Behavior Of Non-player Characters Research Articles

The Application of Artificial Intelligence in Game

AI has been extensively applied in various aspects of game development over the past few decades. One critical application is the use of AI algorithms to simulate intelligent decision-making processes for non-player character behavior. Another application is procedural content generation, where AI algorithms generate game content such as level designs or missions to enhance gameplay diversity. This paper aims to explore the application of AI in game development by analyzing specific cases and discussing the current state of the field. The paper also focuses on related topics such as motion and physics simulation, as well as dynamic adaptation. The current state of using AI to enhance realistic motion and physics simulation in game development is further examined. In summary, this paper provides a comprehensive overview of the application of AI in game development, highlighting its significance. The paper concludes with a summary and future outlook of the entire study.

Experiment of Fuzzy Logic on Game System

Abstract: Applications of fuzzy logic in game creation have proven useful, especially when it comes to modelling the behavior of non-player characters (NPCs) in video games. NPC behavior in video games has historically been governed by strict rule-based systems or scripted sequences, which frequently led to restricted and predictable interactions. On the other hand, fuzzy logic adds a degree of flexibility and ambiguity to the decision-making processes of NPCs. By improving the naturalness, adaptability, and emotional resonance of NPC behavior, fuzzy logic may greatly enhance player experiences in games. It provides players with progressively more engaging and dynamic gaming experiences as it continues to be incorporated into game production, opening up new possibilities for interactive narrative, gaming, and immersive virtual worlds. Based on Utku Kose's article, "Developing a Fuzzy Logic Based Game System," I have investigated the use of fuzzy logic in gaming contexts for NPC decision-making in this study.

A modeling environment for reinforcement learning in games

Developing Non-Player Characters, i.e., game characters that interact with the game’s environment autonomously with flexibility to experiment different behavior configurations is not a trivial task. Traditionally, this has been done with techniques that limit the complexity of the behavior of Non-Player Characters (NPCs), as in the case of using Navigation Mesh (NavMesh) for navigation behaviors. For this problem, it has been shown that reinforcement learning can be more efficient and flexible than traditional techniques. However, integrating reinforcement learning into current game development tools is laborious, given that a great deal of experimentation and coding is required. For that, we have developed a modeling environment that integrates with a game development tool and allows the direct specification of reward functions and NPC agent components with maximum code reuse and automatic code generation.

Behaving Like Soldiers


 Training soldiers through virtual reality simulations has many benefits, two of which are reduced cost to conduct training operations and the ability to repeatedly perform activities that are not feasible in real life. The use of automated agents in place of human players or instructors has reduced cost. The sophistication of such agents can range from highly scripted scenarios designed to teach a specific concept to advanced artificially intelligent behaviors that react realistically to a wider range of in-game activities. Repeatability with realistic variation in the automated courses of action (COAs) is a primary objective, but often this comes at the cost of consistency to the doctrine (e.g., “what might I expect that support squad do in this case?”) and explainability of behaviors during after action review (e.g., “why did the enemy do that in this situation?”).
 
 
 This paper describes early progress to develop advanced automated forces that are informed by a given corpus of doctrine. They anticipate future states with uncertainty to generate diverse COAs that guide higher level behaviors of nonplayer characters in a simulation, and they may be queried to explain behaviors at various points of the simulation. We are working with the US Army Combat Capabilities Command Soldier Center (CCDC-SC) to develop this technology for future integration into advanced virtual training systems.

Behavior model of an intelligent agent based on a genetic algorithm

A solution of improving the behavior model of a non-player character as an intelligent agent by optimizing input parameters based on a genetic algorithm is presented. The proposed approach includes the development of a non-player character model: a skeleton, rigid bodies, the implementation of a dynamic model based on the Featherstone algorithm, and modeling of the character's behavior based on a genetic algorithm. The formation of a behavior model using a genetic algorithm that simulates the physical properties of a character, taking into account his actions, is proposed. The stages of the genetic algorithm include creating an initial population, fitness score, selection, crossing and mutation. Based on the results of the experiments, the input parameters of the non-player character behavior model were determined, maximizing the cumulative fitness score, which acts as an estimate of the reward, which can be used as initial values for further experiments.
 Keywords: non-player character, intelligent agent, simulation, genetic algorithm

Modeling Self-Adapting and Multicultural Non-Player Characters in Artificial Worlds

