Information Sharing in Large Heterogeneous Teams

61
Information Sharing in Large Heterogeneous Teams Prasanna Velagapudi Robotics Institute Carnegie Mellon University FRC Seminar - August 13, 2009

description

Information Sharing in Large Heterogeneous Teams. Prasanna Velagapudi Robotics Institute Carnegie Mellon University. Large Heterogeneous Teams. 100s to 1000s of agents (robots, agents, people) Shared goals Must collaborate to complete complex tasks Dynamic, uncertain environment. - PowerPoint PPT Presentation

Transcript of Information Sharing in Large Heterogeneous Teams

Page 1: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009

Information Sharing in Large Heterogeneous Teams

Prasanna VelagapudiRobotics Institute

Carnegie Mellon University

Page 2: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 2

Large Heterogeneous Teams• 100s to 1000s of agents

(robots, agents, people)• Shared goals• Must collaborate to

complete complex tasks• Dynamic, uncertain

environment

Page 3: Information Sharing in Large Heterogeneous Teams

Scaling Teams

• Far more data than can be feasibly shared– Amount of information exchanged often grows

faster than amount of available bandwidth• Vague, incomplete knowledge of large parts of

the team– Often not important

• Shared information improves team performance

Page 4: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 4

Search and Rescue

• Air robots, ground robots, human operators• Each is generating information

– Humans Classify objects and issue commands– Robots Explore and map area

• Geometric Random Graph

Page 5: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 5

Search and Rescue

VideoStreams

(320kbps x 24,For operators) Decentralized

Evidence Grid(14kbps x 24,For all agents)

OperatorControl

(<1kbps x 24,For robots)

O(N2)O(N2)

Available throughput: Θ(WN0.5) [Gupta 2000]

Page 6: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 6

Available Network Technologies

Source: William Webb - Ofcom

Page 7: Information Sharing in Large Heterogeneous Teams

Scaling Teams

• We need to deliver information efficiently– Get to the agents that can make use of it most– Don’t waste communication bandwidth

• Key Idea: Different agents have different needs for a given piece of information

Page 8: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 8

Sharing information

• When information generation exceeds network capacity, there are a few options:– Compression/Fusion (Eliminate redundant data)– Structuring (Eliminate overhead costs)– Selection (Eliminate unimportant data)

Page 9: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 9

Related work

• Distributed Data Fusion– Channel filtering (DDF) [Makarenko 04]– Particle exchange [Rosencrantz 03]

• Networking– Gossip[Haas 06], SPIN[Heinzelman 99], IDR[Liu 03]

• Multiagent Coordination– STEAM [Tambe 97]– ACE-PJB-COMM [Roth 05], Reward-shaping

[Williamson 09], dec-POMDP-com [Zilberstein 03]

Page 10: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 10

Domain assumptions

• Information generated dynamically and asynchronously

• Limited bandwidth and memory – With respect to size of team

• Significant local computing• Some predictive knowledge about other

agents’ information needs• Peer-to-peer communications

Page 11: Information Sharing in Large Heterogeneous Teams

11

Domain assumptions

FRC Seminar - August 13, 2009

Inconsistency

Complexity Communication

dec-POMDP-com

Gossip

SPIN, IDR

STEAM

Flooding

Particle ExchangeChannel Filter

Tokens

ACE-PJB-COMM

Reward Shaping

Our domains

Page 12: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 12

Abstract Problem

• Suppose we are given some metric for team performance in a domain:– How much information sharing complexity and

communication is necessary to achieve good performance in a large team?

– How can we characterize the effects of information sharing on performance in large teams?

• Suppose we are given some metric for team performance in a domain:– How much information sharing complexity and

communication is necessary to achieve good performance in a large team?

– How can we characterize the effects of information sharing on performance in large teams?

Page 13: Information Sharing in Large Heterogeneous Teams

13

A simple example

• Two robots (1 static, 1 mobile) in a maze• Limited sensing radius, global communication• Team task: Get mobile robot to goal point• Team performance = battery power

– Movement and communication use power

• How useful is it to the team for the static robot to share its info with the mobile robot?

FRC Seminar - August 13, 2009

Page 14: Information Sharing in Large Heterogeneous Teams

14

A simple example

FRC Seminar - August 13, 2009

Page 15: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 15

A simple example• Without information • With information

Page 16: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 16

A simple example• Without information • With information

The change in path cost is the “utility” of this information

Page 17: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 17

Utility of Information

• Utility: the change in team performance when an agent gets a piece of information

• