M. Tambe, W. Johnson, R. Jones, F. Koss, J. Laird, P. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, page 16(1), 1995.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....and the state of its reasoning process. 1 Introduction Intelligent agents are used in a wide variety of simulation environments where they are expected to exhibit behavior similar to that of a human in the same situation. Examples of such environments include RoboCup[10] air combat simulations [14] and virtual theater[16] Defining agents for simulation environments is a very active research area. The research has resulted in a large number of agent architectures being proposed. Many of the proposed architectures have accompanying languages for defining the behaviors, for example [1, 3, 5, ....

Milind Tambe, W. Lewis Johnson, Randolph Jones, Frank Koss, John Laird, Paul Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....was observed was that an ecient parallel development approach evolved. 1 Introduction Interactive simulation environments with large numbers of intelligent agents are becoming increasingly common. Such simulation environments often have agents playing the roles of humans within the simulation[13]. Examples of this type of environment are air combat simulation[12] disaster management simulation, computer games[3] and RoboCup[7] In general, knowledge of precisely what agents should do in the environment is not an agent developer s area of expertise, rather it is a domain expert s ....

Milind Tambe, W. Lewis Johnson, Randolph Jones, Frank Koss, John Laird, Paul Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15-39, Spring 1995.

....observations of the other agents. An agent therefore needs to monitor the other agents. Observation based coordination (OBC) is a key challenge to the multi agent and multi robot systems. Increasingly, robots and synthetic agents are being deployed in multi agent virtual environments for training [1] and entertainment [2] robotic soccer [3] hazardous cleanup tasks [4] formation maintenance tasks [5,6] and more. Many of these applications rely on agents to coordinate with one another based on their observations of each other [7] OBC is often a challenging process, mainly because it is ....

....process from the literature to illustrate its meaning. 3. 1 Examples from the literature These examples are taken from the ModSAF domain, a high fidelity virtual environment for military training that allows thousands of agents (synthetic and human) to interact in battlefield scenarios [1]. Several coordination scenarios for synthetic helicopter pilots in this domain have been described in [11,8] These rise from the scenario described below. A team of 3 6 helicopters take off from their base and fly in formation (with the fly flight plan (f ) behavior) until they reach an area ....

Milind Tambe, W. Lewis Johnson, Randy Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....military training. Cognitive models as intelligent agents can populate synthetic environments representing some or all of the entities involved in real combats, thus enabling the use of realistic environments for training purposes [14, 19] One such attempt has been the TacAir Soar system [24] which employs cognitive models developed with the Soar cognitive architecture [7, 9] to simulate the behavior of military personnel in fixed wing aircraft missions. The benefits of using TacAir Soar are particularly evident in In Proceedings of the 11th Computer Generated Forces and Behavior ....

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S., and Schwamb, K.: Intelligent agents for interactive simulation environments. AI Magazine, Vol. 16, pp. 15-40, 1995.

....observations of the other agents. An agent therefore needs to monitor the other agents. Observation based coordination (OBC) is a key challenge to the multi agent and multi robot systems. Increasingly, robots and synthetic agents are being deployed in multi agent virtual environments for training [1] and entertainment [2] robotic soccer [3] hazardous cleanup tasks [4] formation maintenance tasks [5,6] and more. Many of these applications rely on agents to coordinate with one another based on their observations of each other [7] OBC is often a challenging process, mainly because it is ....

....process from the literature to illustrate its meaning. 3. 1 Examples from the literature These examples are taken from the ModSAF domain, a high fidelity virtual environment for military training that allows thousands of agents (synthetic and human) to interact in battlefield scenarios [1]. Several coordination scenarios for synthetic helicopter pilots in this domain have been described in [11,8] These rise from the scenario described below. A team of 3 6 helicopters take off from their base and fly in formation (with the fly flight plan ( behavior) until they reach an area ....

Milind Tambe, W. Lewis Johnson, Randy Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....Soar [19] ACT R [2] and EPIC [18] are the ones used more often to guide the construction of cognitive models. JACK is also being increasingly used [6] In some cases, models based on cognitive architectures were built for use for training purposes in military simulations. The TacAir Soar system [35] is a notable example. In TacAir Soar, cognitive models developed with the Soar cognitive architecture [22, 17] simulate the behavior of military personnel in fixed wing aircraft missions. TacAir Soar was successfully used in Stow 97, a large scale simulation exercise in which up to 3,700 ....

