Real-time Strategy Games Research Articles

Playing real-time strategy games by imitating human players’ micromanagement skills based on spatial analysis

Unlike the situation with board games, artificial intelligence (AI) for real-time strategy (RTS) games usually suffers from numerous possible future outcomes because the state of the game is continuously changing in real time. Furthermore, AI is also required to be able to handle the increased complexity within a small amount of time. This constraint makes it difficult to build AI for RTS games with current state-of-the art intelligence techniques. A human player, on the other hand, is proficient in dealing with this level of complexity, making him a better game player than AI bots. Human players are especially good at controlling many units at the same time. It is hard to explain the micro-level control skills needed with only a few rules programmed into the bots. The design of micromanagement skills is one of the most difficult parts in the StarCraft AI design because it must be able to handle different combinations of units, army size, and unit placement. The unit control skills can have a big effect on the final outcome of a full game in professional player matches. For StarCraft AI competitions, they employed a relatively simple scripted AI to implement the unit control strategy. However, it is difficult to generate cooperative behavior using the simple AI strategies. Although there has been a few research done on micromanagement skills, it is still a challenging problem to design human-like high-level control skills. In this paper, we proposed the use of imitation learning based on human replays and influence map representation. In this approach, we extracted huge numbers of cases from the replays of experts and used them to determine the actions of units in the current game case. This was done without using any hand-coded rules. Because this approach is data-driven, it was essential to minimize the case search times. To support fast and accurate matching, we chose to use influence maps and data hashing. They allowed the imitation system to respond within a small amount time (one frame, 0.042s). With a very large number of cases (up to 500,000 cases), we showed that it is possible to respond competitively in real-time, with a high winning percentage in micromanagement scenarios.

Analysing the influence of the fitness function on genetically programmed bots for a real-time strategy game

Finding the global best strategy for an autonomous agent (bot) in a RTS game is a hard problem, mainly because the techniques applied to do this must deal with uncertainty and real-time planning in order to control the game agents. This work describes an approach applying a Genetic Programming (GP) algorithm to create the behavioural engine of bots able to play a simple RTS. Normally it is impossible to know in advance what kind of strategies will be the best in the most general case of this problem. So GP, which searches the general decision tree space, has been introduced and used successfully. However, it is not straightforward what fitness function would be the most convenient to guide the evolutionary process in order to reach the best solutions and also being less sensitive to the uncertainty present in the context of games. Thus, in this paper three different evaluation functions have been proposed, and a detailed analysis of their performance has been conducted. The paper also analyses several aspects of the obtained bots, in addition to their final performance on battles, such as the evolution of the decision trees (behavioural models) themselves, or the influence on the results of noise or uncertainty. The results show that a victory-based fitness, which prioritises the number of victories, contributes to generate better bots, on average, than other functions based on other numerical aspects of the battles, such as the number of resources gathered, or the number of units generated.

A multi-measure approach for connecting cognitive workload and automation

Abstract Eye behavior metrics, time of run, and subjective survey results were assessed during human-computer interaction with high, low, and intermediate system autonomy levels. The results of this study are provided as a contribution to knowledge on the relationship between cognitive workload physiology and automation. Research suggests that changes in eye behavior metrics are related to changes in cognitive workload. Few studies have investigated the relationship between eye behavior metrics physiology measures and levels of automation. A within-subjects experiment involving 18 participants who played an open-source real-time strategy game was conducted. Three different versions of the game were developed, each with a unique static autonomy level designed from Sheridan and Verplank's 10 levels of autonomy (levels 2, 4, and 9). NASA-TLX subject survey ratings, time to complete run, and visual fixation rate were found to be significantly different among automation levels. These findings suggest that assessing visual physiology may be a promising indicator for evaluating cognitive workload when interacting with static autonomy levels. This efforts takes us one step closer to using visual physiology as a useful method for evaluating operator workload in almost real-time. Relevance to industry Potential applications of this research include development of software that integrates adaptive automation to improve human-computer task performance for high cognitive workload tasks (Air Traffic Control, aircraft piloting, process control, information analyst, etc.).

Multiscale Bayesian Modeling for RTS Games: An Application to StarCraft AI

This paper showcases the use of Bayesian models for real-time strategy (RTS) games AI in three distinct core components: micromanagement (units control), tactics (army moves and positions), and strategy (economy, technology, production, army types). The strength of having end-to-end probabilistic models is that distributions on specific variables can be used to interconnect different models at different levels of abstraction. We applied this modeling to StarCraft, and evaluated each model independently. Along the way, we produced and released a comprehensive data set for RTS machine learning.