Modeling the behavior of Non-Player Characters (NPCs) so that they can act properly in an artificial world is a difficult, yet essential task in the development of modern computer games, often requiring never- ending adjustments of handwritten rules in a way to enable NPCs to perform well in the game world. This pa- per introduces an evolutionary method relying on high-level rule-based behaviors, genetic algorithms and speciation. By interconnecting these three areas, we pro- pose a technique that has the potential to provide multicultural NPCs with the ability to adapt themselves to an artificial worl

Dynamic Expansion of Behaviour Trees

Artificial intelligence in games is typically used for creating player's opponents. Manual edition of intelligent behaviors for Non-Player Characters (NPCs) of games is a cumbersome task that needs experienced designers. Our research aims to assist designers in this task. Behaviours typically use recurring patterns, so that experience and reuse are crucial aspects for behavior design. The use of hierarchical state machines allows working on different abstraction levels, sharing transitions and reusing pieces from the more detailed levels. However, the static nature of the design process does not release the designer from the burden to completely specify each behaviour. Our approach applies Case-Based Reasoning (CBR) techniques to retrieve and reuse stored behaviors represented as hierarchical state machines (actually, behaviour trees). In this paper we focus on dynamic retrieval of behaviours taking into account the world state and the underlying goals to select the most appropriate state machine to guide the NPC behaviour. The global behaviour of the NPC is dynamically built in run time querying the CBR system. We exemplify our approach through a serious game, developed by our research group, with gameplay elements from First-Person Shooter (FPS) games.

An AI System for Large Open Virtual World

In recent years, computer games have reached unprecedented level of graphical fidelity to the real world. As the non-player characters (NPCs) in the game world look more and more realistic, players expect them to manifest believable behavior as well. This is accented especially in games that feature large open worlds, which players may explore freely and it is thus not possible to explicitly account for all possible player interactions. In this paper we focus mainly on ambient AI - the logic behind day to day behaviors of NPCs as they sleep, work and entertain themselves in the virtual world. In this context, it is of great importance to build a system that handles many NPCs (up to several hundreds) quickly. In this paper we report on an implementation of a particular AI system that was approved for deployment in an upcoming high-budget game. The system features a hierarchy of control similar to the subsumption architecture and a visual agent-based language inspired by behavior trees. We describe the challenges involved in building such a system and specific design decisions we have made that let us achieve a level of behavioral fidelity unmatched by existing games. Finally we evaluate the performance of the system in a realistic setting.

Demonstration-Based Training of Non-Player Character Tactical Behaviors

State of the art methods for generating non-player character (NPC) tactical behaviors typically depend on hard-coding actions or minimizing a given objective function. In many games however, it is hard to foresee how the NPC should behave to appear intelligent or to accommodate human player preferences for NPC tactics. In this paper we consider an alternative approach, by training NPC tactical behavior via demonstrations. We propose a heuristic search-based planning method based on previously-developed Experience Graphs, which facilitates the use of behavior demonstration data to plan goal-oriented NPC behavior. Our method provides a principled solution to the problem which tolerates some amount of differences in between the training demonstration and the actual problem and yet still grants guarantees on the quality of the solution output.

Building behavioral AI using trust and reputation model based on mask model.

This article describes a concept of artificial intelligence that simulates a behavior of non-player characters or NPCs based on the architecture of the mask model using the algorithm of trust and reputation. The mask model allows you to quickly create different behavioral models using a ready-made set of different character roles and relationships to different entities. The algorithm of trust and reputation changes calculation rules of character‘s actions over time depending on a player actions, thereby creating a dynamic and lively environment. Combining these two models gives an ability to configure different personalities with an extensive decision-making system. Article describes the decision-making algorithms, changes in a character’s reputation and an implementation example.

HRLB⌃2: A Reinforcement Learning Based Framework for Believable Bots

The creation of believable behaviors for Non-Player Characters (NPCs) is key to improve the players’ experience while playing a game. To achieve this objective, we need to design NPCs that appear to be controlled by a human player. In this paper, we propose a hierarchical reinforcement learning framework for believable bots (HRLB⌃2). This novel approach has been designed so it can overcome two main challenges currently faced in the creation of human-like NPCs. The first difficulty is exploring domains with high-dimensional state–action spaces, while satisfying constraints imposed by traits that characterize human-like behavior. The second problem is generating behavior diversity, by also adapting to the opponent’s playing style. We evaluated the effectiveness of our framework in the domain of the 2D fighting game named Street Fighter IV. The results of our tests demonstrate that our bot behaves in a human-like manner.