Tambe, M., Johnson, W. L, Jones, R., M, Koss, F., V, Laird, J., E, Rosenbloom, P. S, and Schwamb, K., B.: Intelligent agents for interactive simulation environments. AI Magazine, Vol. 16, pp. 15-39, 1995.

....detect failures, coordinate, and collaborate. Indeed, the importance of agent monitoring in deployed multi agent systems has long been recognized in theory (e.g. 4, 7, 8] and in practice, ranging from industrial systems (e.g. 12] to virtual environments for training and research (e.g. [21, 22]) to human computer interaction (e.g. 16] and multi agent robotics (e.g. 18, 2] Agent monitoring infrastructure is of particular importance in teams of cooperating agents, since the correct execution of teamwork mandates that team members come to agree on the task that is jointly executed ....

.... these challenges while working on developing robust multi agent teams in two dynamic, complex, domains: ModSAF, a commercially developed, high fidelity virtual environment, where we have been involved in the development of synthetic helicopter pilot agents that carry out a variety of missions [21]) and RoboCup soccer simulation, a dynamic research oriented simulation which requires real time teamwork and coordination, where we have been involved in the development of both soccer playing agents and a coach agent [22] Monitoring Algorithm Complexity As discussed above, agents cannot ....

[Article contains additional citation context not shown here]

Milind Tambe, W. Lewis Johnson, Randy Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....of modelling such as cognitive task analysis. They are used extensively for this purpose, both within Defence and more broadly. Within DSTO alone, there are a number of applications, such as SWARMM [5] OSIE and CAEN. A small sample of the work in this area from outside DSTO include Tac Air Soar [9] (US air combat simulation) the work of Frank Ritter and colleagues on a hand eye model for testing user interfaces [7] and the John Laird s work on modelling a Quake player [3] However despite this wide range of applications, there is no generic human operator agent in each case the agent ....

Milind Tambe, W. Lewis Johnson, Randolph M. Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), 1995.

....behavior is that of believability of agents. The basic premise is that characters should appear to be alive, to think broadly , to react emotionally and with personality to appropriate circumstances. There is a growing graphics and animated agent literature on the believability topic (e.g. see [6 19, 55 58]) and much of this work focuses on using great personality to mask the lack of deeper reasoning ability. However, in this paper we are less interested in the kinesthetics, media and broadly appealing personalities, than we are in the planning, judging, and choosing types of July 2001, Systems ....

....that can recognize and shift play to meta games and not get locked into brittle, narrow rules of play. The ability to do this is increasing due to the convergence and maturation of a number of lines of research in human behavior mo deling [27 46] in agent animation and expressive capabilities [6 19], and in computational sciences in general and for electronic theater and simulators in particular [20 26, 48 49, 55 63] Our goal is to benefit from and to further that integration with the framework presented here, and to focus it upon the task of developing agents embedded in interactive dramas ....

[Article contains additional citation context not shown here]

Tambe, M., Johnson, W.L., Jones, R.M., Koss, F. Laird, J.E., Rosenbloom, P.S. and Schwamb, K. (1995). Intelligent Agents for Interactive Simulation Environments, AI Magazine, 15-37

....environment, in a way that broadly fits a model of practice. For example, Cohen, Greenberg, Hart and Howe s (1989) Phoenix simulation models coordination of fire fighting teams. Hayes Roth, Brownston and Sinco# s (1995) simulations allow for improvisation in games played by the agents. Tambe, Johnson, Jones, Laird, Rosenbloom and Schwamb (1995) describe a tool that models social interactions such as briefing sessions before military missions. In general, such multiagent simulation tools have the following characteristics. f Agents are modeled as active, controlling the work flow by their beliefs, decisions actions, not as resources ....

TAMBE, M., JOHNSON, W. L., JONES, R. M., LAIRD, J. E., ROSENBLOOM,P.S.&SCHWAMB, K. (1995). Intelligent agents for interactive simulation environments. A1 Magazine, 16, 15---39.

