Constructing animation skeletons for quadrupeds is a complicated process, and knowing how to construct an animation skeleton for one type of quadruped does not guarantee the ability to effortlessly do so for another. This project explores how anatomy may easy the task of quadruped animation skeleton setup. Quadruped anatomy has been extensively studied and a method for animation skeleton setup based on anatomical information was explored using Autodesk Maya 2012. This method was based on the assumption that anatomical information could be incorporated into animation skeleton creation, thereby enabling the construction of an animation skeleton structure applicable to quadrupeds of different locomotion. It was discovered that this was hardly the case and the conclusion of this project has been that a better approach to animation skeletons is to construct them depending on the requirements of the project they are intended for rather than seeking to standardise how they should be set up.

Android and iOS are two of the leading smart phone and mobile device operating systems on the present market. Both of these platforms support third party application development. Cloud storage involves storing and accessing data online. The goal of this project involved helping a cloud storage company, with an existing iOS application and a less developed Android application, to update the Android application and simultaneously develop both applications. In addition to producing functionality in the applications it was the goal of this project to determine methods that would be useful in further updating and production of new functionality. While a wide range of tools exist aimed at helping cross platform development none of the frameworks examined fit the requirements of this project. As a result development was done natively on each platform. Attempts were made to translate the iOS code line by line and to simply build from an outline of the iOS functionality. During this process a number of tools that could replace each other on each device were determined. Additionally a number of the challenges in translating between operating systems were discussed. Finally while translating functionality line by line proved ineffective, it was found that noting down some aspects of the iOS code structure and some server call code could be useful for Android development when one was familiar with the iOS code.

I den här rapporten utvärderas ett antal olika collaborative filtering algoritmer för att ta reda på vilken av dess som lämpar sig bäst för att implementera åt PlanEatSmile i en PHP/MySQL-miljö. Målet är att hitta en algoritm som klarar av att skapa rekommendationer i realtid och med hög precision. Därtill är det viktigt att algoritmen ska kunna skapa rekommendationer åt nya användare (cold-start-problemet) samt att algoritmen ska skala väl när sajten växer. De algoritmer som undersöks är user-based collaborative filtering, item-based collaborative filtering och tendencies collaborative filtering. Samtliga algoritmer undersöks i två versioner. En version som utgår ifrån receptbetyget och en version som omvandlar receptbetygen till  ingrediensbetyg. Algoritmerna testas på data insamlat genom en enkät för att avgöra deras precision. För att avgöra skalbarheten testas algoritmerna på simulerad data var olika faktorer såsom antalet användare, antalet betyg och antalet recept varieras. I studien lyckas vi visa att user-based collaborative filtering och tendencies collaborative filtering uppnår den högsta precisionen. Vi visar även att tendenciesalgoritmerna presterar mycket bra med avseende på skalbarhet. Genom att använda tendencies collaborative filtering och slå ut receptbetygen på receptets enskilda ingredienser lyckas vi skapa en algoritm som både uppvisar god precision och mycket god prestanda med avseende på skalbarhet.

This report describes a thesis work in computer science that was carried out at Imagination Studios. Imagination Studios is a motion capture studio that also does animations and has its main customers in the game sector. Motion capture is a way to make games and animated videos look realistic by letting real actors perform and capture scenes that are then superimposed on the animated characters in the game or video. Imagination Studios research and development department needed a tool for their future research that has the possibility to compare captured motions in an easy way. So the task for this thesis was to create a tool that makes it possible for a user to see the differences in captured motions and also come up with a creative graphical user interface. During this thesis some game scenarios were made, and then video sketches to describe how an actor should act during these scenarios. Some shoots doing these scenarios to get some test data for the final program. The final product contains of two different programs that make it possible to compare captured motions. The main program shows positions of different markers in three different graphs representing X, Y and Z coordinates over time and has the possibility to save the data down to a csv-file. The second program shows a marker cloud of the actual motions in 3D where the user has the possibility to show one or multiple takes at the same time for easier comparison.

This report is about the project of improving an existing program in order to make it more useful. The program simulates a kind of production cell that consist of different components that are controlled by a special controller program. By analyzing the program, several aspects that could be improved were found and those were grouped into three different parts of the problem. The first one is about improving the realism and precision of the simulation. Issues such as unrealistic movement of the components and other types of actions that are different than a real production cell are addressed in this part. Detection of unwanted situations is the second part of the problem. Here unwanted movement and contact are some situations that need to be checked. The final part of the problem is about improving the visualization in the simulator. There are several ways to visualize the production cell that would involve to either use 2D images or 3D models. / Denna rapport handlar om projektet att förbättra ett redan existerande program för att göra det mer användbart. Programmet simulerar en sorts produktionscell som består av olika komponenter som kontrolleras av ett speciellt kontroller-program. Genom att analysera programmet, hittades flera aspekter att förbättra och dessa grupperades i tre grupper som utgjorde problemet. Den första delen handlar om att förbättra realismen samt precisionen på simulationen. Problem såsom orealistisk förflyttning av komponenterna och andra typer av händelser som skiljer sig från vad som händer i en riktig produktionscell kommer att tas upp i denna del. Detektion av oönskade situationer är den andra delen av problemet. Här är oönskad rörelse och kontakt några situationer som behöver kontrolleras. Den sista delen av problemet handlar om att förbättra visualiseringen i simulatorn. Det finns flera olika sätt att visualisera produktionscellen som skulle involvera att antingen använda 2D bilder eller 3D modeller.

The present work describes an algorithm to enhance selectivity of metal oxide (MOX) gas sensors. The main objective is to improve gas discrimination performance of MOX sensor using a temperature modulation combined with phase space method. Our aim is to achieve a very good gas discrimination performance based on a fragment of sensor signal, rather than using the response of sensor for entire modulation signal. The basic principle behind this thesis work is that investigating in sensor response and extracting a segment with high class separability from a full modulation cycle of sensor response. The fragment of sensor signal is obtained by variable sliding window. A segment that gives more separable classes is taken for discriminating between a set of given analytes. In this work we demonstrate the developed algorithm with a single, commercially available MOX sensor, Figaro TGS 2620, which is temperature modulated with a sinusoidal waveform.

When making a good decision, people must evaluate the advantages and disadvantages of each option, and take all the alternatives into consideration. But it’s always difficult for most of us. How to make a good decision and make a smart choice becomes more and more important in our daily life. In our thesis work, we have performed a pilot study within a research program in decision analysis regarding representation of utility difference proportion matrices, with the problem of utility aggregation in view.  We have proposed a matrix representation based on a map of maps and implemented it as a Java class. We have also defined a simple measure of inconsistency between utility difference proportions, and implemented this measure as a method in the matrix class. The function of the matrix class has been verified with some simple test programs.

Current predictions from industry envision that within a decade, the Internet will bepopulated by tens of billion of devices. Already today, smart Internet devices havesensors that provide an enormous potential for creating new applications. The chal-lenge at hand is how this information can be shared on the future Internet in order tounlock the full capability of applications to interact with the real world. Therefore,there is an urgent need for scalable and agile support for connecting people, placesand artifacts in applications via a vast number of devices and sensors on the futureInternet. Clearly, this poses a challenge of sharing and thus storage of so-called con-text information. Beyond scalable context storage lays another challenge to identifyand locate devices that are important to the user. In a support for billion of contin-uously changing sensors and actuators, a search engine would not work. Thereforean intelligent way to group devices is required. This thesis deals with mainly three issues: Firstly, propose a method for devicesto be reachable and thus addressable independent of their location in the infrastruc-ture. Secondly, how can the proposed method be used to insure automatic connectiv-ity anywhere between clients and services offered by the device, in particular associ-ated sensors and actuators. Thirdly, how can the grouping and support be combinedand used to dynamically associate sensors from across the Internet with applications,assuming that the aforementioned grouping exists. The proposed solution to the firstissue is to store identifier-locator pairs in an overlay. For the second issue we pro-pose a sensor socket introduced which exploits the identifier/locator pairs to enabledevice mobility. The third issue is addressed by providing a group-cast operation inthe sensor socket. This arrangement allows communication with peers determinedby a grouping algorithm which operates on context information on the context over-lay. Thus we have enabled the creation of automated dynamic clustering of sensorsand actuators in the Internet of Things.The sensor socket is designed as a stand-alone module to support any contextoverlay that provides the same basic functionality. The sensor socket embodiesa support to automatically interconnect and communicate with devices. Using abridging software, remote devices can be dynamically found and inserted into legacylocal area network where current devices can benefit from the connectivity. For fu-ture work the bridge can be extended to actively locate and identify nearby sensorsthat are unable to participate in the overlay network otherwise.

