Retinal ganglion cells (RGCs) serve as a vital connection between the eye and the brain with damage to their axons resulting in loss of vision and/or blindness. Retinal organoids are three-dimensional structures derived from human pluripotent stem cells (hPSCs) which recapitulate the spatial and temporal differentiation of the retina, providing a valuable model of RGC development in vitro. The working hypothesis of these studies is that hPSC-derived RGCs are capable of extensive outgrowth and display target specificity and pathfinding abilities. Initial efforts focused on characterizing RGC differentiation throughout early stages of organoid development, with a clearly defined RGC layer developing in a temporally-appropriate manner expressing a compliment of RGC-associated markers. Beyond studies of RGC development, retinal organoids may also prove useful to investigate and model the extensive axonal outgrowth necessary to reach post-synaptic targets. As such, additional efforts aimed to elucidate factors promoting axonal outgrowth. Results demonstrated significant enhancement of axonal outgrowth through modulation of both substrate composition and growth factor signaling. Furthermore, RGCs possessed guidance receptors that are essential in influencing outgrowth and pathfinding. Subsequently, to determine target specificity, aggregates of hPSC-derived RGCs were co-cultured with explants of mouse lateral geniculate nucleus (LGN), the primary post-synaptic target of RGCs. Axonal outgrowth was enhanced in the presence of LGN, and RGCs displayed recognition of appropriate targets, with the longest neurites projecting towards LGN explants compared to control explants or RGCs grown alone. Generated from the fusion of regionally-patterned organoids, assembloids model projections between distinct regions of the nervous system. Therefore, final efforts of these studies focused upon the generation of retinocortical assembloids in order to model the long-distance outgrowth characteristic of RGCs. RGCs displayed extensive axonal outgrowth into cortical organoids, with the ability to respond to environmental cues. Collectively, these results establish retinal organoids as a valuable tool for studies of RGC development, and demonstrate the utility of organoid-derived RGCs as an effective platform to study factors influencing outgrowth as well as modeling long-distance projections and pathfinding abilities.

Over the last years researchers proposed numerous indoor localisation and navigation systems. However, solutions that use WiFi or Radio Frequency Identification require infrastructure to be deployed in the navigation area and infrastructureless techniques, e.g. the ones based on mobile cell ID or dead reckoning suffer from large accuracy errors. In this Thesis, we present a novel approach of infrastructure-less indoor navigation system based on computer vision Structure from Motion techniques. We implemented a prototype localisation and navigation system which can build a navigation map using area photos as input and accurately locate a user in the map. In our client-server architecture based system, a client is a mobile application, which allows a user to locate her or his position by simply taking a photo. The server handles map creation, localisation queries and pathnding. After the implementation, we evaluated the localisation accuracy and latency of the system by benchmarking navigation queries and the model creation algorithm. The system is capable of successfully navigating in Aalto University computer science department library. We were able to achieve an average error of 0.26 metres for successfully localised photos. In the Thesis, we also present challenges that we solved to adapt computer vision techniques for localisation purposes. Finally we observe the possible future work topics to adapt the system to a wide use. / Forskare har de senaste åren framfört olika inomhusnavigations- och lokaliseringssystem. Dock kräver lösningar som använder WiFi eller radiofrekvens identifikation en utbyggdnad av stödjande infrastruktur i navigationsområdena. Även teknikerna som används lider av precisionsfel. I det här examensarbetet redovisar vi en ny taktik för inomhusnavigation som använder sig av datorvisualiserings Structure from Motion-tekniker. Vi implementerade en navigationssystemsprototyp som använder bilder för att bygga en navigationskarta och kartlägga användarens position. I vårt klient-server baserat system är en klient en mobilapplikation som tillåter användaren att hitta sin position genom att ta en bild. På server-sidan hanteras kartor, lokaliseringsförfrågor och mättningar av förfrågorna och algoritmerna som används. Systemet har lyckats navigera genom Aalto Universitets datorvetenskapsbiblioteket. Vi lyckades uppnå en felmarginal pa 0.26 meter för lyckade lokaliseringsbilder. I arbetet redovisar vi utmaningar som vi har löst för att anpassa datorvisualiseringstekniker for lokalisering. Vi har även diskuterat potentialla framtida implementationer for att utvidga systemet.