....this hierarchical representation captures the flexibility necessary for agents behaving in complex environments. Furthermore, the approach has been used for both stick level control of an aircraft in simulation (Pearson, Huffman, Willis, Laird, and Jones 1993) and higher level tactical flight (Tambe, Johnson, Jones, Koss, Laird, Rosenbloom, and Schwamb 1995) for medium sized (400 3000 rule) real time expert systems, indicating the feasibility of such an approach. 2.2 Addressing Problems with Hierarchical Knowledge Hierarchical knowledge is economical and flexible, but it is not without drawbacks. The most obvious difficulty with hierarchical ....

....and fed back into the empirical evaluation of new capabilities to consider. Because A 5 is the most promising approach at present, an immediate priority is to investigate it in a dynamic domain. The domain currently being considered for this step is a scaled down version of the TacAir Soar (Tambe, Johnson, Jones, Koss, Laird, Rosenbloom, and Schwamb 1995) domain. TacAir Soar uses a hierarchical knowledge representation to control aircraft in a real time simulation at the tactical level. Because of problems such as non contemporaneous constraints and knowledge contention, TacAir Soar is currently used without learning. However, the domain is highly ....

Tambe, M., W. L. Johnson, R. M. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb (1995). Intelligent agents for interactive simulation environments. AI Magazine 16 (1), 15--39.

....enemy threats, re)plan routes to avoid threats and obstacles, etc. The distributed team is composed of diverse agents from four different research groups: A Quickset multimodal command input agent [1] a route planner [15] the Ariadne information agent [12] and eight synthetic helicopter pilots [19]. The team is integrated using the Teamcore multi agent integration architecture [20] which accomplishes integration by wrapping each agent with a proxy that maintains collaboration with other agents (via their own proxies) A distributed application is formed by a team of agents jointly ....

Milind Tambe, W. Lewis Johnson, Randy Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....to identify whether it would be possible to use the toolkit for their application. Conclusions We have briefly described the SIM AGENT toolkit and the way we have used it within the domain of Computer Generated Forces. Without more experience with other toolkits in the same domain, e.g. SOAR (Tambe et al. 1995), it is difficult to draw conclusions about how effective other tools would have been applied to this domain. The flexibility of the toolkit made it easy for us to develop domain specific enhancements, however this came at a cost in overall efficiency. Facilities for making and reasoning about ....

Tambe, M., Johnson, W. L., Jones, R.M., Koss, F., Laird, J.E., Rosenbloom, P.S., Schwamb, K. 1995 Intelligent Agents for Interactive Simulation Environments. AI Magazine 16(1).

....we observed was that an ecient, parallel development approach emerged. 1 Introduction Interactive simulation environments with large numbers of intelligent agents are becoming increasingly common. Such simulation environments often have agents playing the roles of humans within the simulation [17]. Examples of this type of environment are military simulations [16, 8] training [1] computer games [4] and RoboCup [10] In general, knowledge of precisely what agents should do in such environments is not an agent developer s area of expertise, rather it is a domain expert s expertise. It ....

Milind Tambe, W. Lewis Johnson, Randolph Jones, Frank Koss, John Laird, Paul Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15-39, Spring 1995.

....back to the environment, all in real time. The actors reasoning may need to be very complex taking into account a variety of factors including the current situation, a variety of concurrent, potentially con icting goals, team members, opponents, previous actions, resource constraints and so on [ Tambe et al. 1995b ] In order for the simulation in which the actors are embedded to be useful the actors must usually act in a human like manner. Often knowledge of precisely how an actor should act will be expert knowledge hence it is desirable to have domain experts, as opposed to actor experts, specifying ....

....increasing the desirability of easily reusing parts of existing speci cations. The pilot actors need to appear to be intelligent and act realistically in a very complex environment. The actuators for the actor, i.e. the aircraft controls, are extremely complex and allow many degrees of freedom [ Tambe et al. 1995a ] 1.3 Related Work Recently there has been a lot of interest in development methodologies for agents. Examples include methodologies for Belief Desire Intention (BDI) agents [ Kinny and George , 1996 ] for behavior based agents [ Bryson, 1998 ] for distributed multi agent systems [ ....

Milind Tambe, W. Lewis Johnson, Randolph Jones, Frank Koss, John Laird, Paul Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15-39, Spring 1995.

