--Angry Birds is a popular video game in which the player is provided with a sequence of birds to shoot from a slingshot. The task of the game is to destroy all green pigs with maximum possible score. Angry Birds appears to be a difficult task to solve for artificially intelligent agents due to the sequential decision-making, non-deterministic game environment, enormous state and action spaces and requirement to differentiate between multiple birds, their abilities and optimum tapping times. We describe the application of Deep Reinforcement learning by implementing Double Dueling Deep Q-network to play Angry Birds game. One of our main goals was to build an agent that is able to compete with previous participants and humans on the first 21 levels. In order to do so, we have collected a dataset of game frames that we used to train our agent on. We evaluate our agent using results of the previous participants of AIBirds competition, results of volunteer human players and present the results of AIBirds 2018 competition. I NTRODUCTION Angry Birds has been one of the most popular video games for a period of several years. The main goal of the game is to kill all green pigs on the level together with applying as much damage as possible to the surrounding structures.

In this paper we tackle a problem of tile-based combat in the turn-based strategy (space 4X) video game Children of the Galaxy (CotG). We propose an improved version of Monte Carlo tree search (MCTS) called MCTS considering hit points (MCTS_HP). We show MCTS_HP is superior to Portfolio greedy search (PGS), MCTS and NOKAV reactive agent in small to medium combat scenarios. MCTS_HP performance is shown to be stable when compared to PGS, while it is also more time-efficient than regular MCTS. In smaller scenarios, the performance of MCTS_HP with 100 millisecond time limit is comparable to MCTS with 2 seconds time limit. This fact is crucial for CotG as the combat outcome assessment is precursor to many strategical decisions in CotG game. Finally, if we fix the amount of search time given to the combat agent, we show that different techniques dominate different scales of combat situations. As the result, if search-based techniques are to be deployed in commercial products, a combat agent will need to be implemented with portfolio of techniques it can choose from given the complexity of situation it is dealing with to smooth gameplay experience for human players.

The quality of high-level AI of non-player characters (NPCs) in commercial open-world games (OWGs) has been increasing during the past years. However, due to constraints specific to the game industry, this increase has been slow and it has been driven by larger budgets rather than adoption of new complex AI techniques. Most of the contemporary AI is still expressed as hard-coded scripts. The complexity and manageability of the script codebase is one of the key limiting factors for further AI improvements. In this paper we address this issue. We present behavior objects - a general approach to development of NPC behaviors for large OWGs. Behavior objects are inspired by object-oriented programming and extend the concept of smart objects. Our approach promotes encapsulation of data and code for multiple related behaviors in one place, hiding internal details and embedding intelligence in the environment. Behavior objects are a natural abstraction of five different techniques that we have implemented to manage AI complexity in an upcoming AAA OWG. We report the details of the implementations in the context of behavior trees and the lessons learned during development. Our work should serve as inspiration for AI architecture designers from both the academia and the industry.

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.

This paper deals with planning domains that appear in computer games, especially when modeling intelligent virtual agents. Some of these domains contain only actions with no negative effects and are thus treated as easy from the planning perspective. We propose two new techniques to solve the problems in these planning domains, a heuristic search algorithm ANA* and a constraint-based planner RelaxPlan, and we compare them with the state-of-the-art planners, that were successful in IPC, using planning domains motivated by computer games.