Ships and vessel conditions demand regular assessment to maintain their safety. In the traditional environment, their assessment was conducted using surveys and complex and time-consuming operations. But now, with the emergence of Industry 4.0 where intelligence and smart devices serve the imagery, drone-based, and many other alternative methods for inspection, the subject is obtaining considerable interest. The concept is highly effective with low cost and less disruption delivering a safer inspection approach. This study has examined Industry 4.0 technology as a quality inspection technique of a ship/vessel, examined drone-based ship inspection techniques for quality inspection of the ship/vessel without human involvement, to analyse robotic underwater surveillance methods for quality inspection of the ship/vessel, and to identify vision-based corrosion detection techniques for quality inspection of the ship/vessel. In the finding, it was revealed ship inspection through Industry 4.0 technology and other techniques can help the marine industries rely more on automated systems to gather the information that is required to be capable of authenticating process and product conformance also they can reduce human error, risks and uncover useful insights from the gathered vessel/ship data.

Vattenkraften är en viktig del av den svenska elförsörjningen och tillgodoser cirka 45 % av det årliga elbehovet i landet. Vattenkraften utnyttjar höjdskillnaden mellan inflödet och utflödet för att driva turbiner och producera el. I Sverige finns omkring 200 dammar med en fallhöjd större än 15 m och för höga nivåer vid någon av dessa skulle leda till ett dammbrott med omfattande konsekvenser. Ett viktigt område som det investeras mycket i idag är dammsäkerhet vilket går ut på att undvika okontrollerade översvämningar av dammen. Kännedom av fallhöjd och flöde är nödvändiga inom vattenkraft för att kunna beräkna producerad effekt och även vid dimensionering av dammar, vattenvägar och generatorer. Mättningar på inflödet till ett vattendrag anses vara mycket komplicerat och tidskrävande däremot kan utflödet genom utskoven beräknas med matematiska samband. Utskov avser öppningar i dammen som används för att avbörda vatten. Det finns således olika typer av utskov och det som påverkar valen är aspekter som kostnad, väderförhållanden, avbördningsförmåga, etc. För att undvika en möjlig överströmning av en damm dimensioneras ofta utskoven med avseende på vilken mängd vatten de bör avbörda. Ett krav som ofta ställs är att det ska vara möjligt att avbörda en vattendrags högsta vattennivå (HHQ). Några av de mest förekommande varianter av kontrollerade utskov i Sverige är valsdammar, luckdammar och bottenutskov.   Nivån i vattenmagasinet mäts upp med hjälp av sensorer som är placerade på olika platser vid dammen och uppmätta höga nivåer leder till att olika larmfunktioner löser ut. Regleringen av vattennivån sker då genom att avbörda vatten på ett kontrollerat sätt genom utskoven.  Genom att simulera regleringsprocessen av vattennivån i en virtuell miljö kan olika scenarier som kan uppstå vid magasinhantering testas på ett ekonomiskt sätt och hitta lämpliga reglerstrategier. Simuleringar är också en av de verktyg lyft fram inom industri 4.0 vilket kan bidra till framtidens lösningar inom automatiseringsprojekt. I arbetet studeras de olika beståndsdelar i ett vattenkraftverk och en tilltänkt metod presenteras för skapning av en virtuell modell av nivåregleringsprocessen i Siemens SIMIT, styrning av modellen via en virtuell PLC-enhet och visualisering av processen på en HMI-skärm. / Hydropower is an important part of the Swedish electricity supply and meets about 45% of the annual electricity needs in the country. Hydropower uses the height difference between inflow and outflow to power turbines and produces electricity. In Sweden, there are about 200 dams with a drop height greater than 15 m, and too high levels at any of these dams would lead to a dam break with extensive consequences. An important area in which a lot is invested today is dam safety, which is about avoiding uncontrolled flooding of the dam. Knowledge of drop height and water flow is necessary for hydropower to calculate the power produced and when dimensioning dams, waterways, and generators. Saturations on the inflow to a watercourse are very complicated and time-consuming, however, the outflow through the spillway can be calculated with mathematical correlations. Spillways refer to openings in the dam that are used to carry away water volumes. Thus, there are different types of spillways, and what influences the choices are aspects such as cost, weather conditions, etc. To avoid possible flooding of a dam, the spillways are often dimensioned with respect to the amount of water they should carry away. A requirement that is often made is that it must be able to carry away the highest water level (HHQ) of a watercourse.  The level in the water reservoir is measured with the help of sensors that are in different places by the dam and measured high levels lead to different alarm functions being triggered. The regulation of the water level then takes place by carrying away water in a controlled manner through the spillway. By simulating the regulation process of the water level in a virtual environment, different scenarios that can arise during reservoir management can be tested in an economical way and find suitable control strategies. Simulations are also one of the tools highlighted in industry 4.0, which can contribute to future solutions in automation projects. In the work, the various components of a hydropower plant are studied, and an intended method is presented for creating a virtual model of the level control process in Siemens SIMIT, controlling the model via a virtual PLC unit and visualizing the process on an HMI screen.

