Groups of starlings can form impressive shapes as they travel northward together in the springtime. This is among a group of natural phenomena based on swarm behaviour. The research field of artificial intelligence in computer science, particularly the areas of robotics and image processing, has in recent decades given increasing attention to the underlying structures. The behaviour of these intelligent swarms has opened new approaches for face detection as well. G. Beni and J. Wang coined the term “swarm intelligence” to describe this type of group behaviour. In this context, intelligence describes the ability to solve complex problems. The objective of this project is to automatically find exactly one face on a photo or video material by means of swarm intelligence. The process developed for this purpose consists of a combination of various known structures, which are then adapted to the task of face detection. To illustrate the result, a 3D hat shape is placed on top of the face using an example application program.:1 Introduction 1.1 Face Detection 1.2 Swarm Intelligence and Particle Swarm Optimisation Fundamentals 3 Face Detection by Means of Particle Swarm Optimisation 3.1 Swarms and Particles 3.2 Behaviour Patterns 3.2.1 Opportunism 3.2.2 Avoidance 3.2.3 Other Behaviour Patterns 3.3 Stop Criterion 3.4 Calculation of the Solution 3.5 Example Application 4 Summary and Outlook

Particle Swam Optimization (PSO) is especially useful for rapid optimization of problems involving multiple objectives and constraints in dynamic environments. It regularly and substantially outperforms other algorithms in benchmark tests. This paper describes research leading to the application of PSO to the autonomous asset management problem in electronic warfare. The PSO speed provides fast optimization of frequency allocations for receivers and jammers in highly complex and dynamic environments. The key contribution is the simultaneous optimization of the frequency allocations, signal priority, signal strength, and the spatial locations of the assets. The fitness function takes into account the assets' locations in 2 dimensions, maximizing their spatial distribution while maintaining allocations based on signal priority and power. The fast speed of the optimization enables rapid responses to changing conditions in these complex signal environments, which can have real-time battlefield impact. Results optimizing receiver frequencies and locations in 2 dimensions have been successful. Current run-times are between 450ms (3 receivers, 30 transmitters) and 1100ms (7 receivers, 50 transmitters) on a single-threaded x86 based PC. Run-times can be substantially decreased by an order of magnitude when smaller swarm populations and smart swarm termination methods are used, however a trade off exists between run-time and repeatability of solutions. The results of the research on the PSO parameters and fitness function for this problem are demonstrated.

Developing models of the cardiovascular system that simulates the dynamic behavior of a virtual patient’s condition is fundamental in the medical domain for predictive outcome and hypothesis generation. These models are usually described through Ordinary Differential Equation (ODE). To obtain a patient-specific representative model, it is crucial to have an accurate and rapid estimate of the hemodynamic model parameters. Moreover, when adequate model parameters are found, the resulting time series of state variables can be clinically used for predicting the response to treatments and for non-invasive monitoring. In the Thesis, we address the parameter estimation or inverse modeling, by solving an optimization problem, which aims at minimizing the error between the model output and the target data. In our case, the target data are a set of user-defined state variables, descriptive of a hospitalized specific patient and obtained from time-averaged state variables. The Thesis proposes a comparison of both state-of-the-art and novel methods for the estimation of the underlying model parameters of a cardiovascular simulator Aplysia. All the proposed algorithms are selected and implemented considering the constraints deriving from the interaction with Aplysia. In particular, given the inaccessibility