Browsing by study line "Mjukvarusystem"

This thesis introduces Robot Configurator, a software for performing variability configuration for a robot cooperation software named Cooperative Brain Service. The variability configuration problem concerns selecting appropriate combinations of components to form a valid complete product. It is applicable to the realm of software as well, as there are approaches where an instance of software is formed from different pre-made components. Cooperative Brain Service is the main product of Creative and Adaptive Cooperation between Diverse Autonomous Robots (CACDAR), a University of Helsinki research project. One of the main principles of the project is enabling cooperation between various types of robots. In Cooperative Brain Service, this is taken into account by having support for any new robots and actions be added as new modules. In the current implementation, the modules to be loaded are determined during startup, through a manually written JSON configuration file. The problem is, managing such JSON files requires beforehand knowledge of what modules there are implemented in Cooperative Brain Service. As it is crucial to be able to flexibly experiment with different cooperation scenarios in CACDAR, I design Robot Configurator as a tool to assist in the creation and management of these JSON configuration files. The main contribution of this thesis is Robot Configurator, for which I provide an architecture description that describes and documents it from multiple stakeholder perspectives. Additionally, a novel approach to modeling variability involved in the initialization of Cooperative Brain Service is also introduced.

Ohjelmistorobotiikka on joukko teknologioita, joiden avulla voidaan automatisoida rutiiniprosesseja toimistotyössä. Ohjelmistorobotiikan hyödyntämisen päätavoitteena onkin korvata ihmisten toistuvia ja tylsiä toimistotehtäviä automaation avulla. Kun ohjelmistorobotti suorittaa tehtävän prosessia, se jäljittelee ihmisen toimintaa. Tämän pro gradu tutkimuksen aiheena on ohjelmistorobotiikan hyödyntämisen edellytykset. Tutkimuksen tavoitteena on tutkia millaisia taitoja ja kyvykkyyksiä vaaditaan organisaatiolta, jotta se voi automatisoida liiketoimintaansa ohjelmistorobotiikan avulla. Tuloksia voidaan hyödyntää tulevissa ohjelmistorobotiikka-automatisoinneissa. Tämä tutkimus on kvalitatiivinen eli laadullinen semi-strukturoitu haastattelu. Tutkimuksen aineisto kerättiin kyselylomakkeen avulla tehtävistä haastatteluista. Kyselylomakkeen tukena käytettiin kirjallisuuskatsausta. Tutkimuksessa käytettiin taustatiedon hakuun lumipallo-otantaa. Tutkimuksessa hyödyntämisen edellytyksiä tutkittiin kolmesta näkökulmasta; prosessin, asiantuntijan ja organisaation näkökulmasta. Näin ollen tutkimuksen tuloksetkin kuvaavat näitä osa-alueita erikseen. Kuitenkin kaikilla osa-alueilla on oleellinen suhde toisiinsa. Tutkimuksessa saatujen tulosten perusteella tärkeä vaatimus organisaation näkökulmasta automatisoinnille on, että automatisoitava prosessi pystytään kuvaamaan tarkasti ja voidaan ylipäätään automatisoida. Asiantuntijoilta vaaditaan yleisesti loogista ajattelukykyä, teknologiaymmärrystä ja kommunikaatiotaitoa. Edellytettävät taidot ovat riippuvaisia siitä, minkä alan asiantuntija on kyseessä. Roolista riippumatta vaaditaan perusymmärrystä ohjelmoinnista ja RPA-työkaluista. Prosessilta vaaditaan, että se on rutiininomainen, toistuva ja sääntöperusteinen. Automatisoinnin on myös oltava kannattavaa. Tutkimuksen tuloksia voivat hyödyntää kaikki, jotka miettivät työtehtävien automatisointia ohjelmistorobotiikan avulla. ACM Computing Classification System (CCS): •Social and professional topics~Professional topics~Computing and business~Automation •Information systems~Information systems applications~Process control systems

Modern software systems increasingly consist of independent services that communicate with each other through their public interfaces. Requirements for systems are thus implemented through communication and collaboration between different its services. This creates challenges in how each requirement is to be tested. One approach to testing the communication procedures between different services is end-to-end testing. With end-to-end testing, a system consisting of multiple services can be tested as a whole. However, end-to-end testing confers many disadvantages, in tests being difficult to write and maintain. When end-to-end testing should adopted is thus not clear. In this research, an artifact for continuous end-to-end testing was designed and evaluated it in use at a case company. Using the results gathered from building and maintaining the design, we evaluated what requirements, advantages and challenges are involved in adopting end-to-end testing. Based on the results, we conclude that end-to-end testing can confer significant improvements over manual testing processes. However, because of the equally significant disadvantages in end-to-end testing, their scope should be limited, and alternatives should be considered. To alleviate the challenges in end-to-end testing, investment in improving interfaces, as well as deployment tools is recommended.