Currently, robotics is a discipline that is present, and it is becoming more important in daily life and different areas. Moreover, the research in this field is making improvements on the tasks that robots can perform, making it possible to appear in disciplines that have typically been made by humans, such as Art. In this project, it has been developed and implemented a program that allows the creation of paths after processing a picture, and the control of a real robot to follow the generated paths, in this case, the objective is to perform a sketch from a given picture. Nevertheless, it is applicable in many areas that need this kind of application like processing images, identification of trajectories, and path following. Moreover, in this project, it has been developed to simulate in a virtual environment the path planning and all the features of the real robot, which suppose that the user can check trajectories before trying on the real world, avoid problems of collisions or work without needing the physical robot. For that reason, the objective of this project is to contribute to the development of robotics and create a base that could be used in future research or as a source of information for similar projects that will be performed in the future.

In this work, a robust decentralized model predictive control regime for a team of cooperating robot systems is designed. Their assumed dynamics are in continuous time and non-linear. The problem involves agents whose dynamics are independent of one-another, and its solution couples their constraints as a means of capturing the cooperative behaviour required. Analytical proofs are given to show that, under the proposed control regime: (a) Subject to initial feasibility, the optimization solved at each step by each agent will always be feasible, irrespective of whether or not disturbances affect the agents. In the former case, recursive feasibility is established through successive restriction of each agent's constraints during the periodic solution to its respective optimization problem. (b) Each (sub)system can be stabilized to a desired configuration, either asymptotically when uncertainty is absent, or within a neighbourhood of it, when uncertainty is present, thus attenuating the affecting disturbance. In this context, disturbances are assumed to be additive and bounded. Simulations verify the efficacy of the proposed method over a range of different operating environments. / I detta arbete, en robust decentraliserad modell prediktiv kontroll regime förett lag av samverkande robotsystem är utformade. Deras antagnat dynamikär i kontinuerlig tid och olinjär. Problemet involverar agenter vars dynamik äroberoende av varandra, och sina lösning kopplar sina begränsningar som ettmedel för att fånga det kooperativa beteendet som krävs. Analytiska bevis gesför att visa att, enligt det föreslagna kontrollsystemet: (a) med förbehåll förförsta genomförbarhet, kommer optimeringen som löses vid varje steg av varjeagent alltid vara möjligt, oavsett huruvida störningar påverkar agenserna ellerinte. I det förre fallet är rekursiv genomförbarhet etablerad genom successivabegränsningar av varje agents inskränkning under den periodiska lösningentill dess respektive optimeringsproblem. (b) Varje (sub) system kan stabiliserastill en önskad konfiguration, antingen asymptotiskt när osäkerhet saknas,eller inom en närhet av det, när osäkerhet är närvarande, således dämparpåverkanstörning. I detta sammanhang antas störningar vara additiv och avgränsas.Simuleringar verifierar effekten av den föreslagna metoden över ettintervall av olika driftsmiljöer.