of the ODE, we selected gradient-free methods, which do not need to estimate numerically the derivatives. Furthermore, we aim at having a small number of iterations and objective function calls, since these importantly impact the speed of the estimation procedure, and thus the applicability of the knowledge gained through the parameters at the bedside. Moreover, the Thesis addresses the most common problems encountered in the inverse modeling, among which are the non-convexity of the objective function and the identifiability problem. To assist in resolving the latter issue an identifiability analysis is proposed, after which the unidentifiable parameters are excluded. The selected methods are validated using heart failure data, representative of different pathologies commonly encountered in Intensive Care Unit (ICU) patients. The results show that the gradient-free global algorithms Enhanced Scatter Search and Particle Swarm estimate the parameters accurately at the price of a high number of function evaluations and CPU time. As such, they are not suitable for bedside applications. Besides, the local algorithms are not suitable to find an accurate solution given their dependency on the initial guess. To solve this problem, we propose two methods: the hybrid, and the prior-knowledge algorithms. These methods, by including prior domain knowledge, can find a good solution, escaping the basin of attraction of local minima and producing clinically significant parameters in a few minutes. / Utveckling av modeller av det kardiovaskulära systemet som simulerar det dynamiska beteendet hos en virtuell patients är grundläggande inom det medicinska området för att kunna förutsäga resultat och generera hypoteser. Dessa modeller beskrivs vanligtvis genom Ordinary Differential Equation (ODE). För att erhålla en patientspecifik representativ modell är det viktigt att ha en exakt och snabb uppskattning av de hemodynamiska modellparametrarna. När adekvata modellparametrar har hittats kan de resulterande tidsserierna av tillståndsvariabler dessutom användas kliniskt för att förutsäga svaret på behandlingar och för icke-invasiv övervakning. I avhandlingen behandlar vi parameteruppskattning eller invers modellering genom att lösa ett optimeringsproblem som syftar till att minimera följande felet mellan modellens utdata och måldata. I vårt fall är måldata en uppsättning användardefinierade tillståndsvariabler som beskriver en specifik patient som är inlagd på sjukhus och som erhålls från tidsgenomsnittliga tillståndsvariabler. I avhandlingen föreslås en jämförelse av befintlinga och nya metoder. för uppskattning av de underliggande modellparametrarna i en kardiovaskulär simulator, Aplysia. Alla föreslagna algoritmer är valts och implementerade med hänsyn tagna till de begränsningar som finnis i simulatorn Aplysia. Med tanke på att ODE är otillgänglig har vi valt gradientfria metoder som inte behöver uppskatta derivatorna numeriskt. Dessutom strävar vi efter att ha få interationer och funktionsanrop eftersom dessa påverkar hastigheten på estimeringen och därmed den kliniska användbartheten vid patientbehandling. Avhandlingen behandlas dessutom de vanligaste problemen vid inversmodellering som icke-konvexitet och identifierbarhetsproblem. För att lösa det sistnämnda problemet föreslås en identifierbarhetsanalys varefter de icke-identifierbara parametrarna utesluts. De valda metoderna valideras med hjälp av data om hjärtsvikt som är representativa för olika patologier som ofta förekommer hos Intensive Care Unit (ICU)-patienter. Resultaten visar att de gradientfria globala algoritmerna Enhanced Scatter Search och Particle Swarm uppskattar parametrarna korrekt till priset av ett stort antal funktionsutvärderingar och processortid. De är därför inte lämpliga för tillämpningar vid sängkanten. Dessutom är de lokala algoritmerna inte lämpliga för att hitta en exakt lösning eftersom de är beroende av den ursprungliga gissningen. För att lösa detta problem föreslår vi två metoder: hybridalgoritmer och algoritmer med förhandsinformation. Genom att inkludera tidigare domänkunskap kan dessa metoder hitta en bra lösning som undviker de lokala minimernas attraktionsområde och producerar kliniskt betydelsefulla parametrar på några minuter.