The techniques used to program quantum computers are somewhat crude. As quantum computing progresses and becomes mainstream, a more efficient method of programming these devices would be beneficial. We propose a method that applies today’s programming techniques to quantum computing, with program equivalence checking used to discern between code suited for execution on a conventional computer and a quantum computer. This process involves determining a quantum algorithm’s implementation using a programming language. This so-called benchmark implementation can be checked against code written by a programmer, with semantic equivalence between the two implying the programmer’s code should be executed on a quantum computer instead of a conventional computer. Using a novel compiler optimization verification tool named CORK, we test for semantic equivalence between a portion of Shor’s algorithm (representing the benchmark implementation) and various modified versions of this code (representing the arbitrary code written by a programmer). Some of the modified versions are intended to be semantically equivalent to the benchmark while others semantically inequivalent. Our testing shows that CORK is able to correctly determine semantic equivalence or semantic inequivalence in a majority of cases.

The goal of real-time rendering is to produce synthetic and usually photorealistic images from a virtual scene as part of an interactive application. A scene is a set of light sources and polygonal objects. Photorealism requires a realistic simulation of light, but it contains a recursive problem where light rays can bounce between objects countless of times. The objects can contain hundreds of thoudands of polygons, so they cannot be processed recursively in real-time. The subject of this thesis is a voxel-based lighting method, where the polygonal scene is processed into a voxel grid. When calculating the indirect bounces of light, we can process a small amount of voxels instead of the vast amount of polygons. The method was introduced in 2011, but it didn't gain much popularity due to its performance requirements. In this thesis, we studied the performance and image quality of the voxel-based lighting algoritm with a modern, low-cost graphics card. The study was conducted through design research. The artefact is a renderer that produces images with the voxel-based algorithm. The results show that the algorithm is capable of a high frame rate of 60 images per second in a full-hd resolution of 1920x720 pixels. However, the algorithm consumes most of the time spent forming the image, which doesn't leave much time to simulate a game world for example. In addition, the voxelization of the scene is a slow operation, which would require some application-specific optimizations to be performed every frame in order to support dynamically moving objects. The image quality improves greatly when compared to a scene that doesn't calculate indirect light bounces, but there is a problem of light bleeding through solid objects.

Cultural Heritage (CH) collections, data, and artefacts used to be available mainly in galleries, libraries, archives, and museums (GLAM). Due to digitisation, huge collections of CH data are now available online. Since the appearance of these data online, CH data have been searched, analysed and presented using various methods and visualisations. However, many of these methods do not take adequate advantage of the nature of CH data. Even when the CH data dimensions are utilised, the choices of visualisation sometimes fall short. This thesis is a part of the Finnish Archaeological Finds Recording Linked Open Database (SuALT) project. The project aims to develop digital Web services to cater for archaeological finds made by members of the public, especially metal detectorists through ‘citizen science’. Therein, a portion of an innovative prototype that tries to enable serendipitous knowledge discovery by making searching and visualising CH data seamless is presented. The artefact would serve as part of the basis of the FindSampo portal. The prototype is not implemented from scratch, but it is based on the Sampo model, the Sampo-UI framework and several third-party libraries and visualisations. The prototype includes mainly a user-centric faceted search engine, keyword searches to filter the facet options. It also includes data-analytic tools for visualising the filtered data on a table, a timeline chart, pie chart, line chart, maps and an option to export the results in CSV form for further analysis in external tools. However, this thesis focuses on the faceted search engine and the timeline visualisation that presents the spatio-temporal nature of the SuALT data. The portion of the prototype was evaluated based on three principles: serendipity, generosity, and criticality. And the result of the user experience survey suggests that the prototype could provide a good starting point to explore the find collection, make access to individual find information easier, improves the serendipitous discovery of archaeological find data and also make the find data analysis and interpretation easier.