....for trainees to emulate, as well as providing a path for the acquisition of tacit knowledge. 1. Introduction Computer simulation has been used for rehearsing aeroplane manoeuvres and tactical operations, for determining component or answering specific questions about equipment requisitions [8]. In these operational simulations, intelligent agents interact with each other and possibly a human pilot in a virtual environment. The computer controlled pilots flying in simulations must respond realistically, in real time, to mission or tactical situations. These dynamic, interactive ....

Tambe, M. et. Al. Intelligent Agents for Interactive Simulation Environments. AI Magazine, Spring (1995).

....other agents that it is expected to recognize. Second order recognition (I recognize that she he has recognized my plan) complicates this significantly and in complex domains will quickly become unwieldy. There are other significant implemented agent systems for modelling warfare TAC AIR SOAR [ Tambe et al. 1995# Laird et al. 1994 ] is a notable example that implements a different model of cognition. The SOAR ar chitecture [ Laird et al. 1987# Newell, 1991 ] supports a multi layered view of cognition and may provide some features that may assist in the developmentofintegrated plan recognition. 2 ....

M. Tambe, W. L. Johnson, R. M. Jones, F. Koss, J. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16, Spring 1995.

....the name of a plan, you need to look at automatic ways of getting from the name of the plan to knowing it s in operation at the pattern or low level. The importance of agent recognition models is particularly emphasised for multi aeroplane air mission simulations. The tac airSOAR architecture [2] identifies agent perception as an important factor. Recently, the STEAM system and STOW 97 exercises have included battlefield missions involving thousands of virtual and agents [3] Hidden Markov Models have been used for multi agent soccer playing simulations in the Robocup league [4] although ....

Tambe, M., and Johnson, W. J., and Jones, R. M., and Koss, K., and Laird, J. E., and Rosenbloom, P. S., and Schwamb, K. Intelligent Agents for Interactive Simulation Environments. AI Magazine, pp15-39, Spring (1995).

....which it is derived (e.g. Newell Simon, 1972) has been applied both within artificial intelligence as a vehicle for constructing knowledge intensive systems, and within psychology for the modelling of human cognition. Although more work has been done on the artificial intelligence side (e.g. Tambe, Johnson, Jones, Koss, Laird, Rosenbloom Schwamb, 1995), there has been a steady stream of work exploring how Soar offers state of the art accounts of human empirical phenomena (Altmann, 1996; Altmann John, in press; Howes Young, 1996; Lewis, 1993, 1997a, 1997b; Miller Laird, 1996; Polk Newell, 1995; Wiesmeyer, 1992; Wiesmeyer Laird, 1993) ....

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S. & Schwamb, K. (1995) Intelligent agents for interactive simulation environments. AI Magazine, 16, 15-40.

....support reasoning about coordination. Rather, all capabilities for coordination are implemented as problem solving knowledge. The system in which these agents are constructed is called TacAir Soar. An earlier version of TacAir Soar, with only limited coordination capabilities, has been described byTambe et al. 1995) and a preliminary discussion of coordination in TacAir Soar was presented by Laird 2 et al. 1994) This domain is an appropriate test bed for our thesis because there are existing procedures used by the military to coordinate behavior, with the intent to minimize deliberation and communication ....

M. Tambe, W. L. Johnson, R. M. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligentagents for interactivesimulation environments. AI Magazine, 16(1), 1995.

....yield sub optimal solutions. Enhanced goal prioritization will be included in future PS versions. 6. FAILURE RESPONSE RA provides a two level failure response an immediate reactive response, and a longer term deliberative response. This is typical of many autonomy architectures (e.g. Soar [7], Guardian [8] The fast, real time reactive behavior is implemented by EXEC and MIR. If this fails to solve the problem within the time and resource constraints of the current plan, then the failure can endanger future goals in the plan. In this case EXEC puts the spacecraft in standby, PS is ....

Tambe, M., Johnson, W.L., Jones, R.M., Koss, F., Laird, J.E., Rosenbloon, P.S., and Schwamb, K. 1995. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15-39.

No context found.

M. Tambe, W. Johnson, R. Jones, F. Koss, J. Laird, P. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, page 16(1), 1995.

