ArticlePDF Available

Capture the Flag: Military Simulation Meets Computer Games

Authors:
Marc S. Atkin
Marc S. Atkin
Marc S. Atkin
  • This person is not on ResearchGate, or hasn't claimed this research yet.
David Westbrook at University of Massachusetts Amherst
David Westbrook
Paul R. Cohen at University of Arizona
Paul R. Cohen
Download full-text PDF
Read full-text
Download citation

Abstract and Figures

Some of the more complex AI testbeds are not that different from computer games. It seems that both sides, AI and game design, could profit from each other's technology. We go a first step in this direction by presenting a very general agentcontrol architecture (HAC: Hierarchical Agent Control), a toolkit for designing an action hierarchy. It supports action abstraction, a multi-level computational architecture, sensor integration, and planning. It is particularly well suited to controlling large numbers of agents in dynamic environments.
Figures - uploaded by David Westbrook
Author content
All figure content in this area was uploaded by David Westbrook
Content may be subject to copyright.
Content uploaded by David Westbrook
Author content
All content in this area was uploaded by David Westbrook on Oct 23, 2012
Content may be subject to copyright.
Follow
Move
Apply
Force
Errors and Sensor Data
Control
Physical
Schemas
Primitive
Actions
Domain
Specific
Actions
Move
Mobile Offense
Move-To-Point
Harass
Attack
... The proposed methods are studied and demonstrated through three applications of increasing complexity. The first application consists of two problems well known in the context of multi-agent systems, foraging [28,33,65] and capture the flag [2,6]. Both scenarios require, independently, the cooperation and repeated interaction of mobile agents with limited sensing and acting capabilities, which makes these problems useful in the study of different issues in systems involving cooperative agents. ...
... In this sense the task is similar to a generalized version of the robotic capturethe-flag game [6], with minerals representing a multitude of flags for a longer, iterated game, as opposed to the one-shot version of capturing a single flag once and for all. in an idle state. Once they receive a "start signal", they begin searching for resources. ...
View
... CADET focuses on course of action analysis, which is the most timeconsuming aspect of the Military Decision Making Process. Other researchers (e.g., [8, 9, 10, 4]) are addressing a different (and preceding) phase of the process, the very interesting and challenging problem of generating the high-level maneuver COA. In addressing the style of interactions between the human and the decision aid, we de-emphasize the mixed-initiative, incremental style (even though CADET allows such a style) in favor of a rapid style of generating a complete plan from a high-level COA, followed by manual modifications. ...
Decision Aids for Adversarial Planning in Military Operations: Algorithms, Tools, and Turing-test-like Experimental Validation
Article
  • Jan 2016
Use of intelligent decision aids can help alleviate the challenges of planning complex operations. We describe integrated algorithms, and a tool capable of translating a high-level concept for a tactical military operation into a fully detailed, actionable plan, producing automatically (or with human guidance) plans with realistic degree of detail and of human-like quality. Tight interleaving of several algorithms -- planning, adversary estimates, scheduling, routing, attrition and consumption estimates -- comprise the computational approach of this tool. Although originally developed for Army large-unit operations, the technology is generic and also applies to a number of other domains, particularly in critical situations requiring detailed planning within a constrained period of time. In this paper, we focus particularly on the engineering tradeoffs in the design of the tool. In an experimental evaluation, reminiscent of the Turing test, the tool's performance compared favorably with human planners.
View
... of the complexity of large-scale multiagent coordination, we typically choose four typical application domains: urban search and rescue (USAR), controlling wide area search munitions (WASMs) [31], RoboCup [32], and large agent team in strategy game and scheduling [33]. For example, in USAR, coordination is mainly focused on heterogeneous teams and there are only one or two tasks for each of plan templates. ...
Modeling and Simulation of Complex Network Attributes on Coordinating Large Multiagent System
Article
Full-text available
With the expansion of distributed multiagent systems, traditional coordination strategy becomes a severe bottleneck when the system scales up to hundreds of agents. The key challenge is that in typical large multiagent systems, sparsely distributed agents can only communicate directly with very few others and the network is typically modeled as an adaptive complex network. In this paper, we present simulation testbed CoordSim built to model the coordination of network centric multiagent systems. Based on the token-based strategy, the coordination can be built as a communication decision problem that agents make decisions to target communications and pass them over to the capable agents who will potentially benefit the team most. We have theoretically analyzed that the characters of complex network make a significant difference with both random and intelligent coordination strategies, which may contribute to future multiagent algorithm design.
View
... HAC [1] is a framework for building agents that act in dynamic environments. HAC, among others, deals with the problems of reacting to changing environments in a timely manner, integrating reactive and cognitive processes to achieve abstract goals and interleaving planning with execution. ...
