Issues and approaches in the design of collective autonomous agents
Robotics and Autonomous SystemsPublished 1 December 1995
Maja J. Matarić
Citations272
SJR quartileQ1
SJR score1.17
SNIP1.99
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The most prominent directions of research are overviewed, key terms are defined, and the approach to controlling group behavior and its relation to the field as a whole is described.
Abstract
The problem of synthesizing and analyzing collective autonomous agents has only recently begun to be practically studied by the robotics community. This paper overviews the most prominent directions of research, defines key terms, and summarizes the main issues. Finally, it briefly describes our approach to controlling group behavior and its relation to the field as a whole.
Keywords
Computer ScienceEngineering
Computers in PhysicsIntroduction to Applied Nonlinear Dynamical Systems and Chaos
3,833 Citations1990Stephen Wiggins, D.S. Mazel
Texts in applied mathematicsIntroduction to Applied Nonlinear Dynamical Systems and Chaos
2,797 Citations1990Stephen Wiggins
Distributed Optimization by Ant Colonies
2,580 Citations1992Alberto Colorni, Marco Dorigo +3 more
A distributed problem solving environment is introduced and its use to search for a solution to the travelling salesman problem is proposed.
Intelligence Without Reason
1,556 Citations2018Rodney A. Brooks
It is claimed that the state of computer architecture has been a strong influence on models of thought in Artificial Intelligence over the last thirty years.
Medical Entomology and ZoologyThe complexity of robot motion planning
1,398 Citations1988John Canny
John Canny resolves long-standing problems concerning the complexity of motion planning and, for the central problem of finding a collision free path for a jointed robot in the presence of obstacles, obtains exponential speedups over existing algorithms by applying high-powered new mathematical techniques.
The International Journal of Robotics ResearchAutomatic Synthesis of Fine-Motion Strategies for Robots
898 Citations1984Tomás Lozano‐Pérez, Matthew T. Mason +1 more
A formal approach to the synthesis of compliant-motion strategies from geometric descriptions of assembly operations and explicit estimates of errors in sensing and control is described.
Physica D Nonlinear PhenomenaCo-evolving parasites improve simulated evolution as an optimization procedure
775 Citations1990W. Daniel Hillis
An example of how simulated evolution can be applied to a practical optimization problem and more specifically, how the addition of co-evolving parasites can improve the procedure by preventing the system from sticking at local maxima is shown.
Studies in cognitive systemsFom Local Actions to Global Tasks: Stigmergy and Collective Robotics
452 Citations2000Ralph Beckers, Owen Holland +1 more
This paper presents a series of experiments where a group of mobile robots gather 81 randomly distributed objects and cluster them into one pile through stigmergy, a principle which allows indirect communication between agents through sensing and modification of the local environment which determines the agents’ behaviour.
Co-evolving parasites improve simulated evolution as an optimization procedure
448 Citations1991W. Daniel Hillis
Interaction and Intelligent Behavior.
388 Citations1994Maja J. Matarić
A novel formulation of reinforcement learning is proposed that makes behavior selection learnable in noisy, uncertain multi-agent environments with stochastic dynamics, and enables and accelerates learning in complex multi-robot domains.
Elsevier eBooksReward Functions for Accelerated Learning
377 Citations1994Maja J. Matarić
A methodology for designing reinforcement functions that take advantage of implicit domain knowledge in order to accelerate learning in situated domains characterized by multiple goals, noisy state, and inconsistent reinforcement is proposed.
Cooperative multi-robot box-pushing
370 Citations2002Maja J. Matarić, Martin Nilsson +1 more
This paper presents and experimentally demonstrate an approach that utilizes cooperation at three levels: sensing, action, and control, and takes the advantage of a simple communication protocol to compensate for the robots' noisy and uncertain sensing.
National Conference on Artificial IntelligenceOn the synthesis of useful social laws for artificial agent societies
270 Citations1992Yoav Shoham, Moshe Tennenholtz
This paper argues that the notion of social law is not epiphenomenal, but rather should be built into the action representation; it then offers such a representation.
