers sometimes need to identify test points and components on thecircuit board that needs troubleshooting. AR can help in this process. In this study, anexperiment will be conducted to see if circuit boards can be well suited as a marker-based target. Marker-based target is an image that enables object tracking for examplewith a Head-mounted display, in this case a HoloLens.The experiment will be conducted with a test group of 10 people where all will beallowed to test a HoloLens in order to locate components or test points on a circuitboard. This method will then be compared to a traditional method by identifying thesame components or test points with a PDF. The test persons will also have to answer aquestionnaire that looks at whether this method is suitable for identifying componentsusing AR. An application will be created that implements the method and that the testpersons will be able to use in the localization process. The results show that the traditional method in comparison with the HoloLens is equivalent for locating components,but the habit of using HoloLens had an influence on the result.

Artificial Intelligence (AI) is increasingly used in different parts of society for providing decision support in various activities. The agricultural sector is anticipated to benefit from an increased usage of AI and smart devices, a concept called smart farming technologies. Since the agricultural sector faces several simultaneous challenges, such as shrinking marginals, complicated pan-European regulations, and demands to mitigate the environmental footprint, there are great expectations that smart farming will benefit both individual farmers and industry stakeholders. However, most previous research focuses only on a small set of characteristics for implementing and optimising specific smart farming technologies, without considering all possible aspects and effects. This thesis investigates both technical and non-technical opportunities and hurdles when implementing AI in Swedish agricultural businesses. Three sectors in agriculture are scrutinized: arable farming, milk production and beef production. As a foundation for the thesis, a literature review revises former research on smart farming. Thereafter, an interview study with 27 respondents both explores the susceptibility and maturity of smart farming technologies and provides examples of technical requirements of three chosen applications of AI in agriculture. Findings of the study include a diverse set of aspects that both enable and obstruct the transition. Main identified opportunities are the importance smart farming has on the strategic agendas of several industry stakeholders, the general trend towards software technology as a service through shared machinery, the vast amount of existing data, and the large interest from farmers towards new technology. Contrasting, the thesis identifies main hurdles as technical and legislative challenges to data ownership, potential cybersecurity threats, the need for a well-articulated business case, and the sometimes lacking technical knowledge within the sector. The thesis concludes that the macro trend points towards a smart farming transition but that the speed of the transformation will depend on the resolutions for the identified obstacles.

I delar av byggbranschen finns idag ett gemensamt problem som kan beskrivas med termen dålig framförhållning. Detta försvårar möjligheten att se bygglogistiken i ett helhetsperspektiv, både för leverantören och entreprenören, men tyvärr även för övriga aktörer inblandade i byggproduktionsprojekt. Att uppnå bra bygglogistik med ökad hänsyn till miljön kräver ett gott samarbete mellan samtliga parter. Bygglogistik är planering flera (många) veckor framåt, i byggbranschen är det vanligt att det uppkommer oväntade händelser som leder till att tidsplaner för olika delar i byggprojektet justeras. I dagsläget är det lite för ofta som justeringar sker ”i sista minuten”. Detta leder till dyrare frakter, svårigheter att klara av en bra leveransprecision vilket i sin tur påverkar effektiviteten i bygglogistiken negativt. Problemen som upptäckts för de olika processerna inom bygglogistiken på Cramo är beskrivna med åtgärdsförslag, rekommendationer för effektivisering/förbättring av bygglogistik under rubriken probleminventering. För detta arbete har faktainsamlingen omfattat metoderna i lean construction som JIT- Just in time och jämförelser mellan olika tilläggstjänsterna som används i byggbranschen idag. Även fakta om kommunikationssystemet, TMS, samt läran om transport och materialhantering har undersökts och analyserats. Fakta från intervjuerna som beskriver TMS, transport metoder och lösningar samt hur hantering av material sker och vad man har för mål i framtiden har redovisats i bakgrunden, resultatet samt i analysen. Bygglogistik är kortfattat en samlings ord för alla aktiviteter som leder till att rätt produkt ska vara på rätt plats vid rätt tillfälle. “Tänk efter före”(Sundin, 2019) / In the construction industry, there is a general industry problem that can be described by the term poor advance planning. This makes it difficult to see the construction logistics in a holistic perspective, both for the supplier and the contractor, but unfortunately also for other parties involved in construction production projects. Achieving good building logistics with consideration for the environment requires good cooperation between all parties. Building logistics is planning several (many) weeks ahead, in the construction industry it is common that unexpected events arise that lead to adjusting timetables for different parts of the construction project. At present, it is a little too often that adjustments are made "at the last minute". This leads to more expensive freight, difficulties in coping with a good delivery precision, which in turn negatively affects the efficiency of construction logistics. Problems that has been discovered for the different parts of the process within building logistics in CRAMO are presented with solutions, recommendation of efficiency/improvement for building logistics under the title problem inventory,”problem inventering”. In problem inventory, these are described while action proposals are presented later in the report during the analysis. For this work, the facts have included the methods in lean construction as JIT- Just in time and comparisons between various additional services used in the construction industry today. Facts about the communication system, TMS, and the doctrine of transport and material handling have also been investigated and analyzed. Facts from the interviews that describe TMS, transport methods and solutions as well as how material handling takes place and what one has for goals in the future have been reported in the background, the result and in the analysis. Building logistics is briefly a term for all activities that lead to the right product being in the right place at the right time. "Think ahead" (Sundin, 2019)