A real-time strategy game, “GALLIPOLI WARS,” as a centennial tribute to the Gallipoli Campaign (1915–2015)

The same authors had reported a review of game development processes and methodologies in an earlier paper. In that paper, game design elements, such as the players, story, rules, objectives, procedures, conflict, and challenge, and their effect over gameplay were treated. Technical components in that paper included the render engine and rendering pipeline (fixed function pipeline and flexible pipeline), physics engine and physics-related techniques (collision detection, ray-casting, etc.), game codes (for game mechanics, artificial intelligence, scenario creation, and management) and artwork contents (game level, three-dimensional (3D) models, two-dimensional maps for shaders, skeletal animation, and audio assets) that form a game. Along with their development processes and approaches, all of these technical and artwork components are applied on a case problem and are summarized in this paper. The case problem considered is a small part of Gallipoli or Dardanelles Wars in Ottoman history, as a real-time 3D strategy game.

Assessing the economic effects of server launches in free-to-play MMO games

In MMO (massively multi-player online) games, the interaction among thousands of players can produce both positive and negative network effects. It is therefore crucial for a company managing MMO games to establish how the entire user base should be distributed across different game instances (‘servers’) in order to optimize the players’ game experience and, ultimately, the company’s revenues. While splitting the user base onto different game servers is a common measure in the industry, there is a conspicuous lack of clear guidelines as to when launching new game servers would be advisable. Our work notably fills this gap: for a popular MMO real-time strategy game, our counterfactual simulation shows that a division of the user base across different game servers can lead to additional revenues if implemented at the right time. For the game in focus, our results indicate that launching multiple new game servers would be beneficial in terms of revenue when the game was new, and additional servers every 90–120 days when it was in decline.

Unsupervised Learning of HTNs in Complex Adversarial Domains

While Hierarchical Task Networks are frequently cited as flexible and powerful planning models, they are often ignored due to the intensive labor cost for experts/programmers, due to the need to create and refine the model by hand. While recent work has begun to address this issue by working towards learning aspects of an HTN model from demonstration, or even the whole framework, the focus so far has been on simple domains, which lack many of the challenges faced in the real world such as imperfect information and real-time environments. I plan to extend this work using the domain of real-time strategy (RTS) games, which have gained recent popularity as a challenging and complex domain for AI research.

Predicting Opponent's Production in Real-Time Strategy Games With Answer Set Programming

The adversarial character of real-time strategy (RTS) games is one of the main sources of uncertainty within this domain. Since players lack exact knowledge about their opponent's actions, they need a reasonable representation of alternative possibilities and their likelihood. In this article we propose a method of predicting the most probable combination of units produced by the opponent during a certain time period. We employ a logic programming paradigm called Answer Set Programming, since its semantics is well suited for reasoning with uncertainty and incomplete knowledge. In contrast with typical, purely probabilistic approaches, the presented method takes into account the background knowledge about the game and only considers the combinations that are consistent with the game mechanics and with the player's partial observations. Experiments, conducted during different phases of StarCraft: Brood War and Warcraft III: The Frozen Throne games, show that the prediction accuracy for time intervals of 1–3 min seems to be surprisingly high, making the method useful in practice. Root-mean-square error grows only slowly with increasing prediction intervals—almost in a linear fashion.

Applying thematic analysis to define an awareness interpretation for collaborative computer games

ContextCollaborative computer games have evolved from single-player to massively multiplayer awareness-demanding games, usually involving collaboration to achieve team goals. As a consequence of such evolution, these players should be provided with awareness information that enables them to perform collaborative tasks with other team members. ObjectiveThe objective of this work is the analysis of current awareness interpretations in order to develop an awareness interpretation that collects the awareness needs of such games. MethodThis analysis has been conducted by means of a step-by-step Thematic Analysis of current interpretations that led us to extract the most relevant awareness elements defined in existing interpretations. The developed awareness interpretation was empirically evaluated by means of several surveys aimed at assessing whether the implementation of the interpretation elements in a game would improve the players enjoyment. ResultsThe Thematic Synthesis Analysis concluded that none of the current awareness interpretations can deal properly with collaborative computer games, specifically due to collaboration and social & group dynamics. This Thematic Synthesis Analysis led us to coin Gamespace Awareness, a new awareness interpretation based on a combination of the previously analyzed awareness interpretations, which is suitable for collaborative computer games. The interpretation was positively evaluated for two games, namely a first person shooter and a real-time strategy game. ConclusionsGamespace Awareness combines the potential awareness elements needed for collaborative computer games, making it possible to identify the awareness requirements of these games from the very beginning.