The use of unmanned aerial vehicles (UAVs) is well established in the military sector with great advantages in modern warfare. The concept of using UAV swarms has been discussed over two decades, but is now seeing its first real system used by the Israel defence forces. There is no exact definition what a swarm is, but it is proposed that it should satisfy the following three requirements. A swarm should have limited human control, the number of agents in a swarm should be at least three and the agents in the swarm should cooperate to perform common tasks. The complexity of controlling multiple autonomous UAVs gives way to the problem of how to take advantage of the operators cognitive and tactical abilities to control a swarm to effectively conduct military reconnaissance missions. The method of using behaviour trees as a control structure was derived from previous work in swarm systems. A behaviour tree is a structured way of organising and prioritising actions of autonomous systems. Behaviour trees are similar to finite state machines (FSMs) with the advantages of being modular, reactive, and with better readability. Three different behaviour trees with increasing complexity was created and simulated in the game engine Unity. A fourth more real life behaviour tree was created and used as a basis for discussing the strength and weaknesses of using behaviour trees against previous work. Using behaviour tree as a unifying structure for creating a swarm that integrates the tactical abilities of an operator with the strength of an autonomous swarm seems promising. The proposed method of using behaviour trees i suggested to be used as a platform for discussing the swarm desired functions and to create a common vision for both operators and engineers how a swarm should function. / Användning av drönare är väletablerad inom det militära och ger stora fördelar i dagens moderna krigsföring. Idén om att använda en svärm av drönare har diskuterats under de senaste två decennierna, men först nu sett sin första riktiga system som använts av Israels försvarsmakt. Det finns ingen exakt definition av vad en svärm är, men det har föreslagits att en svärm ska uppfylla de följande tre kraven. En svärm ska ha begränsad mänsklig interaktion, antalet agenter ska vara minst tre och svärmen ska samarbeta för att lösa gemensamma uppgifter. Svårigheterna med att styra en autonom svärm ger upphov till hur man ska utnyttja en operatörs kognitiva och taktiska förmåga för att styra en autonom drönarsvärm för att effektivt utföra militära spaningsuppdrag. Utifrån tidigare arbete inom styrning av svärmar verkade beteende träd som en lovande metod. Beteendeträd är ett strukturerat sätt att organisera och prioritera beteenden för ett autonomt system. Beteendeträd har många likheter med ändliga tillståndsmaskiner, men fördelarna att vara modulära, responsiva och mer lättläsliga. Tre olika beteendeträd med ökande komplexitet skapades och deras funktionalitet simulerades i Unity. Ett fjärde mer verklighetstroget beteendeträd skapades och användes som underlag för att diskutera beteendeträds styrkor och svagheter i jämförelse med tidigare arbeten. Användningen av beteendeträd för att förena den mänskliga operatören med det autonoma systemet verkar lovande. Den föreslagna metoden att använda beteendeträd för att styra en svärm är tänkt att användas som ett gemensamt underlag för att möjliggöra att operatörer och ingenjörer kan ha en gemensam bild hur en svärm ska fungera.

Beim Designprozess elektromagnetischer Aktoren ist die zuverlässige Bestimmung der zu erwartenden Verluste von großer Bedeutung. Während ohmsche Verluste sehr einfach bestimmt werden können, stellen Eisen-/Hystereseverluste häufig einen Unsicherheitsfaktor dar. Hier sind Herstellerangaben meist nur für einige wenige Arbeitspunkte bei harmonischem Betrieb vorhanden. Für den Einsatz in numerischen Berechnungen bei der Auslegung und Simulation solcher Aktoren ist eine detaillierte Beschreibung der ferromagnetischen Hysterese notwendig. Zu diesem Zweck werden häufig das Jiles-Atherton-Hysteresemodell und dessen Weiterentwicklungen eingesetzt. Aufgrund der Vielzahl an verfügbaren modifizierten Varianten wurde im Rahmen dieses Beitrages zunächst untersucht, welche Modellversionen zueinander kompatibel sind. So wird die Verwendung statischer und dynamischer Hysteresemodelle sowie die jeweilig dazu passende inverse Modellform bei konsistenter Parametrierung ermöglicht. Weiterhin wird die Parameteridentifikation anhand experimentell ermittelter Hysteresekurven für verschiedene Werkstoffe mit Hilfe der Particle-Swarm-Optimization vorgestellt. / The reliable determination of the expected losses is important for the design process of electromagnetic actuators. While resistive losses can be determined very easily, iron/hysteresis losses often represent an uncertainty factor. Manufacturer’s specifications are usually only available for a few operating points with harmonic excitation. A detailed description of the ferromagnetic hysteresis is necessary for the use in numerical calculations in the design and simulation of such actuators. For this purpose, the Jiles-Atherton hysteresis model and its further developments are often used. Due to the large number of available modified variants, at first an examination on which model versions are compatible with each other has been performed. This allows the use of static and dynamic hysteresis models as well as the corresponding inverse model form with consistent parameterization. Furthermore, the parameter identification based on experimentally determined hysteresis curves for different materials is presented using particle swarm optimization.

Optimising task allocation in networked systems helps in utilising available resources. When working with unstable and heterogeneous networks, task scheduling can be used to optimise task completion time, energy efficiency and system reliability. The dynamic nature of networks also means that the optimal schedule is subject to change over time. The heterogeneity and variability in network design also complicate the translation of setups from one network to another. Discrete Particle Swarm Optimisation (DPSO) is a metaheuristic that can be used to find solutions to task scheduling. This thesis will explore how DPSO can be used to optimise job scheduling in an unstable network. The purpose is to find solutions for networks like the ones used on trains. This in turn is done to facilitate trajectory planning calculations. Through the use of an artificial neural network, we estimate job scheduling costs. These costs are then used by our DPSO meta heuristic to explore a solution space of potential scheduling. The results focus on the optimisation of batch sizes in relation to network reliability and latency. We simulate a series of unstable and heterogeneous networks and compare completion time. The baseline comparison is the case where scheduling is done by evenly distributing jobs at fixed sizes. The performance of the different approaches is then analysed with regards to usability in real-life scenarios on vehicles. Our results show a noticeable increase in performance within a wide range of network set-ups. This is at the cost of long search times for the DPSO algorithm. We conclude that under the right circumstances, the method can be used to significantly speed up distributed calculations at the cost of requiring significant ahead of time calculations. We recommend future explorations into DPSO starting states to speed up convergence as well as benchmarks of real-life performance. / Optimering av arbetsfördelning i nätverk kan öka användandet av tillgängliga resurser. I instabila heterogena nätverk kan schemaläggning användas för att optimera beräkningstid, energieffektivitet och systemstabilitet. Då nätverk består av sammankopplade resurser innebär det också att vad som är ett optimalt schema kan komma att ändras över tid. Bredden av nätverkskonfigurationer gör också att det kan vara svårt att överföra och applicera ett schema från en konfiguration till en annan. Diskret Particle Swarm Optimisation (DPSO) är en meta heuristisk metod som kan användas för att ta fram lösningar till schemaläggningsproblem. Den här uppsatsen kommer utforska hur DPSO kan användas för att optimera schemaläggning för instabila nätverk. Syftet är att hitta en lösning för nätverk under liknande begränsningar som de som återfinns på tåg. Detta för att i sin tur facilitera planerandet av optimala banor. Genom användandet av ett artificiellt neuralt nätverk (ANN) uppskattar vi schemaläggningskostnaden. Denna kostnad används sedan av DPSO heuristiken för att utforska en lösningsrymd med potentiella scheman. Våra resultat fokuserar på optimeringen av grupperingsstorleken av distribuerade problem i relation till robusthet och letens. Vi simulerar ett flertal instabila och heterogena nätverk och jämför deras prestanda. Utgångspunkten för jämförelsen är schemaläggning där uppgifter distribueras jämnt i bestämda gruperingsstorlekar. Prestandan analyseras sedan i relation till användbarheten i verkliga scenarion. Våra resultat visar på en signifikant ökning i prestanda inom ett brett spann av nätverkskonfigurationer. Det här är på bekostnad av långa söktider för DPSO algoritmen. Vår slutsats är att under rätt förutsättningar kan metoden användas för att snabba upp distribuerade beräkningar förutsatt att beräkningarna för schemaläggningen görs i förväg. Vi rekommenderar vidare utforskande av DPSO algoritmens parametrar för att snabba upp konvergens, samt undersökande av algoritmens prestanda i verkliga miljöer.

No description available.

Hydropower systems often contain complex river systems which cause the simulations and analyses of a hydropower operation to be computationally heavy. The complex river system is referred to as something called a Detailed model. By creating a simpler model, denoted the Equivalent model, the computational issue could be circumvented. The purpose of this Equivalent model is to emulate the results of the Detailed model. This thesis computes the Equivalent model for a large hydropower system using Particle Swarm Optimisation- algorithm, then evaluates the Equivalent model performance. Simulations are performed on ten rivers in Sweden, representing four trading areas for one year, October 2017 – September 2018. Furthermore, the year is divided into Quarterly and Seasonal periods, to investigate whether the Equivalent model changes over time. The Equivalent model performance is evaluated based on the relative power difference and computational time compared to the Detailed model. The relative power difference is 4%23% between Equivalent and Detailed models, depending on the period and trading area, with the computational time can be reduced by more than 90%. Furthermore, the Equivalent model changes over time, suggesting that when the year is divided appropriately, the Equivalent model could perform better. The relative power difference results indicate that the Equivalent model performance can still be improved by dividing the periods more appropriately, other than Quarterly or Seasonal. Nevertheless, the results provide a satisfactory Equivalent model, based on the faster computation time and a reasonable relative power difference. Finally, the Equivalent model could be used as a foundation for further analyses and simulations. / Vattenkraftsystem består ofta av komplexa älvsystem som gör att simuleringar och analyser av vattenkraftens operation blir beräkningsmässigt tunga. Det komplexa älvsystem kallas en Detaljeraded modell. Genom att skapa en enklare modell, betecknas som en Ekvivalent modell, beräkningsproblemen kan kringgås. Syftet med denna Ekvivalenta modell är att emulera resultaten av den komplexa Detaljerade modellen. Detta examensarbete beräknar den Ekvivalenta modellen för ett stort vattenkraftssystem med hjälp av Particle Swarm Optimisation- algorithmen, och utvärderar modellprestandan hos Ekvivalenten. Simuleringar utförs på tio älvar i Sverige, som representerar fyra handelsområden under ett år, från oktober 2017 september 2018. Dessutom är året uppdelat i kvartals- och säsongsperioder för att undersöka om den Ekvivalenta modellen förändras över tid. Denna Ekvivalenta modell utvärderas baserat på den relativa effektskillnaden och beräkningstiden jämfört med den Detaljerade modellen. Den relativa effektskillnaden är 4% 23% mellan de Ekvivalenta och Detaljerade modellerna, beroende på period och handelsområde, och beräkningstiden minskas med mer än 90%. Vidare ändras Ekvivalenta modellen över tiden, vilket tyder på att när året delas upp på rätt sätt kan den Ekvivalenta modellen prestera ännu bättre. De relativa effektskillnaderna indikerar att vissa perioder fortfarande kan förbättras genom att dela upp perioden mer korrekt. Trots allt, förser resultanten en tillfredsställande Ekvivalent modell som har en mer effektiv beräkningstid och rimliga effektskillnader. Slutligen skulle den Ekvivalenta modellen kunna användas som en grund för ytterligare analyser och simuleringar.

This research is focused on the integration problem of process planning and scheduling in steel heat treatment operations environment using artificial intelligent techniques that are capable of dealing with such problems. This work addresses the issues involved in developing a suitable methodology for scheduling heat treatment operations of steel. Several intelligent algorithms have been developed for these propose namely, Genetic Algorithm (GA), Sexual Genetic Algorithm (SGA), Genetic Algorithm with Chromosome differentiation (GACD), Age Genetic Algorithm (AGA), and Mimetic Genetic Algorithm (MGA). These algorithms have been employed to develop an efficient intelligent algorithm using Algorithm Portfolio methodology. After that all the algorithms have been tested on two types of scheduling benchmarks. To apply these algorithms on heat treatment scheduling, a furnace model is developed for optimisation proposes. Furthermore, a system that is capable of selecting the optimal heat treatment regime is developed so the required metal properties can be achieved with the least energy consumption and the shortest time using Neuro-Fuzzy (NF) and Particle Swarm Optimisation (PSO) methodologies. Based on this system, PSO is used to optimise the heat treatment process by selecting different heat treatment conditions. The selected conditions are evaluated so the best selection can be identified. This work addresses the issues involved in developing a suitable methodology for developing an NF system and PSO for mechanical properties of the steel. Using the optimisers, furnace model and heat treatment system model, the intelligent system model is developed and implemented successfully. The results of this system were exciting and the optimisers were working correctly.