....Academic Publishers. Printed in the Netherlands. 1. Introduction Teamwork has been a growing area of agent research and development in recent years, seen in a large number of multi agent applications, including autonomous multi robotic space missions [10] virtual environments for training [35] and education [17] distributed resource allocation [19] and software agents on the Internet [34] With the growing importance of teamwork, there is now a critical need for tools to help humans analyze, evaluate, and understand team behaviors. Indeed, in multi agent domains with tens or even ....

....and is used remotely by teams preparing for these competitions. Although ISAAC was initially applied in RoboCup, ISAAC s techniques are intended to apply in other team domains such as agent teams in foraging and exploration [4] distributed resource allocation [19] and battlefield simulations [35]. For instance, in this article, we demonstrate the generality of ISAAC s techniques by applying it to the analysis of communication actions of a team of software agents [38] The team is engaged in the task of the simulated evacuation of civilians trapped in a hostile territory. Here ISAAC can ....

[Article contains additional citation context not shown here]

Tambe, M. Johnson, W. L., Jones, R., Koss, F., Laird, J. E., Rosenbloom, P.S., Schwamb, K.: Intelligent Agents for Interactive Simulation Environments. AI Magazine, 16(1) (Spring), 1995.

.... students in real time[32] In the arena of entertainment, recent work has focused on real time, dynamic interactivity among multiple agents within virtual reality environments[5, 12, 17] Similarly, in the arena of training, there is a recent thrust on dynamic, real time interactive simulations[24, 26, 33]. In these simulations, humans may interact with tens or hundreds of intelligent agents, as they participate in realistic traffic environments that simulate traffic jams and pedestrians[8] or air traffic control environments that simulate multiple aircraft on airfields[19] or large scale combat ....

....the dynamic behavior of agents based on this architecture. Section 7 presents related work, and Section 8 concludes. 2 Agent Tracking in a Real world Setting Our investigation of agent tracking is based on an on going effort to build intelligent pilot agents for a synthetic combat environment[26]. This environment is based on a commercially developed simulator called ModSAF[7] which has already been used in an operational military exercise involving human participants. ModSAF provides a synthetic yet real world setting for studying a broad range of challenging issues in agent tracking. ....

[Article contains additional citation context not shown here]

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

.... and then understanding why it does or does not work (see [ Hanks et al. 1993 ] for a related discussion) In step with this approach, we are investigating agent tracking in the context of our on going effort to build intelligent pilot agents for a real world synthetic air combat environment [ Tambe et al. 1995 ] This environment is based on a commercially developed simulator called ModSAF [ Calder et al. 1993 ] which has already been used in an operational military exercise involving expert human pilots. For an illustrative example of agent tracking in this environment, consider the scenario in ....

....enables RESC to be situated in its present as it tracks an agent s actions. Subsequently, RESC s ambiguity resolution and real time properties are described in Section 3. These descriptions are provided in concrete terms, using an implementation of the pilot agents in a system called TacAir Soar [ Tambe et al. 1995 ] built using the Soar architecture [ Newell, 1990; Rosenbloom et al. 1991 ] We assume some familiarity with Soar s problem solving model, which involves applying operators to states to reach a desired state. 2 Tracking Flexible Goal driven and Reactive Behaviors In an environment such as ....

[Article contains additional citation context not shown here]

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and or a fee. techniques we used to successfully model the behavior of military pilots [1, 6]. However, as we developed the Quakebot, we found that improving the behavior of the bot required more and more specialized tactics. In addition, when we presented our work to game developers, they invariably asked, Does it anticipate the human players actions If it did, that would be really ....

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S., and Schwamb, K. (1995), Intelligent Agents for Interactive Simulation Environments, AI Magazine, 16 (1), 15-39.

....conditional effects or duration. The sufficiency of this representation for constraints D1 D3 (and efficient matching [Doorenbos, 1993] has been demonstrated for complex, real time domains including control of simulated aircraft [Pearson et al. 1993] and tactical air combat [Laird et al. 1995; Tambe et al. 1995] An IMPROV agent will use this operator knowledge to both plan and behave in its world. Errors in the planning knowledge lead to errors in behavior, so it is planning knowledge that IMPROV learns to correct. In IMPROV, planning does not create a monolithic plan. Instead, situation dependent ....