The focus of this study was to evaluate different search engines on Swedish text. Information retrieval is widely used by both people and organizations, and it is important to be able to efficiently retrieve needed information at the right time. The study determined that relevance and speed are the most important factors in search engines. The evaluation measures the precision and recall which are relevance measurements, and speed of two search engines, Elastic search and MarkLogic. The evaluation has determined that there is no significant difference in the relevance of the retrieved results between the engines. The evaluation has also determined that there is a statistically significant difference in speed between the engines, with Elastic search outperforming MarkLogic. Both search engines performed very well in terms of successful searches, meaning to return a relevant document in the first 20 results. Both engines succeeded in fulfilling the information need 96% of the time. / Fokus för denna studie var att utvärdera olika sökmotorer på svensk text. Informationshämtning används i stor utsträckning av både människor och organisationer, och det är viktigt att effektivt kunna hämta nödvändig information vid rätt tidpunkt. Studien fastställde att relevans och hastighet är de viktigaste faktorerna för sökmotorer. Utvärderingen mäter precision och recall som är relevansmätvärden och responstid som hastighetmätvärde för två sökmotorer, Elasticsearch och MarkLogic. Utvärderingen har visat att det inte finns någon signifikant skillnad i relevansen av de hämtade resultaten mellan motorerna. Utvärderingen har också visat att det finns en statistiskt signifikant skillnad i hastighet mellan motorerna, där Elasticsearch överträffar MarkLogic. Båda sökmotorerna presterade väldigt bra när det gäller lyckade sökningar, vilket innebär att returnera ett relevant dokument i de första 20 resultaten. Båda motorerna lyckas uppfylla informationsbehovet 96% av tiden.

Many people around the world live in islands with mountainous lands. These people plant upland rice for their staple food. However, farming in these disadvantageous farms requires high input and gains low yield. One possible solution for improving productivity is by utilizing precision seeding. This study is focused on developing a robot for precision seeding of rice that is suitable for farming in rough terrain. A bioinspired human arm robot for picking seeds and making a hole on the ground was developed in accordance with Aalborg method for bioinspired design. Development of a mobile platform to carry the robot arm was done using hybrid wheel and leg (wheg) wheels and skid steering locomotion. The NugalBot could climb obstacle and perform precision rice seed singulation. However, the performance of the mechanisms for making hole and for covering the hole was dependent on the hardness of the soil.

Quantization of Neural Networks is popular technique for adopting computation intensive Deep Learning applications to edge devices. In this work, low bit mixed precision quantization of FPN-Resnet18 model trained for the task of semantic segmentation is explored using Cityscapes and Arriver datasets. The Hessian information of each layer in the model is used to determine the bit precision for each layer and in some experiments the bit precision for the layers are determined randomly. The networks are quantization-aware trained with bit combinations 2, 4 and 8. The results obtained for both Cityscapes and Arriver datasets show that the quantization-aware trained networks with the low bit mixed precision technique offer a performance at par with the 8-bit quantization-aware trained networks and the segmentation performance degrades when the network activations are quantized below 8 bits. Also, it was found that the usage of the Hessian information had little effect on the network’s performance. / Kvantisering av Neurala nätverk är populär teknik för att införa beräknings-intensiva Deep Learning -applikationer till edge-enheter. I detta arbete utforskas låg bitmixad precisionskvantisering av FPN-Resnet18-modellen som är utbildad för uppgiften för semantisk segmentering med hjälp av Cityscapes och Arriverdatauppsättningar. Hessisk information från varje lager i modellen, används för att bestämma bitprecisionen för respektive lager. I vissa experiment bestäms bitprecision för skikten slumpmässigt. Nätverken är kvantiserings medvetna utbildade med bitkombinationer 2, 4 och 8. Resultaten som erhållits för både Cityscapes och Arriver datauppsättningar visar att de kvantiserings medvetna utbildade nätverken med lågbit blandad precisionsteknik erbjuder en prestanda i nivå med 8-bitars kvantiseringsmedvetna utbildade nätverk och segmenteringens prestationsgrader när nätverksaktiveringarna kvantiseras under 8 bitar. Det visade sig också att användningen av hessisk information hade liten effekt på nätets prestanda.

