Results 1 -
7 of
7
The RoboCup 2013 Drop-In Player Challenges: A Testbed for Ad Hoc Teamwork
, 2014
"... Ad hoc teamwork has recently been introduced as a general challenge for AI and especially multiagent systems [15]. The goal is to en-able autonomous agents to band together with previously unknown teammates towards a common goal: collaboration without pre-coordination. While research to this point ..."
Abstract
-
Cited by 9 (6 self)
- Add to MetaCart
(Show Context)
Ad hoc teamwork has recently been introduced as a general challenge for AI and especially multiagent systems [15]. The goal is to en-able autonomous agents to band together with previously unknown teammates towards a common goal: collaboration without pre-coordination. While research to this point has focused mainly on theoretical treatments and empirical studies in relatively simple domains, the long-term vision has always been to enable robots or other autonomous agents to exhibit the sort of flexibility and adaptability on complex tasks that people do, for example when they play games of “pick-up” basketball or soccer. This paper chronicles the first evaluation of autonomous robots doing just that: playing pick-up soccer. Specifically, in June 2013, the authors helped organize a “drop-in player challenge” in three different leagues at the international RoboCup competition. In all cases, the agents were put on teams with no pre-coordination. This paper documents the structure of the challenge, describes some of the strategies used by participants, and analyzes the results.
Give a hard problem to a diverse team: Exploring large action spaces
- In Proceedings of of the 28th Conference on Artificial Intelligence, AAAI’14
, 2014
"... Recent work has shown that diverse teams can outper-form a uniform team made of copies of the best agent. However, there are fundamental questions that were not asked before. When should we use diverse or uniform teams? How does the performance change as the action space or the teams get larger? Hen ..."
Abstract
-
Cited by 5 (4 self)
- Add to MetaCart
(Show Context)
Recent work has shown that diverse teams can outper-form a uniform team made of copies of the best agent. However, there are fundamental questions that were not asked before. When should we use diverse or uniform teams? How does the performance change as the action space or the teams get larger? Hence, we present a new model of diversity for teams, that is more general than previous models. We prove that the performance of a diverse team improves as the size of the action space gets larger. Concerning the size of the diverse team, we show that the performance converges exponentially fast to the optimal one as we increase the number of agents. We present synthetic experiments that allow us to gain further insights: even though a diverse team out-performs a uniform team when the size of the action space increases, the uniform team will eventually again play better than the diverse team for a large enough ac-tion space. We verify our predictions in a system of Go playing agents, where we show a diverse team that im-proves in performance as the board size increases, and eventually overcomes a uniform team.
Communicating with Unknown Teammates
- Proceedings of the 13th Adaptive and Learning Agents workshop
, 2013
"... ABSTRACT Past research has investigated a number of methods for coordinating teams of agents, but, with the growing number of sources of agents, it is likely that agents will encounter teammates that do not share their coordination methods. Therefore, it is desirable for agents to form an effective ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
(Show Context)
ABSTRACT Past research has investigated a number of methods for coordinating teams of agents, but, with the growing number of sources of agents, it is likely that agents will encounter teammates that do not share their coordination methods. Therefore, it is desirable for agents to form an effective ad hoc team. This research tackles the problem of communication in ad hoc teams, introducing a minimal version of the multiagent, multi-armed bandit problem with limited communication between the agents. This abstract summarizes theoretical results that prove that this problem setting can be solved in polynomial time when the agent knows the set of possible teammates, and the empirical results that show that the problems can be solved in practice.