A Pattern Mining Approach to Study Strategy Balance in RTS Games

Whereas purest strategic games such as Go and Chess seem timeless, the lifetime of a video game is short, influenced by popular culture, trends, boredom, and technological innovations. Even the important budget and developments allocated by editors cannot guarantee a timeless success. Instead, novelties and corrections are proposed to extend an inevitably bounded lifetime. Novelties can unexpectedly break the balance of a game, as players can discover unbalanced strategies that developers did not take into account. In the new context of electronic sports, an important challenge is to be able to detect game balance issues. In this paper, we consider real-time strategy (RTS) games and present an efficient pattern mining algorithm as a basic tool for game balance designers that enables one to search for unbalanced strategies in historical data through a knowledge discovery in databases (KDD) process. We experiment with our algorithm on StarCraft II historical data, played professionally as an electronic sport.

A Semi-Markov Decision Model for Recognizing the Destination of a Maneuvering Agent in Real Time Strategy Games

Recognizing destinations of a maneuvering agent is important in real time strategy games. Because finding path in an uncertain environment is essentially a sequential decision problem, we can model the maneuvering process by the Markov decision process (MDP). However, the MDP does not define an action duration. In this paper, we propose a novel semi-Markov decision model (SMDM). In the SMDM, the destination is regarded as a hidden state, which affects selection of an action; the action is affiliated with a duration variable, which indicates whether the action is completed. We also exploit a Rao-Blackwellised particle filter (RBPF) for inference under the dynamic Bayesian network structure of the SMDM. In experiments, we simulate agents’ maneuvering in a combat field and employ agents’ traces to evaluate the performance of our method. The results show that the SMDM outperforms another extension of the MDP in terms of precision, recall, andF-measure. Destinations are recognized efficiently by our method no matter whether they are changed or not. Additionally, the RBPF infer destinations with smaller variance and less time than the SPF. The average failure rates of the RBPF are lower when the number of particles is not enough.

Enhancing Video Games Policy Based on Least-Squares Continuous Action Policy Iteration: Case Study on StarCraft Brood War and Glest RTS Games and the 8 Queens Board Game

With the rapid advent of video games recently and the increasing numbers of players and gamers, only a tough game with high policy, actions, and tactics survives. How the game responds to opponent actions is the key issue of popular games. Many algorithms were proposed to solve this problem such as Least-Squares Policy Iteration (LSPI) and State-Action-Reward-State-Action (SARSA) but they mainly depend on discrete actions, while agents in such a setting have to learn from the consequences of their continuous actions, in order to maximize the total reward over time. So in this paper we proposed a new algorithm based on LSPI called Least-Squares Continuous Action Policy Iteration (LSCAPI). The LSCAPI was implemented and tested on three different games: one board game, the 8 Queens, and two real-time strategy (RTS) games, StarCraft Brood War and Glest. The LSCAPI evaluation proved superiority over LSPI in time, policy learning ability, and effectiveness.

Hybrid Pathfinding in StarCraft

Micromanagement is a very important aspect of real-time strategy (RTS) games. It involves moving single units or groups of units effectively on the battle field, targeting the most threatening enemy units and use the unit's special abilities when they are the most harmful for the enemy or the most beneficial for the player. Designing good micromanagement is a challenging task for AI bot developers. In this paper, we address the micromanagement subtask of positioning units effectively in combat situations. Two different approaches are evaluated, one based on potential fields and the other based on flocking algorithms. The results show that both the potential fields version and the flocking version clearly increases the win percentage of the bot, but the difference in wins between the two is minimal. The results also show that the more flexible potential fields technique requires much more hardware resources than the more simple flocking technique.

Correlations among Social Anxiety, Self-Esteem, Impulsivity, and Game Genre in Patients with Problematic Online Game Playing.

ObjectiveRecent studies of online game addiction have suggested that social interaction and impulsivity are critical factors for the etiology and progress of online game addiction. We hypothesized that the genre of the online game is associated with impulsivity and sociality in individuals with online game addictions.MethodsIn total, 212 patients with problematic online game playing were divided into four groups by game genre: 1) massive multiplayer online role playing game (MMORPG), 2) real-time strategy (RTS), 3) first-person shooter (FPS), and 4) other. Their symptoms and characteristics were assessed using 8 scales and 2 tests to estimate self-esteem, impulsiveness, comorbidity, social interaction status, and cognitive function.ResultsThe mean social anxiety score was highest in the MMORPG group and lowest in the FPS group. The mean self-esteem score was highest in the RTS group. Social anxiety score was positively correlated with Internet addiction score in the MMORPG group, and the self-esteem score was positively correlated with Internet addiction score in the RTS group.ConclusionThe genre of online game was not associated with impulsivity, but social anxiety status varied significantly with game genre, and differences in social anxiety were especially pronounced in patients playing the MMORPG (highest social anxiety) and FPS (lowest social anxiety) game genres. In addition, self-esteem was highest in the RTS game genre.

