Cheap automatic robotic vacuum cleaners keep their cost down by having less sensors and many of them focus on using a single frontal bumper sensor, this makes it important to be able to get good coverage of a room with no knowledge of said room. This paper investigates whether the algorithm boustrophedon is enough to get a good coverage of a simple room with a maximum of two furnitures and only 90 degree corners. A graphical simulation were made to test the algorithms that are commonly used in cheap automatic robotic vacuum cleaners to compare them with the result of only using boustrophedon. The results show that the best algorithms are the non-deterministic random walk and all combined. Boustrophedon tends to get stuck when the room is not empty and it only cleans half the room when starting in the middle of the room, while being the fastest and gets most coverage in an empty room when starting in a corner. / Billiga automatiska dammsugare håller kostnaderna nere genom att minska antalet sensorer och många av dem använder endast en stötfångare med stötsensor framtill, vilket gör det viktigt att kunna få en bra täckning av ett rum utan vetskap om rummet. I denna rapport undersöks om algoritmen boustrophedon räcker för att få en bra täckning av ett enkelt rum med som mest två möbler och endast 90 graders hörn. En grafisk simulering gjordes för att testa de algoritmer som vanligtvis används i billiga automatiska dammsugare för att jämföra dem med resultatet av att endast använda boustrophedon. Resultaten visar att de bästa algoritmerna är den icke-deterministiska random walk och alla algoritmer kombinerade. Boustrophedon tenderar att fastna när rummet inte är tomt och den rengör bara hälften av rummet när man börjar i mitten av rummet, men den är snabbast och får mest täckning i ett tomt rum när man börjar i ett hörn.

This thesis is about the usage of a machine learning model for self-driving vehicles running on a small mobile computing unit, in this case, a smartphone. We use NVIDIA’s PilotNet model which is a simple feed-forward machine learning model for steering a vehicle. Their testing is conducted on a real-world vehicle and the model is used for lane keeping. Instead, we’ve adopted to be able to drive around an oval track that has no lane marking. The primary goal was to be able to run the model on a smartphone and this was a simple task as we’ve seen that the inference time is small enough that it can run at 30 FPS. As of now the model only generate a good steering output in the testing phase with prerecorded data and was only able to complete a corner on one side of the track. / Arbetet handlar om användning av maskininlärning för självkörande fordon med en liten datorenhet, i detta fallet en smartphone. Vi använder oss av NVIDIAs PilotNet model som är en simpel feed-forward modell för att styra ett fordon. I dagsläget kan modellen generera en bra styrsignal från inspelad data under simulerad testning. Vid testning på själva fordonet kan den slutföra en av fyra svängar på testbanan. PilotNet är ursprungligen utvecklat och testat för körfältshållning. Vårt testområde har ingen vägmarkering som visar vart fordonet kan svänga, vilket gav den en stor uppgift att lära sig. Vi ser att en liten enhet som en smartphone kan köra modellen enkelt och tiden mellan ingången på bilden till generering av styrsignal kan vara i 30FPS som gör att den kan köra i realtid.

Det finns en rad fördelar med RPA. Trots många fördelar med RPA, är många organisationer tveksamma till implementering av tekniken. Även de som implementerar, uppnår inte alltid de förväntade fördelarna eller i vissa fall misslyckats totalt med projektet (Wewerka & Reichert 2021). Det kan vara svårt att identifiera rätt processer som är lämpliga för att automatisera. Att implementera processautomatisering i kommunala verksamheter är inte bara utmanande i praktiken utan själva idén med processautomation som anses vara välfärdssystemets räddare bör också övervägas med viss försiktighet (Toll et al. 2021).  Mer eller mindre står alla svenska kommuner inför stora utmaningar såsom ökade utgifter och minskade intäkter samt brist på personal. Därför undersöker allt fler kommuner möjligheten att en RPA-robot tar över administrativa arbetsuppgifter för att kunna möta framtidens utmaningar. Denna studie syftar till att belysa och beskriva vilka möjligheter och utmaningar uppstått vid RPA-implementering i svenska kommuner.  Studiens frågeställning har besvarats med hjälp av en kvalitativ metod där semistrukturerade intervjuer har använts för datainsamling. En konceptuell modell som är framtaget av Santos et al. (2020) har använts som ett schema för att genomföra och analysera implementering av RPA i svenska kommuner.  Sammanfattningsvis kan det konstateras att RPA möjliggör effektivisering av verksamheterom det kan tas fram tydliga mål med RPA. Genom att låta en RPA-robot utföra arbetsuppgifter, frigörs tid för personalen. Dock saknar kommunerna strategier för hur den tiden som frigörs ska användas vilket är en utmaning som behöver prioriteras.  Studien kom fram till att processkartläggningen har varit en utmaning för kommunerna men samtidigt anses vara en nyckelfaktor för en framgångsrik RPA-implementering. Vidare haridentifierats att otydliga ansvarsområden i projektet kan leda till att projektet misslyckas. För att lyckas med RPA-implementering, har identifierats två huvudroller som ska vara med i projektteamet: en affärsanalytiker och en RPA-utvecklare.  RPA anses inte vara det bästa lösningen på automation utan annan teknik som är mer stabil förespråkas när möjlighet finns. För att göra en bedömning vilka processer är bäst lämpad att automatiseras, har kommit fram att processen behöver vara regelstyrd, standardiserad, digitalt, ha anpassad inloggningssätt för robot, undvika uppdateringar/upphandlingarna på den närmsta tiden samt ha stor volym. Det krävs ofta ett förarbete för att förbättra processen innan automatisering. RPA-projektet kan sägas att det drivs som en iterativ process vilket krävs involvering av verksamheten i hela implementeringsprocessen. Det medför att verksamheterna får bättre förståelse om RPA-utveckling och kan komma med fler förslag på processer som kan automatiseras. / There are many benefits to Robotic Process Automation (RPA). Despite the benefits, many organizations are hesitant to implement the technology. Even those who implement it, do not always attain the expected outcomes, or in some cases have failed the project altogether (Wewerka & Reichert 2021). It can be difficult to identify processes suitable for RPA automation. Implementing process automation in municipal activities is not only challenging in day-to-day operations but the very idea of process automation is considered to be the savior of the system welfare should be considered with caution (Toll et al. 2021). Many Swedish municipalities are facing foremost challenges, such as increased expenses and reduced income, let alone the lack of human resources. Consequently, more and more municipalities are exploring the possibility of an RPA robot taking over administrative tasksto meet impending challenges. This study outlines and describes the opportunities and challenges that arose during RPA implementation in Swedish municipalities. The study's question has been answered using a qualitative method where semi-structured interviews have been used for data collection. A conceptual model developed by Santos et al. (2020) has been used as a scheme to carry out and analyze the implementation of RPA in Swedish municipalities. To sum up, it can be stated that RPA enables the efficiency of the process as long as clear goals can be developed with RPA. By having an RPA robot perform certain tasks the designated human resources will be freed up. However, the municipalities lack strategies on how to best utilize the freed resources and the time gained, which is a challenge that needs to be prioritized. The study concluded that the process mapping has been a challenge for the municipalities, a factor that at the same time is considered crucial for a successful implementation of RPA. Furthermore, it has been identified that unclear duties and responsibilities in the project can lead to project failure. Two main roles have been identified as vital for a project team to succeed with RPA implementation: a business analyst and an RPA developer.  RPA is not considered the best solution for automation, and this is why other more stable technologies are backed whenever possible. The following criteria have been used to assess which processes are best suited to automation: (1) the process needs to be rule-based, (2) standardized, (3) digital, (3) have an adapted login method for robots, (4) avoid updates/procurements in a near future, and (5) have a large volume. Preliminary work is often required to improve the process before automation. The RPA project as such should run as an iterative process, and that requires the involvement of the business in the entire implementation process. This means that businesses have a better understanding of RPA development and can come up with more suggestions for the automation of certain processes.