Wireless Mesh Networks (WMNs) have emerged as a promising network technology, which combines the benefits of cellular networks and Wireless Local Area Networks (WLANs). In a WMN mesh routers wirelessly relay traffic on behalf of other mesh routers or clients and thereby provide coverage areas comparable to cellular networks, while having the low complexity and low costs of WLANs. As Voice over IP (VoIP) is a very important Internet service, it is critical for the success of WMNs to support high quality VoIP. However, currentWMNs are not adapted well for VoIP. The capacity and scalability of single-radio WMNs is low, especially for small packet transmissions of VoIP calls, because the MAC and PHY layer overhead for small packets is high. The scalability of multiradio/multi-channel WMNs is usually higher, since fewer nodes contend for a channel. However channel scheduling might be required, which can lead to excessive delay and jitter and result in poor VoIP quality. In this thesis we investigate how to deliver high quality VoIP in single radio and multi-radio networks by using cross-layer optimization. For single radio WMNs, we consider the use of IP packet aggregation and IEEE 802.11e transmission opportunities. We conclude that IP packet aggregation greatly improves the capacity and thereby the scalability of WMNs. We show that the key for providing good quality is to artificially delay packets prior to aggregation. We propose a distributed cross-layer optimization system, which, based on Fuzzy Logic Inference, derives an aggregation delay that enhances the capacity and quality. For multi-radio/multi-channel WMNs, we demonstrate the importance of qualityof- service-aware channel scheduling. We develop a quality-of-serviceaware channel scheduler that compared to a basic round-robin scheme significantly reduces jitter and in that way increases VoIP quality. Our analysis shows that there is a trade-off between the jitter of high priority VoIP traffic and the throughput of background TCP traffic. The proposed optimizations significantly increase the capacity of singleradio and multi-radio WMNs. This allows network operators to serve more users with an existing mesh infrastructure or provide better service delivery to existing users.

This thesis presents image processing solutions in FPGA based embedded vision systems. Image processing is a demanding process but the information that can be extracted from images is very useful and can be used for many tasks like mapping and navigation, object detection and recognition, collision detection and more. Image processing or analysis involves reading images from a camera system, improve an image with respect to colour fidelity and white balance, removing distortion, extracting salient information. The mentioned steps are often referred to as low to medium level image processing and involve large amounts of data and fairly simple algorithms suitable for parallel processing. Medium to high level processing involves a reduced amount of data and more complex algorithms. Object recognition which involves matching image features to information stored in a database is of higher complexity. A vision system can be used in anything from a car to industry processes to mobile robots playing soccer or assisting people in their homes. A vision system often works with video streams that are processed to find pieces that can be handled in an industry process, detect obstacles that may be potential hazards in traffic or to find and track landmarks in the environment that can be used to build and navigate from. This involves large amount of calculations and this is a problem, even though modern computers are fast they may still not be able to execute the desired algorithms with the frequency wanted. Even if there are computers that are fast enough they are bulky and require a lot of power. They are not suitable for incorporating on small mobile robots. In this thesis I will present the image processing sequence to give an understanding of the complexity of the processes involved and I will discuss some processing platforms suitable for image processing. I will also present my work that is focused on image algorithm implementations for reconfigurable hardware suitable for mobile robots with requirements on speed an power consumption.