Agent-Based Coordination of Sensor Networks

28
Agent-Based Coordination of Sensor Networks Alex Rogers School of Electronics and Computer Science University of Southampton [email protected]

description

Agent-Based Coordination of Sensor Networks. Alex Rogers School of Electronics and Computer Science University of Southampton [email protected]. Overview. Decentralised Coordination Landscape of Algorithms Optimality vs Communication Costs Local Message Passing Algorithms - PowerPoint PPT Presentation

Transcript of Agent-Based Coordination of Sensor Networks

Page 1: Agent-Based Coordination of Sensor Networks

Agent-Based Coordination of Sensor Networks

 Alex Rogers

School of Electronics and Computer ScienceUniversity of Southampton

[email protected]

Page 2: Agent-Based Coordination of Sensor Networks

Overview

• Decentralised Coordination• Landscape of Algorithms

– Optimality vs Communication Costs• Local Message Passing Algorithms

– Max-sum algorithm– Graph Colouring

• Example Application– Wide Area Surveillance Scenario

• Future Work & Sensor Testbed

Page 3: Agent-Based Coordination of Sensor Networks

Decentralised Coordination

Agents

• Multiple conflicting goals and objectives• Discrete set of possible actions• Some locality of interaction

Page 4: Agent-Based Coordination of Sensor Networks

Decentralised Coordination

Sensors

• Multiple conflicting goals and objectives• Discrete set of possible actions• Some locality of interaction

Page 5: Agent-Based Coordination of Sensor Networks

Decentralised Coordination

Agents

• Multiple conflicting goals and objectives• Discrete set of possible actions• Some locality of interaction

Page 6: Agent-Based Coordination of Sensor Networks

Decentralised Coordination

Agents

Central point of controlDecentralised control and coordination through local computation and message passing.• Speed of convergence, guarantees of optimality,

communication overhead, computability

No direct communication Solution scales poorly Central point of failure Who is the centre?

Page 7: Agent-Based Coordination of Sensor Networks

Landscape of Algorithms

Complete Algorithms

DPOPOptAPOADOPT

Communication Cost

Optimality

Probability Collectives

Iterative Algorithms

Best Response (BR)Distributed Stochastic

Algorithm (DSA) Fictitious Play (FP)

Message Passing

Algorithms

Sum-ProductAlgorithm

Page 8: Agent-Based Coordination of Sensor Networks

Sum-Product Algorithm

Variable nodes

Function nodes

Factor Graph

A simple transformation:

allows us to use the same algorithms to maximise social welfare:

Find approximate solutions to global optimisation through local computation and message passing:

Page 9: Agent-Based Coordination of Sensor Networks

Graph Colouring

Agentfunction / utility

variable / state

Graph Colouring Problem Equivalent Factor Graph

Page 10: Agent-Based Coordination of Sensor Networks

Graph Colouring

Equivalent Factor Graph

Utility Function

Page 11: Agent-Based Coordination of Sensor Networks

Max-Sum Calculations

Variable to Function: Information aggregation

Function to Variable: Marginal Maximisation

Decision:Choose state that maximises

sum of all messages

Page 12: Agent-Based Coordination of Sensor Networks

Graph Colouring

Page 13: Agent-Based Coordination of Sensor Networks

Graph Colouring

Page 14: Agent-Based Coordination of Sensor Networks

Optimality

Page 15: Agent-Based Coordination of Sensor Networks

Communication Cost

Page 16: Agent-Based Coordination of Sensor Networks

Robustness to Message Loss

Page 17: Agent-Based Coordination of Sensor Networks

Hardware Implementation

Page 18: Agent-Based Coordination of Sensor Networks

Energy-Aware Sensor Networks

Page 19: Agent-Based Coordination of Sensor Networks

Wide Area Surveillance Scenario

Dense deployment of sensors to detect pedestrian and vehicle activity within an urban environment.

Unattended Ground Sensor

Page 20: Agent-Based Coordination of Sensor Networks

Energy Constrained Sensors

Maximise event detection whilst using energy constrained sensors:– Use sense/sleep duty cycles

to maximise network lifetime of maintain energy neutral operation.

– Coordinate sensors with overlapping sensing fields.

time

duty cycle

time

duty cycle

Page 21: Agent-Based Coordination of Sensor Networks

Energy-Aware Sensor Networks

Page 22: Agent-Based Coordination of Sensor Networks

Energy-Aware Sensor Networks

Page 23: Agent-Based Coordination of Sensor Networks

Empirical Evaluation

Page 24: Agent-Based Coordination of Sensor Networks

Autonomous Mobile Sensors

Page 25: Agent-Based Coordination of Sensor Networks

Future Work• Continuous action spaces

– Not limited to discrete actions

• Bounded Solutions– Prune edges from the cyclic

factor graph to reveal a tree– Run Max-Sum on this tree– Calculate a bound on how far

this solution is from the real optimal solution Factor Graph

Page 26: Agent-Based Coordination of Sensor Networks

Publications

• Farinelli, A., Rogers, A., Petcu, A. and Jennings, N. R. (2008) Decentralised Coordination of Low-Power Embedded Devices Using the Max-Sum Algorithm. In: Proceedings of the Seventh International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS-08), Estoril, Portugal.

• Waldock, A., Nicholson, D. and Rogers, A. (2008) Cooperative Control using the Max-Sum Algorithm. In: Proceedings of the Second International Workshop on Agent Technology for Sensor Networks, Estoril, Portugal.

• Farinelli, A., Rogers, A. and Jennings, N. (2008) Maximising Sensor Network Efficiency Through Agent-Based Coordination of Sense/Sleep Schedules. In: Proceedings of the Workshop on Energy in Wireless Sensor Networks in conjunction with DCOSS 2008, Santorini, Greece.

Page 27: Agent-Based Coordination of Sensor Networks

SunSPOT Network

• Chipcon 2431 SoC– 8051 processor, 8KB RAM

• SunSPOT network– Java enabled, 180 MHz

32bit ARM– Accelerometers, light,

temperature sensors– Programming over-the-air

Page 28: Agent-Based Coordination of Sensor Networks

Questions?