Most story-based games today have manually-scripted non-player characters (NPCs) and the scripts are usually simple and repetitive since it is time-consuming for game developers to script each character individually. ScriptEase, a publicly-available author-oriented developer tool, attempts to solve this problem by generating script code from high-level design patterns, for BioWare Corp.'s role-playing game Neverwinter Nights. The ALeRT algorithm uses reinforcement learning (RL) to automatically generate NPC behaviours that change over time as the NPCs learn from the successes or failures of their own actions. This thesis aims to provide a new learning mechanism to game agents so they are capable of adapting to new behaviours based on the actions of other agents. The new on-line RL algorithm, ALeRT-AM, which includes an agent-modeling mechanism, is applied in a series of combat experiments in Neverwinter Nights and integrated into ScriptEase to produce adaptive behaviour patterns for NPCs.

No description available.

Background. Game developers are continuously searching for new ways of populating their vast game worlds with competent and engaging Non-Player Characters (NPCs), and researchers believe Deep Reinforcement Learning (DRL) might be the solution for emergent behavior. Consequently, fusing NPCs with DRL practices has surged in recent years, however, proposed solutions rarely outperform traditional script-based NPCs. Objectives. This thesis explores a novel method of developing an adversarial DRL NPC by combining Reinforcement Learning (RL) algorithms. Our goal is to produce an agent that surpasses its script-based opponents by first mimicking their actions. Methods. The experiment commences with Imitation Learning (IL) before proceeding with supplementary DRL training where the agent is expected to improve its strategies. Lastly, we make all agents participate in 100-deathmatch tournaments to statistically evaluate and differentiate their deathmatch performances. Results. Statistical tests reveal that the agents reliably differ from one another and that our learning agent performed poorly in comparison to its script-based opponents. Conclusions. Based on our computed statistics, we can conclude that our solution was unsuccessful in developing a talented hostile DRL agent as it was unable to convey any form of proficiency in deathmatches. No further improvements could be applied to our ML agent due to the time constraints. However, we believe our outcome can be used as a stepping-stone for future experiments within this branch of research.

Climbing in video games is a game mechanic that, after the introduction of three-dimensional movement, became more and more common and extended the movement options of the player character. Climbing, as used in this thesis, is a very specific action that can only be performed by interacting with a climbable gameplay element in the game world. A visual analysis is applied to investigate how objects in action-adventure games depict climbable gameplay elements: specifically, whether the way the games communicate climbability through visual communications is consistent or not. Four games are included in this study and the results point towards an overarching consistency for each of the games' use of visual cues to communicate climbability. However, the consistency is affected by the gameplay experience of each game. The linear games use very clear and controlled visual cues to describe the climbable gameplay elements, and the player often only interacts with a climbable gameplay element once per section, similar to an obstacle course. The games that emphasize exploration of a gameplay area, akin to a playground, place climbable gameplay elements around the environment for the player to freely interact with and have less clear visual cues. Climbable gameplay elements, like ledges, are found to be unique in the interaction they provide, and this is proven by the many different methods of visual communication that are used to convey the affordance of climbing. / Klättrande i dataspel är en spelmekanik som efter introduktionen av tredimensionell rörelse, blivit allt vanligare och utvidgat möjligheterna för spelarkaraktärens rörelser. Klättrande, som använt i denna uppsats, är en mycket specifik handling som endast kan utföras genom att interagera med ett klättringsbart spelelement i spelvärlden. En visuell analys appliceras för att undersöka hur objekt i action-äventyrsspel avbildar klättringsbara spelelement. Specifikt, huruvida sättet spelen kommunicerar klättringsbarhet genom visuell kommunikation är konsekvent eller ej. Fyra spel är inkluderade i denna studie och resultaten pekar mot att det finns en övergripande följdriktighet för varje spels användande av visuella signaler för att kommunicera klättringsbarhet. Följdriktigheten är dock påverkad av spelupplevelsen i varje spel. Dem linjära spelen använder mycket tydliga och kontrollerade visuella signaler för att beskriva klättringsbara spelelement, och spelaren interagerar oftast endast med ett klättringsbart spelelement en gång per sektion, likt en hinderbana. Spel som betonar utforskande av ett spelområde, likt en lekplats, har mindre tydliga visuella signaler. Klättringsbara spelelement, såsom avsatser, upptäcks vara unika i interaktionen de förmedlar, och detta påvisas av de många olika metoderna för visuell kommunikation som används för att förmedla klättringsbarhet.