This thesis examines two methods for monocular visual odometry, FAST + KLT and ORBSLAM2, in the case of underwater environments.This is done by implementing and testing the methods on different underwater datasets. The results for the FAST + KLT provide no evidence that this method is effective in underwater settings. However, results for the ORBSLAM2 indicate that good performance is possible whenproperly tuned and provided with good camera calibration. Still, thereremain challenges related to, for example, sand bottom environments and scale estimation in monocular setups. The conclusion is therefore that the ORBSLAM2 is the most promising method of the two tested for underwater monocular visual odometry. / Denna uppsats undersöker två metoder för monokulär visuell odometri, FAST + KLT och ORBSLAM2, i det särskilda fallet av miljöer under vatten. Detta görs genom att implementera och testa metoderna på olika undervattensdataset. Resultaten för FAST + KLT ger inget stöd för att metoden skulle vara effektiv i undervattensmiljöer. Resultaten för ORBSLAM2, däremot, indikerar att denna metod kan prestera bra om den justeras på rätt sätt och får bra kamerakalibrering. Samtidigt återstår dock utmaningar relaterade till exempelvis miljöer med sandbottnar och uppskattning av skala i monokulära setups. Slutsatsen är därför att ORBSLAM2 är den mest lovande metoden av de två testade för monokulär visuell odometri under vatten.

In recent years, customers' life stylish and behaviour have been changed and their tendencies become more diverse than before. As well as, market researches show a growing trend of special demands in their inclination. Accordingly, the companies compete to meet market demands for the reason of satisfying and maintaining the loyalty of their customers. This has led researchers to seek flexible methods and optimum solutions in production processes. In this regard, the manufacturing flow is one of the places in which flexibility and optimization can cause improvement is the assembly line. So, to meet the increasing diversity of market demand, assembly line processes must have the flexibility and ability to dynamically change product specifications and features while minimizing the operating and production processes' costs. For this reason, companies use solutions for designing assembly lines based on designing a flexible manufacturing system (FMS) for fast and better managing material feeding and handling. The FMS is a manufacturing method that is developed to easily can compatible with changes in production in the type and quantity of the product that has been manufactured. The performance of an FMS depends on the precise and efficient scheduling of material handling equipment. In this regard, the use of automated guided vehicles (AGV) instead of conveyors in an assembly line provid effectiveness and transformative.The aim of this research is to obtain the optimal numbers of AGVs required for the purpose of material handling at assembly lines. To solve the problem, a mathematical model has been developed. The objective function of the model is the number of AGVs. Accordingly, a set of variables and constraints have been determined have been considered to solve the linear optimization equation. In addition, the influence of workstations’ buffer capacity on material handling activities and the number of AGVs are studied by using this developed mathematical model.In order to evaluate experiments and analysis of the problem, Python programming software and its libraries have been used to calculate rapidly and accurately equations of the model. Preliminary results show that according to the layout of the production line and demand/production planning, as well as determining the size of input and output buffers of workstations, the proposed mathematical model calculates the minimum number of required vehicles and by changing the size of buffers we achieve different results. The aforementioned IT artefact is implemented in the direction of industrial digitalization and can be developed as application software in the eventuality. And further, this master project is desirable to take performance to continue sustainable development principles. / <p>Examensarbetet är på 16,5 hp:</p><p>Master Degree Project in Virtual Product RealizationOne year Level 16.5 ECTSSpring term 2022</p>

Robots are increasingly utilized in industry for their capability to perform repetitive,complex tasks in environments unsuitable for humans. This surge in robotic applicationshas spurred research into Multi-Robot Systems (MRS), which aim to tackle complex tasksrequiring collaboration among multiple robots, thereby boosting overall efficiency. However,MRS introduces multifaceted challenges that span various domains, including robot perception,localization, task assignment, communication, and control. This dissertation delves into theintricate aspects of task assignment and path planning within MRS.The first area of focus is on multi-robot navigation, specifically addressing the limitationsinherent in current Multi-Agent Path Finding (MAPF) models. Traditional MAPF solutionstend to oversimplify, treating robots as holonomic units on grid maps. While this approachis impractical in real-world settings where robots have distinct geometries and kinematicrestrictions, it is important to note that even in its simplified form, MAPF is categorized as anNP-hard problem. The complexity inherent in MAPF becomes even more pronounced whenextending these models to non-holonomic robots, underscoring the significant computationalchallenges involved. To address these challenges, this thesis introduces a novel MAPF solverdesigned for non-holonomic, heterogeneous robots. This solver integrates the hybrid A*algorithm, accommodating kinematic constraints, with a conflict-based search (CBS) for efficientconflict resolution. A depth-first search approach in the conflict tree is utilized to accelerate theidentification of viable solutions.The second research direction explores synergizing task assignment with path-finding inMRS. While there is substantial research in both decentralized and centralized task assignmentstrategies, integrating these with path-finding remains underexplored. This dissertation evaluatesdecoupled methods for sequentially resolving task assignment and MAPF challenges. Oneproposed method combines the Hungarian algorithm and a Traveling Salesman Problem (TSP)solver for swift, albeit suboptimal, task allocation. Subsequently, robot paths are generatedindependently, under the assumption of collision-free navigation. During actual navigation, aNonlinear Model Predictive Controller (NMPC) is deployed for dynamic collision avoidance. Analternative approach seeks optimal solutions by conceptualizing task assignment as a MultipleTraveling Salesman Problem (MTSP), solved using a simulated annealing algorithm. In tandem,CBS is iteratively applied to minimize the cumulative path costs of the robots.

Rapporten presenterar ett projekt inom utveckling av ett autonomt elektroniskthjälpmedel för shoppingmiljön, där hjälpmedlet ska kunna följa en användare undervistelsen. Projektet är strukturerat i två huvudsakliga segment, där det första segmentet involverar användarens lokaliseringsprocess. Prototypen är utformad för attdetektera de största röda objekten inom dess synfält med hjälp av ett kamerabaseratsystem. Ett kamerabaserat system var ett enkelt system att implementera och fungerade bra som en sensor för att detektera objekt. Det andra segmentet fokuserar påatt följa användare på ett förutbestämt avstånd. Ett tillägg i systemet var regleringför att få en mjukare körning och en PID-regulator implementerades för att reglerahastigheten. Rapportens avslutande resultat visar på ett fullt fungerande system därsamtliga uppsatta mål har uppnåtts. I resultatdelen finns även hur väl prototypenföljer användaren under en färd på 5, 10 och 15 meter. Den genomsnittliga standardavvikelsen mellan enheten och användaren hamnade på cirka fem centimeter. Ävenen video presenteras i resultatet och visar hur väl prototypen följer användaren. Idiskussionen diskuteras utvecklingsmöjligheter i projektet för framtida arbeten. / The report presents a project in the development of an autonomous electronic assistance device for the shopping environment, where the prototype should be ableto follow an user during the stay. The project is structured in two main segments,with the first segment involving the user localization process. The prototype is designed to detect the largest red objects within its field of view using a camera-basedsystem. A camera-based system was a simple system to implement and worked wellas a sensor to detect objects. The second segment focuses on following users at apredetermined distance. An addition to the system was regulation to achieve smoother movement, and a PID controller was implemented to regulate speed. The finalresults of the report shows a fully functioning system where all set goals have beenachieved. In the results section, you will also find how well the prototype followsthe user during a journey of 5, 10 and 15 meters. The average standard deviationbetween the device and the user ended up being about five centimeters. A video isalso presented in the results, demonstrating how well the prototype follows the user.The discussion part discusses development opportunities in the project for futurework.