ICAGENT: Balancing between reactivity and deliberation
Article
Full-text available
The aim of this paper is to present a framework,for developing intelligent agents that act in dynamic,and unpredictable environments,by balancing between deliberation and reaction. Acting reactively, an agent executes routine tasks without considering its options and thinking about alternatives. Acting deliberatively, an agent deals with complex goals that need careful, although not detailed planning and reconciles its desires and its intentions. This paper emphasizes,on,the utility of plan management techniques, as well as on the integrated treatment of agents’ mental structures during reaction and deliberation. The framework,is thoroughly explained and the behaviour of an agent in a specific environment is examined.
View
... Combat server enables Human players and/or CGF software to control combat force, which play against each other within the simulation environment [2]. Such an environment should support generation of combat forces (Blue and Red) with units having configurable number of personal & ammunition, ranges for sensing and firing weapon. ...
Engineering Emergent behavior in Computer Generated Forces (CGF) using Multi-Agent System
Article
Full-text available
Over the previous decade, there has been a significant amount of work done on the development of intelligent, Computer Generated Forces (CGF) capable of combat behavior within synthetic battlefields. One of the pioneering efforts is credited to Ilachinski who developed ISAAC (Irreducible Semi-Autonomous Adaptive Combat) in 1997. The central thesis of ISAAC is that land combat can be thought of as a complex adaptive system - combat forces are composed of large numbers of nonlinearly interacting parts and are organized in a command and control hierarchy. ISAAC is intended to be used as a conceptual playground in which to explore high-level emergent behaviors arising from various low-level interaction rules. However in such a system it is very hard to engineer the emergent behavior. This paper attempts to develop an approach that tries to significantly reduce laborious trial and error experimentation usually required to engineer the emergent behavior. A prototype of synthetic battlefield environment and a generic Multi-Agent based CGF has been built to demonstrate the approach. A toolkit named JACK has been used for implementation.
View
Enlivening Port Scanning Exercises with Capture the Flag and Deduction
Chapter
  • Sep 2020
Designing engaging exercises when students do not yet possess a lot of knowledge can be difficult. We show how we draw on students’ prior knowledge, along with basic introductory concepts, to design an elemental (but fun) port scan exercise in an introductory security testing module. While “capture the flag” is a security industry standard for exercises, it can require a lot of in-depth knowledge to properly implement and complete. Using basic computer science concepts such as ports and ASCII values, we design a simplified capture the flag exercise where students can make use of deductive reasoning to complete the game. Overall, the exercise was received favourably by the students who found it challenging but enriching.
View
View
KSCO 2002: Second International Conference on Knowledge Systems for Coalition Operations
Article
  • May 2002
This document contains the final proceedings for the interdisciplinary conference KSCO-2002 which covered the following topics: Innovative theory and techniques for coalition formation and "virtual organizations"; Applications and requirements for knowledge-based coalition planning and operations; Knowledge-based approaches to command and control; Knowledge-based approaches to coalition logistics; Applications and requirements for knowledge-based coalition planning; Knowledge-based approaches to OOTW (peace keeping, disaster relief or other humanitarian operations); Multi-agent systems and the concept of agency in coalitions; Tools and techniques for knowledge-based simulation and modeling of coalition operations; Security and maintenance of private information or knowledge in coalition operations; Autonomous vs. centrally managed coalition operations.
View
Course of Action Simulation, Testing, Evaluation and Revision (COASTER)
Article
Full-text available
  • Jan 2004
This effort developed simulation, visualization and evaluation technology for air campaign planning. A simple war game called Capture the Flag was used to build and evaluate military Courses of Action based on the developed technology. The Capture the Flag system comprises a simulator, a control architecture, and a planner. The system provides graphical user interfaces for users to observe and participate in war games. A variety of visualizations, instrumentation, and other software support the execution and analysis of experiments. Dynamic maps were developed to visualize the results of hundreds of Monte Carlo simulation trials. This visualization technique, referred to as probabilistic dynamic mapping, relates dynamical properties of a war game to the probability of winning the game. A method for unsupervised clustering of univariate, categorical time series was also developed. This method is referred to as Bayesian Clustering by Dynamics.
