20.500.12556/RUL-158698 Hibridna metoda za razvoj prepričljivega vedenja agentov v videoigrah Hybrid method for implementing believable agent behaviour in video games V industriji računalniških iger smo bili v zadnjem desetletju priča napredkom v računalniški grafiki, zvočni produkciji in količini uporabniških vsebin, za razvoj vedenja agentov v video igrah pa se še vedno uporabljajo tradicionalne metode, ki so v industriji prisotne že od samega začetka. Za ohranjanje rasti industrije in zadovoljstva končnih uporabnikov morajo v prihodnosti inovacije prodreti tudi na druga področja znotraj industrije. Naslednje generacije video iger ne bodo več zaznamovale višje resolucije tekstur, temveč realistično avtonomno vedenje agentov, ki bodo zmožni učenja in prilagajanja dani situaciji. V magistrskem delu predstavimo novo hibridno metodo za razvoj prepričljivega vedenja agentov v video igrah in jo primerjamo z obstoječimi pristopi. Metodo smo osnovali na podlagi vedenjskih dreves, končnih avtomatov in metode spodbujevalnega učenja Q-learning. V predstavljeni hibridni metodi združimo prednosti naštetih pristopov in hkrati odpravimo nekatere slabosti. Prepričljivost nove metode ovrednotimo s pomočjo vprašalnika, njeno učinkovitost in ustreznost pa primerjamo z uveljavljenimi pristopi razvoja vedenja inteligentnih agentov. Over the last decade, the video game industry has witnessed advancements in computer graphics, sound production and the quantitiy of available content. However, the development of agent behaviour in video games still relies on traditional methods that have been present in the industry since its very inception. To sustain the industry's growth and meet end-user satisfaction requirements in the future, innovations must also be introduced in other areas within the industry. The next generation of video games will no longer be marked by higher texture resolutions, but by realistic autonomous behaviour of agents, capable of learning and adapting to given situations. In this master's thesis we present a new hybrid method for developing believable agent behaviour in video games and compare it with existing approaches. We based the method on behaviour trees, finite state machines, and the reinforcement learning method Q-learning. The presented hybrid method combines the advantages of the listed approaches while addressing some of their weaknesses. We evaluate the behaviour believability of the new method using a questionnaire, and compare its efficiency and suitability with established approaches to developing intelligent agent behaviour. računalniške igre inteligentni agenti prepričljivost vedenja spodbujevalno učenje odločitvena drevesa končni avtomati video games intelligent agents behaviour believability reinforcement learning decision trees finite-state machines true false false Slovenski jezik Angleški jezik Magistrsko delo 2024-06-19 12:35:01 2024-06-19 12:35:05 2024-07-12 09:56:08 0000-00-00 00:00:00 2024 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 37064 201583363 Simonic_Arne_-_Hibridna_metoda_za_razvoj_prepricljivega_vedenja_agentov_v_videoigrah.pdf Simonic_Arne_-_Hibridna_metoda_za_razvoj_prepricljivega_vedenja_agentov_v_videoigrah.pdf 1 3968D4AFBDB347D5B31C0C66A414E453 d5d73f545a793da97e95d69ef9ba392f040c691526225d523d851ffb154f0e41 97bc8045-2e27-11ef-aa1b-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=186410 Fakulteta za računalništvo in informatiko Fakulteta za elektrotehniko 0 0 0