Johtamisjärjestelmät ovat johtamisen tueksi luotuja työkaluja, joilla pystytään arvioimaan tilannekuvaa, suunnittelemaan operaatioita ja toimintalinjoja sekä vastaanottamaan ja lähettämään tietoa toimintaympäristöön. Puolustusorganisaatioissa erilaisia simulaattoreita käytetään kouluttamaan henkilöstöä tositilanteita varten oikeassa tai virtuaalisessa ympäristössä. Virtuaaliympäristössä kalliiden ja vaikeasti luotavien tilanteiden harjoittelu pystytään toteuttamaan kustannustehokkaasti ja mielekkäästi. Tutkielman tarkastelukohteena on se, miten ja millä tekniikoilla on kannattavinta mahdollistaa simulaattoreiden ja johtamisjärjestelmien yhteentoimivuus. Samalla selvitetään konkreettiset hyödyt, jotka muodostuvat simulaattoreiden ja johtamisjärjestelmien yhteistoiminnan tuloksena saavutettavien kyvykkyyksien muodossa. Osana tutkielmaa on toteutettu kokeellinen järjestelmä hyödyntäen uutta johtamisjärjestelmien ja simulaattorien yhteentoimivuuden standardia C2SIM:iä. Järjestelmän tarkoituksena on tukea standardin tuotantokelpoisuuden analyysia ja kyvykkyyksien havainnollistamista. Kokeellisen järjestelmän toteutuksen ja sen analyysin perusteella C2SIM todettiin kykenemättömäksi tuotantokäyttöön erityisesti puutteellisen tuen vuoksi. Ainakaan kokeellisessa järjestelmässä mukana olleet simulaattorijärjestelmät eivät tue standardia täydellisesti sen nykymuodossa, eikä käyttämämme johtamisjärjestelmä tarjonnut liitännäistä ollenkaan. Johtamisjärjestelmien ja simulaattoreiden yhteentoimivuuden tukemiseksi on täten suositeltava joko eri järjestelmille erikseen räätälöityjä ratkaisuja niiden ohjelmistokehitystyökaluja hyödyntäen tai yhdistelmäratkaisuja, jotka hyödyntävät useampaa olemassa olevaa standardia.

Single-page application (SPA) model has become a popular way of building web-applications. It makes the user experience of a website more similar to desktop-applications. This is achieved by not having to make a request to backend for each page navigation and operation. However the SPA model brings some challenges of distributed data management to basic web-applications. Managing distributed consistency is a perennial research topic in computer science. Yet this has received little attention in single-page application context. This thesis compares single-page applications to distributed databases and aims to identify techniques from them that could be used in single-page applications. The comparison begins by looking at different distributed databases and analysing which of them matches closest to the single-page application model. Then the comparison focuses on two topics. Techniques used by distributed databases for replicating data from site to another are presented and compared to techniques used in web application in communication between the server and the browser. Next topic of the thesis is to study how consistency related models like ACID-properties along with isolation levels and anomalies are realized or manifested in single-page applications. As a result of the comparison it was observed that the data transfer methods used in distributed databases and web-applications were somewhat different from each other. Distributed systems favor push model for replication and replication is automatic from application developers perspective. Web applications ofter use pull-model and implementing replication is application developers responsibility. A set of consistency anomalies that can be manifested in single-page applications where found while analysing the consistency topic. The findings give a good starting point for developing libraries that could solve some of the problems that were found.

When developing an application using a microservice architecture the application consists of multiple distributed independent and loosely coupled services. These services then communicate with each other through a network in order to form a functioning application. Benefits of developing an application as a set of independent services as opposed to a single monolithic application are numerous. The services may be developed and deployed independently, which enables, for example, the usage of different programming language for a specific service. Services designed for specific tasks are also usually relatively small in size and as such easier to develop, understand and test. The challenges of building an application utilising a microservice architecture as opposed to the traditional monolithic one include identifying suitable functionalities that can be extracted into a service. Also, end-to-end testing of the extracted functionality becomes challenging. Throughout this thesis the most important benefits and challenges of the microservice architecture are investigated with a literature review as well as in practice with a case study. During the case study a specific functionality in a largely monolithic application was transformed into a microservice. The benefits and challenges that became evident during the process are covered in the thesis.

Modern software systems often produce vast amounts of software usage data. Previous work, however, has indicated that such data is often left unutilized. This leaves a gap for methods and practices that put the data to use. The objective of this thesis is to determine and test concrete methods for utilizing software usage data and to learn what use cases and benefits can be achieved via such methods. The study consists of two interconnected parts. Firstly, a semi-structured literature review is conducted to identify methods and use cases for software usage data. Secondly, a subset of the identified methods is experimented with by conducting a case study to determine how developers and managers experience the methods. We found that there exists a wide range of methods for utilizing software usage data. Via these methods, a wide range of software development-related use cases can be fulfilled. However, in practice, apart from debugging purposes, software usage data is largely left unutilized. Furthermore, developers and managers share a positive attitude towards employing methods of utilizing software usage data. In conclusion, software usage data has a lot of potential. Besides, developers and managers are interested in putting software usage data utilization methods to use. Furthermore, the information available via these methods is difficult to replace. In other words, methods for utilizing software usage data can provide irreplaceable information that is relevant and useful for both managers and developers. Therefore, practitioners should consider introducing methods for utilizing software usage data in their development practices.