This compilation thesis presents two main contributions in path planning and obstacle avoidance, as well as an integration of the proposed modules with other frameworks to enable resilient robotic missions in complex environments.In general, through different types of robotic missions it is important to have a collision tolerant and reliable system, both regarding potential risks from collisions with dynamic and static obstacles, but also to secure the overall mission success.%Particularly, a common trend in the presented work is safety regarding collisions with dynamic and static obstacles, as well as reliable overall systems that are capable of executing missions. The work included in this thesis presents the risk-aware path planner D$^*_+$ that is capable of planning traversable paths for both ground and aerial robots. D$^*_+$ is developed on top of D$^*$-lite with a risk layer close to occupied space, modeling the unknown areas as a risk, and is implemented with a dynamic map to enable updates and adjustments to a changing environment. The risk layer aids in solving two common challenges with path planning for real robots: a) it creates a safety margin that gives free space between the path and obstacles so that robots with the corresponding size can follow the path, and b) it masks smaller holes in walls that occur when building maps from real data. Using a dynamic map makes it possible to use D$^*_+$ for an exploration mission, it also enables for the re-planning of the path if the environment changes for example, if an obstacle suddenly blocks a path, a new path will be planned. D$^*_+$ have been tested in different real-life experiments with both an Unmanned Areal Vehicle (UAV) and a quadruped-legged robot and shown to produce traversable paths in different application scenarios, such as exploration, return to base, and navigation on known maps. This thesis also presents an obstacle avoidance architecture for velocity objects, structured around an object detection and tracking scheme that is combined with non-linear model predictive controller (NMPC) to plan the avoidance maneuver. %that uses a Convolutional Neural Network to detect obstacles that are tracked so they can be avoided by a non-linear model predictive controller (NMPC).In this case, the detection is done with the Convoluitonal Neural Network (CNN) You Only Lock Once v4 (YOLO) where the most certain human is tracked with a Kalman filter, and the velocity of the human is estimated.The proposed scheme models the object motion as constant velocity, which is utilized from the NMPC to plan control inputs for the robot to avoid the identified obstacle. A merit of the approach is that the avoidance maneuver does not only consider the current identification position, but also considers the motion prediction of the object. This avoidance framework proved to be capable to avoid non-cooperative obstacles, such as humans moving towards it.Due to the fact that the avoidance is starting when a future collision is predicted, the avoidance maneuver is started early enough to avoid obstacles with a higher velocity than a classic ``static obstacle'' radius approach can handle. An additional aim of this thesis is to showcase that the proposed contributions can be applied in full robotic missions/frameworks. Thus, this thesis presents a search and rescue mission with a quadruped-legged robot and a UAV on a partially known map to find the location of survivors and other objects of related interest. In this mission, the quadruped-legged robot carries the UAV to the edge of the known map from where it launches the UAV that then explores and detects any survival and other relevant objects.Also, an autonomy solution, based on Boston dynamics' quadruped-legged robot Spot, for enabling a map-based navigation in confined environments has been developed and tested. This Spot solution enables the robot to navigate to a user-selected point, rotate in the desired direction, and instruct the UAV, in the combined search and rescue mission, to take off.

This project is being carried out in collaboration with Parker Hannifin. The main objective of this project is to achieve a simulation of an automatic assembly cell for the assembly of the CRVs in the P70 hydraulic valve assembly line. The robotic cell uses the line layout to perform the assembly. It consists of a robot, an automatic feeder system, a conveyor system, an assembly table, and an aligner stand. The tool used to perform the assembly is a combination of two tools mounted on a multi-tool. The first tool is a vacuum gripper, and the second tool is a drill driver. The time taken by the robot to complete one assembly of a P70 module of three sections, with a size of six CRVs on each side, A and B, is eighty-eight seconds which is faster than the speed of a manual operator. The throughput rate of the assembly cell is forty-one complete assemblies per hour, which equals one hundred and twenty-three sections per hour. This result shows clear improvement comparing it to the manual assembly throughput rate. The minimum time for the automatic feeder system to request a refill is approximately twenty-two and a half minutes. A minimal change is required when a new specification is added to the assembly line.