1. Antarctic krill (Euphausia superba) form aggregations known as swarms that vary greatly in size and density. Six acoustic surveys were conducted as part of multidisciplinary studies at two study sites, the western and eastern core boxes (WCB and ECB), during the 1997, 1998 and 1999 austral summers, at South Georgia. A quantitative, automated, image processing algorithm was used to identify swarms, and calculate swarm descriptors, or metrics. In contrast to acoustic surveys of aggregations of other pelagic species, a strong correlation (r = 0.88, p = 0.02, 95% C.I.= 0.24 to 0.99) between the number of krill swarms and the mean areal krill density [rho.hat] was found. Multivariate analysis was used to partition swarms into three types, based on contrasting morphological and internal krill density parameters. Swarm types were distributed differently between inter-surveys and between on and off-shelf regions. This swarm type variation has implications for krill predators, by causing spatial heterogeneity in swarm detectability, suggesting that for optimal foraging to occur, predators must engage in some sort of adaptive foraging strategy. 2. Krill predator-prey interactions were found to occur at multiple spatial and temporal scales, in a nested, or hierarchical structure. At the largest inter-survey scale, an index of variability, I, was developed to compare variation in survey-scale predator sightings, sea temperature and [rho.hat]. Using I and a two-way ANOVA, core box, rather than year, was found to be a more important factor in determining species distribution. The absence of Blue-petrels (Halobaena caerulea) and the elevated number of Antarctic fur seals (Arctocephalus gazella) suggest that 1998 was a characterised by colder than average water surrounding South Georgia, and a high [rho.hat] in the ECB. At the smaller, intra-survey scales (<80 km, <5 day), the characteristic scale (distances in which predator group size, or krill density were similar, L_s) were determined. For krill and predators L_s varied by survey and the L_s of krill also varied by depth within a survey. Overlap in L_s were stronger between predator species than between a predator species and krill, indicating predators were taking foraging cues from the activity of predators, rather than from the underlying krill distribution. No relationship was found between swarm characteristics and predator activity, suggesting either there is no relationship between krill swarms and predators, or that the predator and acoustic observation techniques may not be appropriate to detect such a relationship. 3. To overcome the 2-D sampling limitations of conventional echosounders, a multibeam echosounder (MBE) observed entire swarms in three-dimensions. Swarms found in the nearshore environment of Livingston Island situated in the South Shetland Islands, exhibited only a narrow range of surface area to volume ratios or roughnesses (R = 3.3, CV = 0.23), suggesting that krill adopt a consistent group behaviour to maintain swarm shape. Generalized additive models (GAM) suggested that the presence of air-breathing predators influenced the shape of a krill swarm (R decreased in the presence of predators: the swarm became more spherical). A 2D distance sampling framework was used to estimate the abundance, N, and associated variance of krill swarms. This technique took into account angular and range detectability (half-normal, [sigma_r.hat] = 365.00 m, CV = 0.16) and determined the vertical distribution of krill swarms to be best approximated by a beta-distribution ([alpha.hat] = 2.62, [CV.hat] = 0.19; [beta.hat] = 2.41, [CV.hat] = 0.15), giving the abundance of swarms in survey region as [N.hat] = 5,062 ([CV.hat] = 0.35). This research represents a substantial contribution to developing estimation of pelagic biomass using MBEs. 4. When using a single- or split-beam missing pings occur when the transmit or receive cycles are interrupted, often by aeration of the water column, under the echosounder transducer during rough weather. A thin-plate regression spline based approach was used to model the missing krill data, with knots chosen using a branch and bound algorithm. This method performs well for acoustic observations of krill swarms where data are tightly clustered and change rapidly. For these data the technique outperformed the standard MGCV GAM, and the technique is applicable for estimating acoustically derived biomass from line transect surveys.