Influence of head-up displays’ characteristics on user experience in video games

Head-up displays (HUD) are important parts of visual interfaces of virtual environments such as video games. However, few studies have investigated their role in player–video game interactions. Two experiments were designed to investigate the influence of HUDs on player experience according to player expertise and game genre. Experiment 1 used eye-tracking and interviews to understand how and to what extent players use and experience HUDs in two types of commercial games: first-person shooter and real-time strategy games. Results showed that displaying a permanent HUD within the visual interface may improve the understanding of this environment by players. They also revealed that two HUD characteristics, namely composition and spatial organization, have particular influence on player experience. These critical characteristics were manipulated in experiment 2 to study more precisely the influence of HUD design choices on player experience. Results showed that manipulation of design of these HUD characteristics influences player experience in different ways according to player expertise and game genre. For games with HUDs that are perceived as very useful, the higher player expertise is, the more player experience is influenced. Recommendations for video game design based on these results are proposed.

SELF-EVALUATION OF RTS TROOP’S PERFORMANCE

This paper demonstrates the research results obtained from a comparison of Evolutionary Programming (EP) and hybrid Differential Evolution (DE) and Feed Forward Neural Network (FFNN) algorithms in the Real Time Strategy (RTS) computer game, namely Warcraft III. The main aims of this research are to: test the feasibility of implementing EP and hybrid DE into RTS game, compare the performances of EP and hybrid DE, and generate gaming RTS controllers autonomously, an issue primarily of reinforcement/troops balancing. This micromanagement issue has been overlooked since last decade. Experimental results demonstrate success with all aims: both EP and hybrid DE could be implemented into the Warcraft III platform, and both algorithms used able to generate optimal solutions.

Techno-orientalism: imagining Asia in speculative Fiction, History, and Media ed. by David s. Roh, Betsy Huang, and Greta A. Niu (review)

Reviewed by: Techno-orientalism: imagining Asia in speculative Fiction, History, and Media ed. by David s. Roh, Betsy Huang, and Greta A. Niu John Cheng (bio) Techno-orientalism: imagining Asia in speculative Fiction, History, and Media, edited by David s. Roh, Betsy Huang, and Greta A. Niu. New Brunswick, N.J.: Rutgers University Press, 2015. x + 260 pp. $34.95 paper. ISBN: 978-0-8135-7063-1. This anthology usefully introduces an emergent area of Asian American literary and cultural criticism that seeks to bridge what are often separate realms and discourses of analysis: Asian American studies and critical race studies on the one hand and speculative fiction studies and media studies on the other. Its editors explore aspects of what they, drawing from a 1995 David Morley and Kevin Robins essay, call techno-Orientalism: “the phenomenon of imagining Asia and Asians in hypo- or hypertechnological terms in cultural productions and political discourse” (2). This formulation extends Edward Said’s notion of Orientalism to the technological modern and how Asia, for the West, contains and configures its associated anxieties. While “Orientalism arrests Asia in traditional, and often premodern imagery,” the editors explain, “techno-Orientalism presents a broader, dynamic, and often contradictory spectrum of images, [End Page 382] constructed by the East and West alike, of an ‘Orient’ undergoing rapid economic and cultural transformations” (3). With its particular concern to imagine future possibility and temporal alternative, speculative fiction (or sf, as it is abbreviated) richly expresses these tensions in several media and forms; still, the editors argue, much of sf’s critical tradition has not applied Orientalism’s lens to technology’s sublime. This volume seeks to redress that critical vacuum. The collection’s two sections parallel the editors’ dual aims: “Iterations and Instantiations” charts examples and aspects of techno-Orientalism “over time and across genres” (16), while “Reappropriations and Recuperations” offers opportunities for “critical reappropriations in texts that self-referentially engage with Asian images” (7). With an introduction, conclusion, and fifteen contributor essays within two hundred twenty-five pages, most of its pieces are brief; consequently, some are largely descriptive. Many of the essays are, nevertheless, useful interventions. Several authors chart new, less obvious territory connecting the technological and the Asian/Orient. Aimee Bahng explores the “cruel optimism”—using Lauren Berlant’s notion—of Asian futures forecast through an economics founded on neoliberal financial speculation in Sonny Liew’s graphic novel Malinky Robot. Warren Liu counterposes the technological order of clocks and automata with the natural evolutionary order of geologic strata to excavate the “queer” futures and asynchronous dynamics that Asian racial traces reveal within the apparent temporal stability of William Gibson and Bruce Sterling’s foundational steampunk novel The Difference Engine. Other contributors recover historical examples of this same linkage: Kenneth Hough on the sublime combination of war, engines, and the modern in the formation of turn-of-the-twentieth-century fears of Japanese invasion; Jason Crum on authoritative knowledge about the Orient and its industrial production for mass audiences in late 1930s Mercury Theatre Company radio broadcasts. Still others examine Orientalist themes within emergent technological entertainment. Steve Choe and Se Young Kim consider their presence in discourses about video gaming and gamer culture, specifically the real-time strategy (RTS) game StarCraft, that connect Asian “gamer death” to broader assumptions about what constitutes “fair play”; Dylan Yeats delves into the militaristic link of Asian subjects/objects within Hollywood films, first-person shooter (FPS) video games, and actual military recruitment. While individual pieces assay the subject and concerns the editors introduce, as a whole the collection does not fully achieve their expressed aims. In actual geographic coverage, the “Asia” imagined from its subtitle defaults to East Asia. The editors mention India and South Asians in their introduction and conclusion, but none of the contributors take South or Southeast Asia—with the exception of Bahng and brief references to sites of American conflict—or authors from those regions as their subjects. How might they have considered [End Page 383] the reappropriately imagined Philippines, Viet Nam, and India in the sf of Charles Tan, Vu Kim Dung, or Amitav Ghosh, for instance? Similarly the essays might have more directly engaged scholarship in...