M. Tambe, W. L. Johnson, R. M. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. To appear in AI Magazine, 1995.

....Intelligence. pp. 928 935. Copyright c 1998, American Association for Arti cial Intelligence (www.aaai.org) All rights reserved. architectures (Firby 1987; George Lansky 1987; Laird, Newell, Rosenbloom 1987) for execution tasks in complex, dynamic environments. For example, TacAir Soar (Tambe et al. 1995), which has been successfully used to simulate human pilots in large scale distributed simulations, has over 450 tasks and subtasks in hierarchies that can grow to over 10 levels. Figure 1 shows how the task of intercepting an enemy plane is decomposed until a relatively simple decision can be ....

Tambe, M.; Johnson, W. L.; Jones, R. M.; Koss, F.; Laird, J. E.; Rosenbloom, P. S.; and Schwamb, K. 1995. Intelligent agents for interactive simulation environments. AI Magazine 16(1):15-39.

....similar to Steve s. Because our contribution lies in the integration of Steve s capabilities, rather than novel methods for individual capabilities, we do not discuss related work on these individual capabilities #e.g. plan construction and execution, student monitoring, and explanation#. Tambe et al. #1995# designed a Soar #ghter pilot that can engage in air combat with other simulated pilots. Their pilot can execute missions, monitor other pilots as they execute their missions #Tambe Rosenbloom 1995#, and explain the rationale behind actions #Johnson 1994#. There are many di#erences between ....

Tambe, M.; Johnson, W. L.; Jones, R. M.; Koss, F.; Laird, J. E.; Rosenbloom, P. S.; and Schwamb, K. 1995. Intelligent agents for interactive simulation environments.

....in teams rose out of growing frustration with the signi cant software maintenance e orts in two of our application domains. In the ModSAF domain, a high delity battle eld virtual environment (Calder et al. 1993) we have been involved in the development of synthetic helicopter pilots (Tambe et al. 1995). In the RoboCup soccer simulation domain (Kitano et al. 1997) wehave been involved in developing synthetic soccer players (Marsella, Adibi, Al Onaizan, Kaminka, Muslea, Tallis, Tambe, 1999) The environments in both domains are dynamic and complex, and have many uncertainties: the behavior of ....

Tambe, M., Johnson, W. L., Jones, R., Koss, F., Laird, J. E., Rosenbloom, P. S., & Schwamb, K. (1995). Intelligent agents for interactive simulation environments. AI Magazine, 16 (1).

....that we can actually peek under the hood at the agents reasoning processes, and the various factors and mental states that lead to the agents external behavior. The Situational Awareness Panel (SAP) is a graphical user interface to aid in the observation, use, and development of TacAir Soar [1, 2, 3] intelligent agents that control simulated military aircraft. The SAP allows a human operator or engineer to view various representations of an intelligent agent s current perception, awareness, goals, and intentions during an exercise. Currently, the SAP includes four sub displays (see Figure 1) ....

M. Tambe, W. L. Johnson, R. M. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, & K. B. Schwamb: "Intelligent Agents for Interactive Simulation Environments". AI Magazine, Vol. 16, No. 1, pp. 15-39, 1995.

No context found.

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine,

No context found.

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S., & Schwamb, K. B. (1995) "Intelligent agents for interactive simulation environments." AI Magazine, 16(1), 15-39.

....results spanning many domains and more than #fteen years #see, for example, #Rosenbloom et al. 1993# for thorough coverage of the #rst half of this#. In addition, the core of the work in Soar over the past #veto ten years has concerned autonomous, real time systems see, for example, #Tambe et al. 1995# for which boundedness is a critical requirement for usable learning. Chunking, as implemented and extensively investigated in Soar, is already a variant of EBL #Rosenbloom Laird, 1986#; however, for this work wehave replaced it with a more standard version of EBL in order to more easily ....

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S., & Schwamb, K. B. #1995#. Intelligent agents for interactive simulation environments. AI Magazine, 16:15#39.