No description available.

ITC/USA 2008 Conference Proceedings / The Forty-Fourth Annual International Telemetering Conference and Technical Exhibition / October 27-30, 2008 / Town and Country Resort & Convention Center, San Diego, California / The U.S. Army Research Laboratory (ARL) is exploring technologies to provide low-cost precision fires, applicable across both direct and indirect fire weapon systems. One of these applications involves a forward observer (FO) designating the target with a laser spot and a seeker on-board the munition detecting the reflected energy to allow terminal guidance. This approach, referred to as semi-active laser (SAL) guidance, has been utilized on numerous air-delivered munitions to include bombs, missiles and projectiles. However, the cost of these systems, driven by high quality optics, high sensitivity detectors and specialized electronics, has hampered their migration into gun-fired munitions such as mortars, artillery and grenades. To explore, develop and demonstrate minimal cost solutions, ARL invested in an Army Technical Objective (ATO) called Smaller, Lighter, Cheaper Munition Components (SLCMC). Specifically, SAL seeker hardware, predicated upon commercial components (COTS) and mass production techniques, is being prototyped for use with gun launched projectiles and laser target designators. The seeker system is comprised of several printed circuit board boards, a microprocessor, a quad-photo detector and, a molded optical lens unit. This seeker is designed to rapidly update the projectile boresight angle, interface with other strap-down sensors, and feed data into an on-board guidance, navigation & control (G,N&C) system to allow for projectile maneuvers. The seeker design and basic characteristics are discussed and presented through-out the paper and presentation.

ITC/USA 2015 Conference Proceedings / The Fifty-First Annual International Telemetering Conference and Technical Exhibition / October 26-29, 2015 / Bally's Hotel & Convention Center, Las Vegas, NV / Northrop Grumman Corporation's B-2 Flight Test Instrumentation team is revamping its entire Data Acquisition System (DAS) to be an Ethernet based network (EBN) system that will provide simplified wiring, higher speeds, greater capacity, and control over the data. The old system became obsolete in terms of capability and maintainability. New on-board avionic systems also demand that the Flight Test Instrumentation group (INSTR) accommodate fiber and high speed Ethernet data. In addition, the footprint and location for INSTR systems and components will be moved to remote areas. INSTR engineering selected the Teletronics Technology Corporation's Ethernet networked Data Acquisition Units (DAUs), known as MnDAUs, as the core system. Prior to the first flight utilization of the new INSTR DAS will undergo extensive lab and field testing to assure flight test effectiveness and the accuracy of all necessary data products. The goal is to acquire and employ the best system available while avoiding costly lessons.

To enhance test security of high stakes tests, it is vital to understand the way various exposure control strategies function under various IRT models. To that end the present dissertation focused on the performance of several exposure control strategies under the generalized partial credit model with an item pool of 100 and 200 items. These procedures are relatively easy to implement and have shown promise as an alternative to more complex exposure control strategies. Through unique algorithms these procedures select an item for administration from a subset of items in the item pool. The five procedures examined for efficacy were the modified within .10 logits, restricted modified within .10 logits, randomesque, restricted randomesque, and progressive restricted procedures. The modified within .10 logits, restricted modified within .10 logits, and randomesque, and restricted randomesque procedures select an item for administration from a subset of optimal items. To test the effect of the number of items available for selection in this subset, 3, 6, and 9 items were made available for selection in these procedures. Maximum information item selection was used as a base line, no exposure control, condition. The progressive restricted, restricted randomesque, and restricted modified within .10 logits procedures were found to optimally protect test security while not significantly degrading measurement precision. The restricted forms of the randomesque and modified within .10 logits procedures proved superior to their base procedures, particularly in controlling average maximum exposure rate. The incrementation of item group size in the modified within .10 logits, restricted modified within .10 logits, and randomesque, and restricted randomesque procedures demonstrated that increasing the item group size provided better test security while not significantly degrading measurement precision. Additionally, in general, the increase of the item pool size from 100 to 200 improved measurement precision and test security. Implications towards practical application are discussed and directions for future research are suggested. / text