Area of Simulation

Although Multi-Avatar Distributed Virtual Environments (MAVEs) such as Real-Time Strategy (RTS) games entertain daily hundreds of millions of online players, their current designs do not scale. For example, even popular RTS games such as the StarCraft series support in a single game instance only up to 16 players and only a few hundreds of avatars loosely controlled by these players, which is a consequence of the Event-Based Lockstep Simulation (EBLS) scalability mechanism they employ. Through empirical analysis, we show that a single Area of Interest (AoI), which is a scalability mechanism that is sufficient for single-avatar virtual environments (such as Role-Playing Games), also cannot meet the scalability demands of MAVEs. To enable scalable MAVEs, in this work we propose Area of Simulation (AoS), a new scalability mechanism, which combines and extends the mechanisms of AoI and EBLS. Unlike traditional AoI approaches, which employ only update-based operational models, our AoS mechanism uses both event-based and update-based operational models to manage not single, but multiple areas of interest. Unlike EBLS, which is traditionally used to synchronize the entire virtual world, our AoS mechanism synchronizes only selected areas of the virtual world. We further design an AoS-based architecture, which is able to use both our AoS and traditional AoI mechanisms simultaneously, dynamically trading-off consistency guarantees for scalability. We implement and deploy this architecture and we demonstrate that it can operate with an order of magnitude more avatars and a larger virtual world without exceeding the resource capacity of players' computers.

Cognitive Load in Real-Time Strategy Gaming: Human Opponent Versus AI Opponent

This paper examines the cognitive load in real-time strategy gaming subjects along with physiological and subjective measurements. The subjects were college students from a northeastern university in China. This paper is a study into players’ cognitive load measurement when playing against human opponents and game artificial intelligences (game AIs). Using eye trackers, action per minute recorders and questionnaires, we collected qualitative data from participants about their cognitive load through playing against human opponents and game AI opponents. The results indicate that players have more cognitive load when playing against human opponents than against game AI opponents. Secondly, as the players’ gaming skill increased, their cognitive load did not differ substantially when they switched from playing against human players to playing against game AI opponents. A deeper understanding of players’ cognitive load can enable a developer to design a more appropriate game experience.

A fast evaluation method for RTS game strategy using fuzzy extreme learning machine

This paper proposes a fast learning method for fuzzy measure determination named fuzzy extreme learning machine (FELM). Moreover, we apply it to a special application domain, which is known as unit combination strategy evaluation in real time strategy (RTS) game. The contribution of this paper includes three aspects. First, we describe feature interaction among different unit types by fuzzy theory. Second, we develop a new set selection algorithm to represent the complex relation between input and hidden layers in extreme learning machine, in order to enable it to learn different fuzzy integrals. Finally, based on the set selection algorithm, we propose the FELM model for feature interaction description, which has an extremely fast learning speed. Experimental results on artificial benchmarks and real RTS game data show the feasibility and effectiveness of the proposed method in both accuracy and efficiency.