....in teams rose out of growing frustration with the signi cant software maintenance e orts in two of our application domains. In the ModSAF domain, a high delity battle eld virtual environment (Calder et al. 1993) we have been involved in the development of synthetic helicopter pilots (Tambe et al. 1995). In the RoboCup soccer simulation domain (Kitano et al. 1997) we have been involved in developing synthetic soccer players (Marsella, Adibi, Al Onaizan, Kaminka, Muslea, Tallis, Tambe, 1999) The environments in both domains are dynamic and complex, and have many uncertainties: the behavior of ....

Tambe, M., Johnson, W. L., Jones, R., Koss, F., Laird, J. E., Rosenbloom, P. S., & Schwamb, K. (1995). Intelligent agents for interactive simulation environments. AI Magazine, 16 (1).

....training environments. These agents, in the form of information agents, planning execution agents, middle agents, user agents or embedded agents must often operate in cyberspace (on the internet or intra Gammanets) to interface with relevant information sources, network facilities and other agents[3, 4, 1, 12]. This growth in agent based systems is predicted to be followed by another powerful trend: the reuse of specialized agents as standardized building blocks for large scale systems[4, 6, 2] This prediction is based on two observations. First, software systems are being constructed with ....

....Institute) Queries a database on dynamic obstacles and threats (e.g. enemy missile launchers) Written in Lisp, runs on Unix platforms. Helicopter pilots: developed by Tambe, USC Information Sciences Institute) These pilot agents were earlier developed for distributed interactive simulations[12], and they can fly helicopters along a specified route, land at specified destinations, and mask for observations. The pilot agents used in this work have no teamwork capabilities. Written in Soar, run on a Unix Platform. As seen above, these agents are developed by different research groups, ....

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....to perform the task using this new performance knowledge. In this section KnoMic is described in the context of a flight simulation domain similar to the domains used in behavioral cloning and TacAir Soar, a 5,200 rule, hand coded intelligent agent used in air combat training exercises [7]. 4.1 Expert Environment Loop An expert s interaction with the environment while performing a task can be viewed as a communication loop. The environment sends information to the expert in the form Environment Sensor Inputs Actions Condition Learning Behavior Trace Goal Conditions ....

Tambe, M., Johnson, W. L., Jones R. M., Koss, F., Laird, J. E., Rosenbloom, P. S. and Schwamb, K. (1995). Intelligent agents for interactive simulation environments. AI Magazine 16(1):15-39.

....within a team, etc. For each of these research problems, the presence of multiple cooperative and non cooperative agents, only compounds the difficulty. Consider for instance the challenge of multi agent teamwork, which has become a critical requirement across a wide range of multi agent domains[16, 10, 17]. Here, an agent team must address the challenge of designing roles for individuals (i.e. dividing up team responsibilities based on individuals capabilities) doing so with fairness, and reorganizing roles based on new information. Furthermore, agents must also flexibly coordinate and ....

....computes a direction to shoot the ball into the opponents goal, and a micro plan, consisting of turn or dash actions, to intercept the ball. The lower level does not make any decisions. Instead, all decisionmaking rests with the higher level, implemented in the Soar integrated AI architecture[16]. Once the Soar based higher level reaches a decision, it communicates with the lower level, which then sends the relevant action information to the simulator. Soar s operation involves dynamically executing an operator (reactive plan) hierarchy. The operator hierarchy shown in Figure 2 ....

Tambe, M., W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb: 1995, `Intelligent agents for interactive simulation environments'. AI Magazine 16(1).

No context found.

Tambe, M., Johnson, W.L., Jones, R., Koss, F., Laird, J.E., Rosenbloom, P.S. & Schwamb, K. Intelligent Agents for interactive simulation environments. AI Magazine, 16#1#, 1995.

....structure the state space into an abstract situation based model of behavior that supports interpretation in the face of missing information about agent s actions and goals. 1 Introduction Virtual worlds inhabited by synthetic agents are increasingly being used for training and education (e.g. [12, 7, 16]) These virtual worlds can provide a highly engaging environment in which to develop skills 1 that students can readily apply to real world tasks. In creating these environments, considerable effort is often expended so that the simulated world more faithfully mirrors the real world in which the ....