No description available.

The advancements within robotics in recent years has increased the demand for sophisticated algorithms that can tackle the challenges associated with building robust and safe autonomous systems. The objective of 3D exploration is to enable a robot to explore an unknown environment with a high degree of accuracy while minimizing time and path length. Planners such as Receding Horizon Next Best View Planner (RH-NBVP) and Autonomous Exploration Planner (AEP) have been widely studied in static environments. However, in dynamic environments where obstacles like pedestrians or vehicles can appear, existing solutions either use dynamic motion planners for obstacle avoidance or simply use reactive behavior to avoid collisions like Dynamic Exploration Planner (DEP). This thesis examines how dynamic obstacles can be included in the planning process while performing 3D exploration to make more advantageous decisions regarding both efficiency and overall safety. Considering the uncertain dynamics in the environment is crucial for a robot to act safely, and hence be deployable in a real-world scenario. The suggested solution, Dynamic Autonomous Exploration Planner (DAEP), has been evaluated with other 3D-exploration planners as benchmarks. Experiments demonstrate that static planners struggle in dynamic environments. However, most collisions can be avoided with a simple reactive motion planner. This thesis presents an extension to the static planner AEP that considers dynamic obstacles. The proposed solution samples safer routes to a goal and takes into account historical observations of dynamic obstacles. Finally, a novel potential volumetric information gain is implemented to predict explored volume in the future. The results demonstrate that the extensions to AEP enhance safety planning and improve coverage compared to regular AEP and DEP.

Within Industry 4.0, the concept of digital twins gains significant attention globally as organizations explore their potential in various industries worldwide. International research and development efforts focus on leveraging digital twins to monitor and optimize processes and products' behavior and performance. When applied to automation systems, digital twins primarily support virtual commissioning, enabling the testing of virtual environments before actual processes become operational. However, a research gap exists concerning the utilization of digital twins in maintenance operations post-commissioning. This thesis aims to fill this gap and contribute to the knowledge of how the Real Digital Twin (RDT) can be used to maintain the automation process at Ringhals nuclear power plant. The research methodology involves interviewing maintenance workers to gain insight into their Way-of-Working (WoW) and identify key challenges and requirements. Based on these interviews, various maintenance cases are constructed and subjected to Failure Mode and Effects Analysis (FMEA), a proven method for identifying and evaluating potential failures. The cases are then compared between using and not using RDT to assess its impact on maintenance operations. Emulation is also employed to practically demonstrate the implementation of RDT. The findings highlight the significant benefits of RDT in maintenance operations at Ringhals. RDT enables swift and efficient testing of different operating strategies, leading to the identification of energy-efficient solutions and a reduction in downtime. The comparison of FMEA results demonstrates that RDT enhances the ability to proactively identify and address potential failures, reducing the likelihood of disruptions and ensuring safe and efficient operations. Overall, this thesis showcases the potential of RDT in maintenance operations beyond virtual commissioning, filling the research gap. Integrating RDT into the maintenance workers' Way-of-Working revolutionizes maintenance practices, enhancing efficiency, minimizing downtime, and improving overall operational performance. These findings contribute to the evolving field of Industry 4.0 and digital twin technologies, offering valuable insights for organizations seeking to optimize their maintenance processes and leverage the full potential of digital twins.

No description available.

No description available.

No description available.

No description available.