The Knowledge Engineering ReviewOn distributed artificial intelligence
249 Citations1988Luís Eduardo, Castillo Hern
Two general frameworks, the System conceptual model and the agent conceptual model, are used to compare the different approaches to formalizing some of the more critical capabilities required by multi-agent interaction.
Toward a Practice of Autonomous Systems: Proceedings of the First European Conference on Artificial Life
243 Citations1992Francisco J. Varela, Paul Bourgine
Part 1 Autonomous robots: artificial life and real robots, Rodney A. Brooks concept formation as emergent phenomena, Mukesh J. Patel and Uwe Schnepf distributed adaptive control - a paradigm for designing autonomous agents, Rolf Pfeifer and Paul Verschure taking eliminative materialism seriously - a methodology for autonomous systems research, Tim Smithers an adaptable mobile robot.
Indoor automation with many mobile robots
180 Citations2002Philippe Caloud, Wonyun Choi +3 more
The aim of this paper is to provide an overview of the GOFER project and present the current solutions to many research issues: implementation of nonconflicting sensor systems, man-ro robot and robot-robot communication systems and protocols, contingency-tolerant motion control, multi-ro Bot motion planning, and multi- robot task planning and scheduling.
The International Journal of Robotics ResearchToward a Robot Architecture Integrating Cooperation between Mobile Robots: Application to Indoor Environment
171 Citations1993F.R. Noreils
This architecture exhibits important benefits such as mod ularity, robustness, and robustness (although some modules on the robot fail, it is still able to perform useful tasks), and programmability.
Computers & Mathematics with ApplicationsArtificial life IV: Proceedings of the fourth international workshop on the synthesis and simulation of living systems
167 Citations1995
This book brings together contributions to the Fourth Artificial Life Workshop, held at the Massachusetts Institute of Technology in the summer of 1994, and discusses self-organization and emergent functionality, as well as philosophical issues in A-life (from ontology to ethics).
ScienceCooperative Solution of Constraint Satisfaction Problems
153 Citations1991Scott H. Clearwater, Bernardo A. Huberman +1 more
Experimental results are presented on constraint satisfaction that both test the predictions of a theory of cooperative problem solving and assess the value of cooperation for this class of problems and suggest an alternative methodology for solving constraint satisfaction problems in computer science and distributed artificial intelligence.
Structure decision method for self organising robots based on cell structures-CEBOT
146 Citations2003Toshio Fukuda, Seiya Nakagawa +2 more
An optimal structure decision method that can determine cell type, arrangement, degree of freedom, and link length is proposed that is applicable to fixed-base and mobile-base manipulators.
The MIT Press eBooksAn Adaptive Communication Protocol for Cooperating Mobile Robots
113 Citations1993Holly A. Yanco, Lynn Andrea Stein
A mobile robots engaged in a cooperative task that requires communication is described, initially given a but uninterpreted vocabulary for communication and attempting to perform their task the robots learn a private communication lan guage.
Gradient driven self-organizing systems
102 Citations2002Giulio Sandini, G. Lucarini +1 more
The goal of the experiment is to analyze the performance of simple autonomous robots in reaching the point of highest concentration (i.e. the gas leak) and to analyze different strategies based on different configurations of the sensing devices and reflexive motor controls.
Physica D Nonlinear PhenomenaThe performance of cooperative processes
76 Citations1990Bernardo A. Huberman
For a wide class of problems, there is a highly nonlinear and universal increase in performance due to the interactions between agents, and in some cases this is further enhanced by sharp phase transitions in the topological structure of the problem.
National Conference on Artificial IntelligenceSolving the really hard problems with cooperative search
74 Citations1993Tad Hogg, Colin P. Williams
It is found that simple cooperative methods can often solve hard graph coloring problems near a previously identified transition between under- and overconstrained instances faster than the same number of independent agents.
