Results 1 - 10
of
30
Putting the “smarts” into the smart grid: a grand challenge for artificial intelligence
- Communications of the ACM
"... The phenomenal growth in material wealth experienced in developed countries throughout the twentieth century has largely been driven by the availability of cheap energy de-rived from fossil fuels (originally coal, then oil, and most ..."
Abstract
-
Cited by 46 (7 self)
- Add to MetaCart
(Show Context)
The phenomenal growth in material wealth experienced in developed countries throughout the twentieth century has largely been driven by the availability of cheap energy de-rived from fossil fuels (originally coal, then oil, and most
Asynchronous Algorithms for Approximate Distributed Constraint Optimization with Quality Bounds
"... Distributed Constraint Optimization (DCOP) is a popular framework for cooperative multi-agent decision making. DCOP is NPhard, so an important line of work focuses on developing fast incomplete solution algorithms for large-scale applications. One of the few incomplete algorithms to provide bounds o ..."
Abstract
-
Cited by 23 (5 self)
- Add to MetaCart
Distributed Constraint Optimization (DCOP) is a popular framework for cooperative multi-agent decision making. DCOP is NPhard, so an important line of work focuses on developing fast incomplete solution algorithms for large-scale applications. One of the few incomplete algorithms to provide bounds on solution quality is k-size optimality, which defines a local optimality criterion based on the size of the group of deviating agents. Unfortunately, the lack of a general-purpose algorithm and the commitment to forming groups based solely on group size has limited the use of k-size optimality. This paper introduces t-distance optimality which departs from k-size optimality by using graph distance as an alternative criteria for selecting groups of deviating agents. This throws open a new research direction into the tradeoffs between different group selection
Bounded decentralised coordination over multiple objectives
- In AAMAS
, 2011
"... We propose the bounded multi-objective max-sum algorithm (B-MOMS), the first decentralised coordination algorithm for multi-objective optimisation problems. B-MOMS extends the max-sum message-passing algorithm for decentralised coordination to compute bounded approximate solutions to multi-objective ..."
Abstract
-
Cited by 9 (0 self)
- Add to MetaCart
(Show Context)
We propose the bounded multi-objective max-sum algorithm (B-MOMS), the first decentralised coordination algorithm for multi-objective optimisation problems. B-MOMS extends the max-sum message-passing algorithm for decentralised coordination to compute bounded approximate solutions to multi-objective decentralised constraint optimisation problems (MO-DCOPs). Specifically, we prove the optimality of B-MOMS in acyclic constraint graphs, and derive problem dependent bounds on its approximation ratio when these graphs contain cycles. Furthermore, we empirically evaluate its performance on a multi-objective extension of the canonical graph colouring problem. In so doing, we demonstrate that, for the settings we consider, the approximation ratio never exceeds 2, and is typically less than 1.5 for less-constrained graphs. Moreover, the runtime required by B-MOMS on the problem instances we considered never exceeds 30 minutes, even for maximally constrained graphs with 100 agents. Thus, B-MOMS brings the problem of multi-objective optimisation well within the boundaries of the limited capabilities of embedded agents.
Max/min-sum distributed constraint optimization through value propagation on an alternating dag
- In Proceedings of The Eleventh International Conference on Autonomous Agents and Multiagent Systems
, 2012
"... Distributed Constraint Optimization Problems (DCOPs) are NPhard and therefore the number of studies that consider incomplete algorithms for solving them is growing. Specifically, the Max-sum algorithm has drawn attention in recent years and has been applied to a number of realistic applications. Unf ..."