The Role of Models and Communication in the Ad Hoc Multiagent Team Decision Problem
, 2015
"... Abstract Ad hoc teams are formed of members who have little or no information regarding one another. In order to achieve a shared goal, agents are tasked with learning the capabilities of their teammates such that they can coordinate effectively. Typically, the capabilities of the agent teammates e ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract Ad hoc teams are formed of members who have little or no information regarding one another. In order to achieve a shared goal, agents are tasked with learning the capabilities of their teammates such that they can coordinate effectively. Typically, the capabilities of the agent teammates encountered are constrained by the particular domain specifications. However, for wide application, it is desirable to develop systems that are able to coordinate with general ad hoc agents independent of the choice of domain. We propose examining ad hoc multiagent teamwork from a generalized perspective and discuss existing domains within the context of our framework. Furthermore, we consider how communication of agent intentions can provide a means of reducing teammate model uncertainty at key junctures, requiring an agent to consider its own information deficiencies in order to form communicative acts improving team coordination.
Announced strategy types in multiagent RL for conflict-avoidance in the national airspace
"... Automated conflict-avoidance for unmanned aerial systems is quickly becoming feasible due to near-future advances in communication guarantees. The NextGen Implementation Plan introduced by the FAA lays out requirements for air traffic in the US airspace to be implemented by 2020. This includes manda ..."
Abstract
- Add to MetaCart
(Show Context)
Automated conflict-avoidance for unmanned aerial systems is quickly becoming feasible due to near-future advances in communication guarantees. The NextGen Implementation Plan introduced by the FAA lays out requirements for air traffic in the US airspace to be implemented by 2020. This includes mandates for the introduction of the automatic de-pendent surveillance broadcast (ADS-B) on each system in the airspace, which is a plane-installed system that broad-casts information about a plane in the airspace. Optionally, a pilot can choose to install the traffic information service broadcast (TIS-B) system, which allows a plane to observe ADS-B information for a 15-mile radius. This enhanced communication lends itself well to the for-mation of a multiagent system to model conflict-resolution in the air. But with a variety of different potential reactions by pilots in a conflict situation, a robust response structure must be developed. In this work, we introduce agents to make distributed decisions regarding rerouting for conflict-avoidance, and explore the possible benefit of stereotyping in this domain. We demonstrate also that the availability of avoidance type information can improve performance by reducing agent noise in a simple system reward formulation.
Cooperating with Unknown Teammates in Complex Domains: A Robot Soccer Case Study of Ad Hoc Teamwork
"... Many scenarios require that robots work together as a team in order to effectively accomplish their tasks. However, pre-coordinating these teams may not always be possible given the growing number of companies and research labs creating these robots. Therefore, it is desirable for robots to be able ..."
Abstract
- Add to MetaCart
Many scenarios require that robots work together as a team in order to effectively accomplish their tasks. However, pre-coordinating these teams may not always be possible given the growing number of companies and research labs creating these robots. Therefore, it is desirable for robots to be able to reason about ad hoc teamwork and adapt to new team-mates on the fly. Past research on ad hoc teamwork has fo-cused on relatively simple domains, but this paper demon-strates that agents can reason about ad hoc teamwork in com-plex scenarios. To handle these complex scenarios, we intro-duce a new algorithm, PLASTIC–Policy, that builds on an existing ad hoc teamwork approach. Specifically, PLASTIC– Policy learns policies to cooperate with past teammates and reuses these policies to quickly adapt to new teammates. This approach is tested in the 2D simulation soccer league of RoboCup using the half field offense task. 1
Cooperating with Unknown Teammates in Robot Soccer
, 2014
"... Many scenarios require that robots work together as a team in order to effectively accomplish their tasks. However, pre-coordinating these teams may not always be possible given the growing number of companies and research labs creating these robots. Therefore, it is desirable for robots to be able ..."
Abstract
- Add to MetaCart
(Show Context)
Many scenarios require that robots work together as a team in order to effectively accomplish their tasks. However, pre-coordinating these teams may not always be possible given the growing number of companies and research labs creating these robots. Therefore, it is desirable for robots to be able to reason about ad hoc teamwork and adapt to new teammates on the fly. This paper adopts an approach of learning policies to cooperate with past teammates and reusing these policies to quickly adapt to the new teammates. This approach is applied to the complex domain of robot soccer in the form of half field offense in the RoboCup simulated 2D league. This paper represents a preliminary investigation into this domain and presents a promising approach for tackling this problem.