Journal of Theoretical BiologySocioinformatic processes: MIRROR modelling methodology
57 Citations1985Paulien Hogeweg, B. Hesper
A formalism in which socioinformatic processes can be modelled in a natural way using concepts of MIRROR modelling is discussed, and two very simple principles for the behaviour of individuals are presented and shown to be sufficient to generate “interesting” social structures.
Physica D Nonlinear PhenomenaCollective behavior of predictive agents
48 Citations1990Jeffrey O. Kephart, Tad Hogg +1 more
This work shows that the system performance can be improved if the agents do an adequate job of predicting the current state of the system, and explores two plausible methods for prediction - technical analysis and system analysis, which obtain a successful hybrid of the two prediction methods.
Agents contracting tasks in non-collaborative environments
45 Citations1993Sarit Kraus
This work proposes techniques that provide efficient ways to reach subcontracting in varied situations: the agents have full information about the environment and each other vs. subcontracting when the agents don't know the exact state of the world.
DSpace@MIT (Massachusetts Institute of Technology)On probabilistic strategies for robot tasks
38 Citations1989Michael Erdmann
This thesis introduces active randomization by a robot, both in selecting actions to execute and in focusing on sensory information to interpret, as a basic tool for overcoming uncertainty.
Controlling collective tasks with an ALN
33 Citations2002C. Ronald Kube, Huijuan Zhang +1 more
The authors simplify the behavior arbitration in the collective task by training an ALN implemented using simple combinational logic.
International Joint Conference on Artificial IntelligenceConsenting agents: negotiation mechanisms for multi-agent systems
24 Citations1993Jeffrey S. Rosenschein
Concepts from fields such as decision theory and game theory can provide standards to be used in the design of appropriate negotiation and interaction environments are considered.
Mathematical analysis of behavior systems
22 Citations2005Luc Steels
The paper proposes a mathematical approach for the analysis of behavior systems that rests on viewing a behavior system as a dynamical system whose equilibrium state is attained when the behavior it is responsible for is achieved.
Instinctive behaviors and personalities in societies of cellular robots
21 Citations2002Paolo Dario, F. Ribechini +2 more
A description is presented of the social organization of societies of cellular mobile units featuring instinctive behavior of robotic units, which emphasizes the roles of innate personality, external stimuli, and communication.
Overeager reciprocal rationality and mixed strategy equilibria
20 Citations1993Edmund H. Durfee, Jaeho Lee +1 more
A crucial difference between the approaches is that RMM employs a solution concept that is overeager, which can be reduced by introducing into RMM second-order knowledge about what it knows, in the form of a flexible function for mapping relative expected utility of an option into the probability that the agent will pursue that option.
An Experiment in Mobile Robotic Cooperation
19 Citations1994Lynne E. Parker
The ability of the software architecture to create robust and reliable teams of cooperative mobile robots that are able to function in dynamic, changing environments is demonstrated.
Evolution of Communication in a Population of Simple Machines
19 Citations1990Bruce J. MacLennan
National Conference on Artificial IntelligenceConstrained intelligent action: planning under the influence of a master agent
14 Citations1992Eithan Ephrati, Jeffrey S. Rosenschein
This paper analyzes a particular model of control among intelligent agents, that of nonabsolute control, which involves a "supervisor" agent that issues orders to a "subordinate" agent.
Stationary waves in two-dimensional cyclic swarms
7 Citations2002Qingdao Huang, G. Beni
The authors extend the problem of stationary waves in cyclic swarms to two dimensions and it is shown that for the systems of linear equations describing a difference equation with periodic boundary conditions and specific conservation conditions, one can find an algorithm to solve asynchronously the systemsof linear equations.
eScholarship (California Digital Library)Strategic Social Planning - Looking for Willingness in Multi-Agent Domains
3 Citations1993Miceli, Maria, Cesta, Amedeo
Fuzzy control of group with a leader and their behaviors
3 Citations2002Kenji KUROSU, T. Furuya +1 more
Behaviors described by fuzzy rules were studied to see whether the leader can guide the individual members to form a group, and simulation results lead to the following conclusions.