Abstract
-
Cited by 6 (2 self)
- Add to MetaCart
(Show Context)
Distributed Constraint Optimization Problems (DCOPs) are NPhard and therefore the number of studies that consider incomplete algorithms for solving them is growing. Specifically, the Max-sum algorithm has drawn attention in recent years and has been applied to a number of realistic applications. Unfortunately, in many cases Max-sum does not produce high quality solutions. More specifically, when problems include cycles of various sizes in the factor graph upon which Max-sum performs, the algorithm does not converge and the states that it visits are of low quality. In this paper we advance the research on incomplete algorithms for DCOPs by: (1) Proposing a version of the Max-sum algorithm that operates on an alternating directed acyclic graph (Maxsum_AD), which guarantees convergence in linear time. (2) Identifying major weaknesses of Max-sum and Max-sum_AD that cause inconsistent costs/utilities to be propagated and affect the assignment selection. (3) Solving the identified problems by introducing value propagation to Max-sum_AD. Our empirical study reveals a large improvement in the quality of the solutions produced by Max-sum_AD with value propagation (VP), when solving problems which include cycles, compared with the solutions produced by the standard Max-sum algorithm, Bounded Max-sum and Maxsum_AD with no value propagation.
Asymmetric Distributed Constraint Optimization
"... Abstract. The standard model of distributed constraints optimization problems (DCOPs), assumes that the cost of a constraint is checked by one of the agents involved in the constraint. For DCOPs this is equivalent to the assumption that each constraint has a global cost which applies to each of the ..."
Abstract
-
Cited by 4 (2 self)
- Add to MetaCart
(Show Context)
Abstract. The standard model of distributed constraints optimization problems (DCOPs), assumes that the cost of a constraint is checked by one of the agents involved in the constraint. For DCOPs this is equivalent to the assumption that each constraint has a global cost which applies to each of the participating agents and in other words that all constraints are symmetric. Many multi agent system (MAS) problems involve asymmetric constraints. For example, the gain from a scheduled meeting of two agents is naturally different for each of the participants. In order to solve Asymmetric DCOPs (ADCOPs), one needs to design algorithms in which all agents participating in a constraint independently check the gain for each of them. This naturally brings up the question of privacy, enabling agents to keep their cost (or gain) of constraints private, at least partially. The present paper presents search algorithms for ADCOPs which handle asymmetric constraints in a privacy preserving manner. New versions of Asynchronous Forward Bounding and of Synchronous Branch & Bound are proposed. In addition, two local search algorithms are presented in which agents negotiate moves prior to assigning values. All algorithms are empirically evaluated, and their performance in terms of run-time, network load and solution quality is presented.
ARGUS: A Coordination System to Provide First Responders with Live Aerial Imagery of the Scene of a Disaster (Demonstration
- In Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems
, 2012
"... We present ARGUS, a coordination system for unmanned aerial vehicles (UAVs) deployed to support situational awareness for disaster management settings. ARGUS is based on the max-sum algorithm, a well known decentralised coordination algorithm for multi-agent systems. In this demonstration, we presen ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
(Show Context)
We present ARGUS, a coordination system for unmanned aerial vehicles (UAVs) deployed to support situational awareness for disaster management settings. ARGUS is based on the max-sum algorithm, a well known decentralised coordination algorithm for multi-agent systems. In this demonstration, we present an interactive simulation environment, where a user acting as a first responder submits imagery collection tasks to a team of UAVs, which then use max-sum to assign themselves to the tasks. We then present a set of real flight tests, in which two Hexacopter UAVs again use ARGUS to coordinate over tasks. Our tests indicate that the system responds positively to the dynamism and the heterogeneity of the real world.
A Methodology for Deploying the Max-Sum Algorithm and a Case Study on Unmanned Aerial Vehicles
"... We present a methodology for the deployment of the maxsum algorithm, a well known decentralised algorithm for coordinating autonomous agents, for problems related to situational awareness. In these settings, unmanned autonomous vehicles are deployed to collect information about an unknown environmen ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
(Show Context)
We present a methodology for the deployment of the maxsum algorithm, a well known decentralised algorithm for coordinating autonomous agents, for problems related to situational awareness. In these settings, unmanned autonomous vehicles are deployed to collect information about an unknown environment. Our methodology then helps identify the choices that need to be made to apply the algorithm to these problems. Next, we present a case study where the methodology is used to develop a system for disaster management in which a team of unmanned aerial vehicles coordinate to provide the first responders of the area of a disaster with live aerial imagery. To evaluate this system, we deploy it on two unmanned hexacopters in a variety of scenarios. Our tests show that the system performs well when confronted with the dynamism and the heterogeneity of the real world. 1
Communication-constrained DCOPs: Message approximation in GDL with function filtering
- In AAMAS
, 2011
"... In this paper we focus on solving DCOPs in communication constrained scenarios. The GDL algorithm optimally solves DCOP problems, but requires the exchange of exponentially large messages which makes it impractical in such settings. Function filtering is a technique that alleviates this high communi ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
(Show Context)
In this paper we focus on solving DCOPs in communication constrained scenarios. The GDL algorithm optimally solves DCOP problems, but requires the exchange of exponentially large messages which makes it impractical in such settings. Function filtering is a technique that alleviates this high communication requirement while maintaining optimality. Function filtering involves calculating approximations of the exact cost functions exchanged by GDL. In this work, we explore different ways to compute such approximations, providing a novel method that empirically achieves significant communication savings.
Agent-Based Decentralised Coordination for Sensor Networks using the Max-Sum Algorithm
"... In this paper, we consider the generic problem of how a network of physically dis-tributed, computationally constrained devices can make coordinated decisions to maximise the effectiveness of the whole sensor network. In particular, we propose a new agent-based represen-tation of the problem, based ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
In this paper, we consider the generic problem of how a network of physically dis-tributed, computationally constrained devices can make coordinated decisions to maximise the effectiveness of the whole sensor network. In particular, we propose a new agent-based represen-tation of the problem, based on the factor graph, and use state-of-the-art DCOP heuristics (i.e., DSA and the max-sum algorithm) to generate sub-optimal solutions. In more detail, we for-mally model a specific real-world problem where energy-harvesting sensors are deployed within an urban environment to detect vehicle movements. The sensors coordinate their sense/sleep schedules, maintaining energy neutral operation while maximising vehicle detection probabil-ity. We theoretically analyse the performance of the sensor network for various coordination strategies and show that by appropriately coordinating their schedules the sensors can achieve significantly improved system-wide performance, detecting up to 50 % of the events that a ran-domly coordinated network fails to detect. Finally, we deploy our coordination approach in a realistic simulation of our wide area surveillance problem, comparing its performance to a number of benchmarking coordination strategies. In this setting, our approach achieves up to a 57 % reduction in the number of missed vehicles (compared to an uncoordinated network). This performance is close to that achieved by a benchmark centralised algorithm (simulated annealing) and to a continuously powered network (which is an unreachable upper bound for any coordination approach).
Regret-Based Multi-Agent Coordination with Uncertain Task Rewards
"... Many multi-agent coordination problems can be represented as DCOPs. Motivated by task allocation in disaster response, we extend standard DCOP models to consider uncertain task rewards where the outcome of completing a task depends on its current state, which is randomly drawn from unknown dis-tribu ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
Many multi-agent coordination problems can be represented as DCOPs. Motivated by task allocation in disaster response, we extend standard DCOP models to consider uncertain task rewards where the outcome of completing a task depends on its current state, which is randomly drawn from unknown dis-tributions. The goal of solving this problem is to find a solu-tion for all agents that minimizes the overall worst-case loss. This is a challenging problem for centralized algorithms be-cause the search space grows exponentially with the number of agents and is nontrivial for existing algorithms for standard DCOPs. To address this, we propose a novel decentralized al-gorithm that incorporates Max-Sum with iterative constraint generation to solve the problem by passing messages among agents. By so doing, our approach scales well and can solve instances of the task allocation problem with hundreds of a-gents and tasks.