Traditional shortest path algorithms can efficiently find the optimal paths in graphs using simple heuristics. However, formulating a simple heuristic is challenging under the road network setting since there are multiple factors to consider, such as road segment length, edge centrality, and speed limit. This study investigates how a neural network can learn to take these factors as inputs and yield a path given a pair of origin and destination. The research question is formulated as: Are neural networks applicable to real-time route planning tasks in a roadnetwork?. The proposed metric to evaluate the effectiveness of the neural network is arrival rate. The quality of generated paths is evaluated by time efficiency. The real-time performance of the model is also compared between pathfinding in dynamic and static graphs, using theabove metrics. A staggered approach is applied in progressing this investigation. The first step is to generate random graphs, which allows us to monitor the size and properties of the training graph without caring too many details in a road network. The next step is to determine, as a proof of concept, if a neural network can learn to traverse simple graphs with multiple strategies, given that road networks are in effect complex graphs. Finally, we scale up by including factors that might affect the pathfinding in real road networks. Overall, the training data is optimal paths in a graph generated by a shortest path algorithm. The model is then applied to new graphs to generate a path given a pair of origin and destination. The arrival rate and time efficiency are calculated and compared with that of the corresponding optimal path. Experimental results show that the effectiveness, i.e., arrival rate ofthe model is 90% and the path quality, i.e., time efficiency has a medianof 0.88 and a large variance. The experiment shows that the model has better performance in dynamic graphs than in static graphs. Overall, the answer to the research question is positive. However, there is still room to improve the effectiveness of the model and the paths generated by the model. This work shows that a neural network trained to make locally optimal choices can hardly give a globally optimal solution. We also show that our method, only making locally optimal choices, can adapt to dynamic graphs with little performance overhead. / Traditionella algoritmer för att hitta den kortaste vägen kan effektivt hitta de optimala vägarna i grafer med enkel heuristik. Att formulera en enkel heuristik är dock utmanande för vägnätverk eftersom det finns flera faktorer att överväga, såsom vägsegmentlängd, kantcentralitet och hastighetsbegränsningar. Denna studie undersöker hur ett neuralt nätverk kan lära sig att ta dessa faktorer som indata och finna en väg utifrån start- och slutpunkt. Forskningsfrågan är formulerad som: Är neuronnätverket tillämpliga på realtidsplaneringsuppgifter i ett vägnät?. Det föreslagna måttet för att utvärdera effektiviteten hos det neuronnätverket är ankomstgrad. Kvaliteten på genererade vägar utvärderas av tidseffektivitet. Prestandan hos modellen jämförs också mellan sökningen i dynamiska och statiska grafer, med hjälp av ovanstående mätvärden. Undersökningen bedrivs i flera steg. Det första steget är att generera slumpmässiga grafer, vilket gör det möjligt för oss att övervaka träningsdiagrammets storlek och egenskaper utan att ta hand om för många detaljer i ett vägnät. Nästa steg är att, som ett bevis på konceptet, undersöka om ett neuronnätverk kan lära sig att korsa enkla grafer med flera strategier, eftersom vägnätverk är i praktiken komplexa grafer. Slutligen skalas studien upp genom att inkludera faktorer som kan påverka sökningen i riktiga vägnät. Träningsdata utgörs av optimala vägar i en graf som genereras av en algoritm för att finna den kortaste vägen. Modellen appliceras sedan i nya grafer för att hitta en väg mellan start och slutpunkt. Ankomstgrad och tidseffektivitet beräknas och jämförs med den motsvarande optimala sökvägen. De experimentella resultaten visar att effektiviteten, dvs ankomstgraden av modellen är 90% och vägkvaliteten dvs tidseffektiviteten har en median på 0,88 och en stor varians. Experimentet visar att modellen har bättre prestanda i dynamiska grafer än i statiska grafer. Sammantaget är svaret på forskningsfrågan positivt. Det finns dock fortfarande utrymme att förbättra modellens effektivitet och de vägar som genereras av modellen. Detta arbete visar att ett neuronnätverk tränat för att göra lokalt optimala val knappast kan ge globalt optimal lösning. Vi visar också att vår metod, som bara gör lokalt optimala val, kan anpassa sig till dynamiska grafer med begränsad prestandaförlust.

