Modified Adversarial Hierarchical Task Network Planning in Real-Time Strategy Games

@article{Sun2017ModifiedAH,
  title={Modified Adversarial Hierarchical Task Network Planning in Real-Time Strategy Games},
  author={Lin Sun and Peng Jiao and Kai Xu and Quanjun Yin and Yabing Zha},
  journal={Applied Sciences},
  year={2017},
  volume={7},
  pages={872}
}
The application of artificial intelligence (AI) to real-time strategy (RTS) games includes considerable challenges due to the very large state spaces and branching factors, limited decision times, and dynamic adversarial environments involved. To address these challenges, hierarchical task network (HTN) planning has been extended to develop a method denoted as adversarial HTN (AHTN), and this method has achieved favorable performance. However, the HTN description employed cannot express complex… 
View PDF
9 Citations
Methods Citations
1

9 Citations

HTN Guided Adversarial Planning for RTS Games

This paper proposes a novel method which considers the time constraint when generating the subgame tree by HTN planning and introduces the opponent strategy as the adversarial player and shows the algorithm performs better than originalHTN planning methods.

HISTORY-BASED SINGLE BELIEF STATE GENERATION FOR PARTIALLY OBSERVABLE REAL-TIME STRATEGY GAMES

This paper focuses on extending Hierarchical Task Network (HTN) Planning in partially observable environment, and proposes a partially observable adversarial hierarchical task network planning with repairing algorithm named PO-AHTNR.

HISTORY-BASED SINGLE BELIEF STATE GENERATION FOR PARTIALLY OBSERVABLE REAL-TIME STRATEGY GAMES

This paper focuses on extending Hierarchical Task Network (HTN) Planning in partially observable environment, and proposes a partially observable adversarial hierarchical task network planning with repairing algorithm named PO-AHTNR.

Online Adversarial Planning in μRTS : A Survey

The unique characteristics of μRTS helped for the emergence of interesting online adversarial planning techniques, dealing with multiple levels of abstraction, in fully and partially observable environments.

Artificial Intelligence Techniques on Real-time Strategy Games

The purpose of this paper is to systematically review the artificial intelligence technologies used in R TS games in recent years, including the definition of RTS games, the challenges faced, the platform for research problems, and the artificial Intelligence methods for problem solving.

A Rule-Based AI Method for an Agent Playing Big Two

The experimental results show that the proposed rule-based AI agent can play Big Two well and outperform randomized AI, conventional AI, and human players, with the capability of maximizing the winning score and minimizing the number of cards left when the chance of winning is low.

Automatic hierarchical task network planning system for the Unity game engine

A graphical notation for domain knowledge definition was developed that will allow quick complex behaviour pattern implementation for non-player characters within the game environment.

Evolving interpretable strategies for zero-sum games

AI TECHNIQUES IN RTS GAMES 3 . 1 Classification of RTS games

  • 2018

44 References

Adversarial Hierarchical-Task Network Planning for Complex Real-Time Games

This paper presents an alternative approach to adversarial Hierarchical Task Network planning that combines ideas from game tree search with HTN planning, and presents empirical results for the µRTS game, comparing it to other state of the art search algorithms for RTS games.

Learning HTN Method Preconditions and Action Models from Partial Observations

This paper presents a formal framework and algorithms to acquire HTN planning domain knowledge, by learning the preconditions and effects of actions and precond conditions of methods, and presents the HTN-learner algorithm, which first builds constraints from given observed decomposition trees to build action models and method precondition.

A Repair-Replanning Strategy for HTN-based Therapy Planning Systems ∗

This work presents a strategy to respond to failures during the execution of a treatment plan in the real world, focused on HTN-based therapy planning frameworks, that performs an exception analysis episode to identify both the nature and the complexity of the unexpected event.

HOTRiDE: Hierarchical Ordered Task Replanning in Dynamic Environments

A planning system, called HOTRiDE (Hierarchical Ordered Task Replanning in Dynamic Environments), which interleaves plan generation, execution, and repair in order to work under such circumstances.

UCT for Tactical Assault Planning in Real-Time Strategy Games

This paper investigates the use of UCT, a recent Monte-Carlo planning algorithm for tactical assault planning in real-time strategy games, and presents an evaluation of the approach on a range of tactical assault problems with different objectives in the RTS game Wargus.

Portfolio greedy search and simulation for large-scale combat in starcraft

This paper presents an efficient system for modelling abstract RTS combat called SparCraft, which can perform millions of unit actions per second and visualize them, and presents a modification of the UCT algorithm capable of performing search in games with simultaneous and durative actions.

HTN planning: Overview, comparison, and beyond

A hierarchical task network planner for pathfinding in real-time strategy games

Results show that the hierarchical approach reduces the suboptimality of the LRTA and speeds up the convergence process.

Labeled RTDP: Improving the Convergence of Real-Time Dynamic Programming

This paper introduces a labeling scheme into RTDP that speeds up its convergence while retaining its good anytime behavior, and shows that Labeled RTDP (LRTDP) converges orders of magnitude faster than RTDP, and faster also than another recent heuristic-search DP algorithm, LAO*.

Hierarchical Task Network Plan Reuse for video games

An approach for Plan Reuse is proposed that manipulates the order in which the search tree is traversed by using a similarity function and is tested in the SimpleFPS domain, and shown to be capable of finding (optimal) plans with a decreased amount of search effort on average when re-planning for variations of previously solved problems.