View
Anticipatory Planning With Agents Using Genetic Algorithms and Simulation
Article
  • May 2001
The traditional Military Decision Making Process (MDMP) focuses on developing a few friendly Courses of Action (COAs) against the "most-likely and most-dangerous" enemy COAs. Military planners need a way to incorporate the continuous feed of battle information into the planning process so that they achieve and maintain "option dominance". A new approach to military operations, called Anticipatory Planning and Adaptive Execution, treats planning and execution as a tightly coupled, single process, and replaces reaction to events with anticipation of events. This research develops the methodology for automating the Anticipatory Planning process. A prototype Anticipatory Planning Support System (APSS) has been designed and implemented to provide human planners with an interactive visual development system using simulations to build Plan Descriptions. The primary goals of this implementation are to provide a common representation of the plan, facilitate the planning process, anticipate the flow of the battle, and provide a means for stimulating planning systems.
View
Reactive Reasoning and Planning.
Conference Paper
Full-text available
  • Jan 1987
In this paper, the reasoning and planning capabilities of an au- tonomous mobile robot are described; The reasoning system that controls the robot is designed to exhibit the kind of be- havior expected of a rational agent, and is endowed with the psychological attitudes of belief, desire, and intention. Because these attitudes are explicitly represented, they can be manipu- lated and reasoned about, resulting in complex goal-directed and reflective behaviors. Unlike most planning systems, the plans or intentions formed by the robot need only be partly elaborated before it decides to act. This allows the robot to avoid overly strong expectations about the environment, overly constrained plans of action, and other forms of overcommitment common to previous planners. In addition, the robot is continuously re- active and has the ability to change its goals and intentions as situations warrant. The system has been tested with SRI's au- tonomous robot (Flakey) in a space station scenario involving navigation and the performance of emergency tasks.
View
View
Lanchester Attrition Processes and Theater-Level Combat Models
Article
  • Sep 1981
This paper is an expository survey of some aspects of the theory of Lanchester attrition processes and of applications of these processes in models of theater-level combat. Classical Lanchester differentials-including homogeneous and heterogeneous square and linear laws-are discussed. Underlying assumptions and mathematical forms of analogous continuous time, discrete state space Markov attrition processes are presented. A detailed treatment of a class of simplified ('binomial') attrition equations is given, including underlying assumptions, exact equations for expected attrition and exponential approximations to the exact equations. Applications of the binomial and Markov attrition processes as the basis for attrition calculations in the CONAF Evaluation Model, the IDAGAM I Model, the Lulejian-I Model and the VECTOR-2 Model (4 principal documented models of theater-level combat) are considered.
View
Trial by Fire: Understanding the Design Requirements for Agents in Complex Environments.
Article
  • Sep 1989
  • AI MAG
The authors describe Phoenix from the perspective of their technical aims and motives. They describe the Phoenix task - controlling simulated forest fires - and explain why they use a simulated environment instead of a real, physical one. They discuss the characteristics of the forest fire environment and the constraints they place on the design of agents. They illustrate Phoenix agents controlling a forest fire.
View
A Simulation Substrate for Real- Time Planning
Article
  • Aug 1997
Real-time planning is, generally speaking, problem-solving under time-pressure. In order to test and evaluate real-time planners, scientists need to have environments that pose problems to be solved. This dissertation describes a substrate, called Mess, for building environments and real-time planners. Mess relies on a second system, TCL, to replace CPU time as a way to measure the amount of thinking done by a real-time agent. The key issue with real-time planning simulators is that some of the simulation processes represent real-world events and other processes represent thinking or planning that occurs while the real-world events do. Therefore, there needs to be a correspondence between real-world events and thinking processes.
View
Physical planning and dynamics
  • Jan 1998
  • 4-9
  • M S Atkin
  • P R Cohen
Atkin, M. S., and Cohen, P. R. 1998. Physical planning and dynamics. In Working Notes of the AAAI Fall Symposium on Distributed Continual Planning, 4{9.
AFS and HAC: Domain-general agent simulation and control
  • Jan 1998
  • M S Atkin
  • D L Westbrook
  • P R Cohen
  • G D Jorstad
Atkin, M. S. Westbrook, D. L. Cohen, P. R. and Jorstad, G. D. 1998. AFS and HAC: Domain-general agent simulation and control. In Working Notes of the Workshop on Software Tools for Developing Agents, AAAI-98, 89{95.
  • Jan 1986
  • 1383-1398
  • M P George
  • A L Lansky
George, M. P., and Lansky, A. L. 1986. Procedural knowledge. Proceedings of the IEEE Special Issue on Knowledge Representation 74(10):1383{1398.
Proceedings of The Second International Conference o n A rti cial Intelligence Planning Systems
  • Jan 1994
  • 158-163
  • L Spector
  • J Hendler
Spector, L., and Hendler, J. 1994. The use of supervenience in dynamic-world planning. In Hammond, K., ed., Proceedings of The Second International Conference o n A rti cial Intelligence Planning Systems, 158{ 163.