....the same kind of complex behaviors that human participants would exhibit. For example, Distributed Interactive Simulation (DIS) training sessions involve teams of human students interacting with potentially thousands of synthetic and human agents within a very dynamic battlefield simulation (e.g. [16]) From an instructor s perspective, the use of very dynamic multiagent virtual environments raises several concerns. Among these is the seemingly simple question of what the student teams are currently doing. In a complex simulated world, conditions in the simulation can be difficult to ....

Tambe, M., Johnson, W.L., Jones, R., Koss, F., Laird, J. E., Rosenbloom, P.S. & Schwamb, K. Intelligent Agents for interactive simulation environments. In The AI Magazine, 16(1), Spring, 1995.

....how the functional design constraints map on to cognitively plausible representations and mechanisms, sometimes in surprising ways. Introduction For the past few years, we have been developing a computer system, called TacAir Soar, that flies aircraft in tactical air combat simulations (Tambe et al. 1995). The overall goal of this work is for the system to generate behavior that looks like it is being generated by an expert level human. The evaluation of our system takes place at a very high level of behavior. To test TacAir Soar, we place it in a variety of different situations, flying different ....

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laird, J. E., Rosenbloom, P. S., & Schwamb, K. B. (1995). Intelligent agents for interactive simulation environments. AI Magazine, 16(1), 15--39.

....task. We outline the current state of our TOP implementation and the outstanding issues in developing such a framework. 1 Introduction Agent based systems currently operate in complex, dynamic environments such as user interfaces [18] robotic space missions, virtual training environments [22], and Internet information extraction [28] These agents are often autonomous, heterogeneous, and distributed over a variety of platforms and domains. Yet, users may desire such diverse agents to work together to accomplish novel, complex tasks. Such reuse of existing agents is preferable to ....

Milind Tambe, W. Lewis Johnson, Randolph Jones, Frank Koss, John E. Laird, Paul S. Rosenbloom, and Karl Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

....examples from an analytic search tree and some real world domains are presented. Keywords: Multi agent systems, Teamwork, Persistence, Markov decision processes 1. Introduction Teamwork is critical in many multi agent environments, such as, interactive simulations for training and education[34], RoboCup robotic and synthetic soccer[20] interactive entertainment[13] multi robot deep sea or space exploration or reconnaissance, and internet based information integration. An increasingly important requirement in many of these domains is that of persistent teams, i.e. teams that persist ....

....deep sea or space exploration or reconnaissance, and internet based information integration. An increasingly important requirement in many of these domains is that of persistent teams, i.e. teams that persist over long periods of time. For instance, consider virtual environments for training[34]. Here, the Advanced Concepts Technology Demonstration battlefield simulation exercise (henceforth, referred to as ACTD) jointly conducted in the US and Europe in October 1997, lasted for multiple c fl 1999 Kluwer Academic Publishers. Printed in the Netherlands. FINAL.tex; 5 11 1999; 17:59; p.1 2 ....

Tambe, M., W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb: 1995, `Intelligent agents for interactive simulation environments'. AI Magazine 16(1).

No context found.

Tambe, M., W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb: 1995, `Intelligent agents for interactive simulation environments'. AI Magazine 16(1).

No context found.

Tambe, M., Johnson, W., Jones, R., Kossand, F., Lairdand, J., Rosenbloom, P., Schwamb, K.: Intelligent agents for interactive simulation environments. AI Magazine (1995) 15--39

No context found.

Tambe, M, W.L. Johnson, R.M. Jones, F. Koss, J.E. Laird, P.S. Rosenbloom, K. Schwamb, Intelligent Agents for Interactive Simulation Environments, AI Magazine 16(1):15-39, Spring.1995.

No context found.

Tambe, M., W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb: 1995, `Intelligent agents for interactive simulation environments'. AI Magazine 16(1).

No context found.

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1), Spring 1995.

No context found.

Tambe, M., Johnson, W. L., Jones, R. M., Koss, F., Laid, J. E., Rosenbloom, P. S., & Schwamb, K. (1995). Intelligent Agents for Interactive Simulation Environments. AI Magazine, 16(Spring).

No context found.

M. TAMBE, L. JOHNSON, R. JONES, F. KOSS, J. LAIRD, P. ROSENBLOOM, K. SCHWAMB, "Intelligent Agents for Interactive Simulation Environments", AI Magazine, 16, 15-39 (Spring 1995).

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC