Anshul Goyal

Academic Writings

Human-like Bots for Unreal Tournament 2004 A Q-learning Approach to Refine UT2004 Strategies

Abstract: Defining human-like behavior for bots or non-player characters (NPC) in a real-time game such as Unreal Tournament 2004 (UT2004) has always been a challenging task. Dynamic game environments challenge the bot to behave autonomously without any human supervision or intervention. We highlight how we can use Behavior Trees to define agent’s behavior and use Reinforcement Learning (RL) technique to teach the bot how to choose actions so as to maximize numerical reward. The other motivation behind creating such an agent that can behave like a human is to participate in the Bot Prize competition [3], which is similar to the Turing test for intelligent agents, wherein the bots have to convince the human judges that ’It’ is a human, while playing the First Person Shooter (FPS) UT2004.

Real-Time Planning Using HTN in Stealth Game

Abstract: Video game development have come a long way in terms of incorporating logical reasoning and decision making for Non Player Characters (NPC) either as an individual or a group working together towards a common goal. Planning is a key ability for intelligent agents when working in a group, increasing their autonomy and flexibility through the construction of sequences of actions to achieve their goals. It has been an area of research since the earliest days of Artificial Intelligence (AI) and cognitive science. This paper will explore the concepts of Hierarchical Task Network (HTN) planning algorithm and how it can be effectively used to perform real‐time planning among a group of NPC’s in a stealth‐based game.