Un système multi-agents est composé de plusieurs agents autonomes qui interagissent entre eux dans un environnement commun. Ce mémoire vise à démontrer l’utilisation d’un système multi-agents pour le développement d’un jeu vidéo. Tout d’abord, une justification du choix des concepts d’intelligence artificielle choisie est exposée. Par la suite, une approche pratique est utilisée en effectuant le développement d’un jeu vidéo. Pour ce faire, le jeu fut développé à partir d’un jeu vidéo mono-agent existant et mo- difié en système multi-agents afin de bien mettre en valeur les avantages d’un système multi-agents dans un jeu vidéo. Le développement de ce jeu a aussi démontré l’applica- tion d’autres concepts en intelligence artificielle comme la recherche de chemins et les arbres de décisions. Le jeu développé pour ce mémoire viens appuyer les conclusions des différentes recherches démontrant que l’utilisation d’un système multi-agents per- met de réaliser un comportement plus réaliste pour les joueurs non humains et bien plus compétitifs pour le joueur humain. / A multi-agent system is composed of several autonomous agents that interact with each other in a common environment. This thesis aims to demonstrate the use of a multi- agent system for the development of a video game. First, a justification of the artificial intelligence’s concepts used in this master’s thesis is exposed. Subsequently, a practical approach is used in developping a video game. To do this, the game was developed from an existing single-agent video game and modified into a multi-agent system in order to properly highlight the benefits of a multi-agent system in a video game. The development of this game also demonstrate the application of other concepts in artificial intelligence such as pathfindinig and behaviour trees. In summary, the use of a multi- agent system has achieved a more realistic behavior for the non-human players and a more competitive gameplay for the human player.

Un système multi-agents est composé de plusieurs agents autonomes qui interagissent entre eux dans un environnement commun. Ce mémoire vise à démontrer l’utilisation d’un système multi-agents pour le développement d’un jeu vidéo. Tout d’abord, une justification du choix des concepts d’intelligence artificielle choisie est exposée. Par la suite, une approche pratique est utilisée en effectuant le développement d’un jeu vidéo. Pour ce faire, le jeu fut développé à partir d’un jeu vidéo mono-agent existant et mo- difié en système multi-agents afin de bien mettre en valeur les avantages d’un système multi-agents dans un jeu vidéo. Le développement de ce jeu a aussi démontré l’applica- tion d’autres concepts en intelligence artificielle comme la recherche de chemins et les arbres de décisions. Le jeu développé pour ce mémoire viens appuyer les conclusions des différentes recherches démontrant que l’utilisation d’un système multi-agents per- met de réaliser un comportement plus réaliste pour les joueurs non humains et bien plus compétitifs pour le joueur humain. / A multi-agent system is composed of several autonomous agents that interact with each other in a common environment. This thesis aims to demonstrate the use of a multi- agent system for the development of a video game. First, a justification of the artificial intelligence’s concepts used in this master’s thesis is exposed. Subsequently, a practical approach is used in developping a video game. To do this, the game was developed from an existing single-agent video game and modified into a multi-agent system in order to properly highlight the benefits of a multi-agent system in a video game. The development of this game also demonstrate the application of other concepts in artificial intelligence such as pathfindinig and behaviour trees. In summary, the use of a multi- agent system has achieved a more realistic behavior for the non-human players and a more competitive gameplay for the human player.

Denna rapport går igenom utvecklingen av spelet Metal in Your Brain som skickas in som ett bidrag till Swedish Game Awards (SGA), nordens största spelutvecklar-tävling och hur den artificiella intelligensen är uppbyggd och funktionerar. Den artificiella intelligensen är konstruerad på två sätt: fuzzy logic som står bakom de handlingar som en icke spelande karaktär (NPC) tar beroende på vilken situation den befinner sig i och A*-sökning som används för att en NPC ska kunna söka sig fram till ett mål och ta den kortaste vägen eller fly från spelaren för att gömma sig bakom närmsta skydd. Metal in Your Brain är ett 2D top-down shooter för Windowsdatorer där man tillsammans med två andra spelare ska överleva en rad vågor av fiender och möjligen samtidigt utföra ett antal uppdrag beroende på vilken bana det är. / This report reviews the development of the game Metal inYour Brain that will be submitted to Swedish Game Awards (SGA), Scandinavia's largest game developer competition and how the artificial intelligence is structured and functions. The artificial intelligence is constructed in two ways: fuzzy logic that decides the actions for a non-player character (NPC) will take depending on the situation it find itself in, and A * search that is used for a NPC to be able to seek out a goal and take the shortest route to get there or flee from the player and take cover behind the nearest shelter. Metal In Your Brain is a 2D top-down shooter for Windows computers where you, together with two other players aim to survive a series of waves of enemies and possibly simultaneously perform a number of tasks depending on what level one is at. 1

This thesis deals with a path planning using cellular automata algorithms in a rectangular grid environment. Theoretical part starts with an overview of commonly used approaches for path planning and later on focuses on existing cellular automata solutions and capabilities in detail. Implemented cellular automata algorithms and the commonly used path planning algorithms are together with a map generator described in the practical part. Conclusion of this thesis contains results completed in a special application.