Rapid modeling of human-defined AI behavior patterns in games

The creation of computer and videogames is a challenging and multidisciplinary endeavor, requiring different approaches to integrate different disciplines while keeping relatively low development costs. In this context, domain specific languages (DSLs) are increasingly becoming a valid tool, allowing nonprogrammers to participate in the development process. In this work, we focus on a DSL developed within the Gade4all project, focused on defining the behaviors of non-player characters, proposing a design which allows the participation of users with a lack of programming knowledge to define behavior and interaction complex patterns of in-game enemies in a simple and straightforward way. This preferable to other approaches to defining opponent AIs in games, such as machine learning techniques (which produce advanced AI opponents that do not behave as humans) or rigid rule systems created by programmers rather than behavioral experts. The approach has been tested by comparing the creation of well-known AI patterns using T2GAME with popular game editors, and resulted in significantly reduced development times, in addition to being more approachable for non-programmers such as behavioral psychologists.

Behavior Trees for Modelling Artificial Intelligence in Games: A Tutorial

We provide a tutorial introduction to behavior trees, which are a useful way of structuring artificial intelligence in games. A behavior tree is a model of plan execution that is graphically represented as a tree. A node in a tree either encapsulates an action to be performed or acts as a control flow component that directs traversal over the tree. Behavior trees are appropriate for specifying the behavior of non-player characters and other entities because of their maintainability, scalability, reusability, and extensibility. We describe the main features of behavior trees, show an example of how to create a behavior tree, and briefly survey existing packages for editing behavior trees. We recommend that behavior trees be used when some game designers are not programmers, the conditions governing the behavior are complex, and the NPCs have aspects of behavior in common.

A model of non-player character believability

Abstract In this study we aim to describe in what ways the behaviour of non-player characters (NPCs) affects to what extent the player finds the game experience to be believable. To this end, we have conducted an online survey, where respondents were asked to classify and describe NPCs. Furthermore, we also examined recordings of NPCs in games. These data sources were analysed using a model for NPC social believability in order to describe the effects of NPC behaviour in relation to how different types of NPCs are perceived as being believable. Based on this we were able to construct a model of NPC believability, which describes the NPC’s level of complexity and ability to handle a mutable social context. As described by the model, NPCs are currently less capable of handling changing social contexts. They do, however, show promise, and given current emerging technologies it is feasible that new types of more socially capable NPCs will arise within the near future.

PENERAPAN ALGORITMA COLLISION DETECTION DAN BAYESIAN UNTUK STRATEGI MENYERANG JARAK DEKAT PADA NPC (NON PLAYER CHARACTER)MENGGUNAKAN UNITY 3D

The game is a field of science is growing rapidly and very interesting to learn because review has potential good view of science and commercial aspect. To produce an attractive and realistic games, there are a few things to note. One of them is the attention to the attack strategy, especially for the NPC (Non Player Character). Referring to previous studies that discuss strategies melee attack for NPC (Non Player Character)[1]. This study combines two Collision Detection with Bayesian methods for attacking strategy melee. In this study is to divide some of the behavior of NPCs attack when in the position closest to the enemy. This study applies the algorithm for Bayesian classification assault NPC behavior and collision detection algorithms for decision-making behavior when NPCs hit the player. By merging algorithm aims to improve on the weaknesses of previous research. NPC assault classification is divided into two offensive behavior: attitudes and behavior kick smash. As for the variables used in Bayesian classification is health points, attack points player and the distance obtained from NPC conditions. The results of the implementation of collision detection and Bayesian methods have proved that the collision detection methods can take decisions NPC attack behavior although placed closest to the player. From the test results with the calculation methods confusion matrix gain as much as 90% accuracy.

Supporting the Construction of a GUI Component for Specifying the Behavior of Non-Player Characters in Unity

Unity 3D is a widely used middleware for game development, primarily by small studies. In addition to the core technology, a number of extensions or add-ons have been created by third-party developers to provide additional functionality. Nevertheless, the support Unity 3D provides for building GUI for such add-ons is at a very low level, and for that reason the authors have designed and built UHotDraw, an extensible framework in C# that simplifies the development of GUI and draw applications and editor extensions in Unity 3D.

동물의 세력 투쟁 행동을 이용한 게임 인공 지능 구현

게임에서 인공 지능은 주로 NPC(Non Player Character)와 적의 행동 패턴을 결정하거나 길 찾기에 사용된다. 이러한 인공 지능을 구현하는 경우에 FSM(Finite State Machine)과 플로킹(Flocking) 방법이 사용된다. FSM 방법에서는 상태 개수에 따라 NPC의 행동 개수에도 제한을 받는다. 상태 개수가 너무 적은 경우 플레이어들이 쉽게 NPC의 행동 패턴을 알 수 있으며 너무 많은 경우에는 구현이 복잡하게 된다. Flocking 방법에서는 리더의 결정에 따라 NPC들의 행동이 결정되기 때문에 NPC들의 이동 패턴이나 공격 방향을 쉽게 플레이어들이 알 수 있다. 본 논문에서는 이 문제를 개선하기 위하여 동물의 세력 투쟁 행동(공격, 위협, 의례적인 보여줌, 기피, 복종)들을 NPC에 적용하는 것을 제안하고 이를 Unity3D 엔진을 이용하여 구현한다. 이 논문은 실제감 있는 NPC 인공 지능 제작에 도움을 줄 수 있다. Artificial intelligence in the game is mainly used to determine patterns of behavior of NPC (Non Player Character) and the enemy, path finding. These artificial intelligence is implemented by FSM (Finite State Machine) or Flocking method. The number of NPC behavior in FSM method is limited by the number of FSM states. If the number of states is too small, then NPC player can know the behavior patterns easily. On the other hand, too many implementation cases make it complicated. The NPC behaviors in Flocking method are determined by the leader's decision. Therefore, players can know easily direction of movement patterns or attack pattern of NPCs. To overcome these problem, this paper proposes agonistic behaviors(attacks, threats, showing courtesy, avoidance, submission)in animals to apply for the NPC, and implements agonistic behaviors using Unity3D engine. This paper can help developing a real sense of the NPC artificial intelligence.

The SimpleFPS Planning Domain: A PDDL Benchmark for Proactive NPCs

In this paper we focus on proactive behavior for non-player characters (NPCs) in the first-person shooter (FPS) genre of video games based on goal-oriented planning. Some recent approaches for applying real-time planning in commercial video games show that the existing hardware is starting to follow up on the computing resources needed for such techniques to work well. Nonetheless, it is not clear under which conditions real-time efficiency can be guaranteed. In this paper we give a precise specification of SimpleFPS, a STRIPS planning domain expressed in PDDL that captures some basic planning tasks that may be useful in a first person shooter video game. This is intended to work as a first step towards quantifying the performance of different planning techniques that may be used in real-time to guide the behavior of NPCs. We present a simple tool we developed for generating random planning problem instances in PDDL with user defined properties, and show some preliminary results based on SimpleFPS instances that vary in the size of the domain and two well-known planners from the planning community.

Query-Enabled Behavior Trees

Artificial intelligence in games is typically used for creating player's opponents. Manual editing of intelligent behaviors for nonplayer characters (NPCs) of games is a cumbersome task that needs experienced designers. Our research aims to assist designers in this task. Behaviors typically use recurring patterns, so that experience and reuse are crucial aspects for behavior design. The use of hierarchical structures like hierarchical state machines, behavior trees (BTs), or hierarchical task networks, allows working on different abstraction levels reusing pieces from the more detailed levels. However, the static nature of the design process does not release the designer from the burden of completely specifying each behavior. Our approach applies case-based reasoning (CBR) techniques to retrieve and reuse stored behaviors represented as BTs. In this paper, we focus on dynamic retrieval and selection of behaviors taking into account the world state and the underlying goals. The global behavior of the NPC is dynamically built at runtime querying the CBR system. We exemplify our approach through a serious game, developed by our research group, with gameplay elements from first-person shooter (FPS) games.

Définir des individualités pour des personnages non joueurs

Designing the behaviour of non player characters (NPC), in role-playing video games, is a hard problem both from the programming point of view and from behaviour modelling considerations. We tackle these hurdles with the interaction oriented approach, promoted by the IODA-CoCoA project. It enables us to propose a generic engine suitable to several contexts. This engine is combined with an action selection mechanism based on motivations. This mechanism provides means to define several NPC behaviours by tuning the motivation parameters without new programming code. Our proposition constitutes a behavioural engine dedicated to the modelling of situated character behaviour. Thanks to its genericity, this engine can be used in various environment (ie games) and for various agents, since it adapts to individual specificities and abilities while proposing a diversity in designed behaviours.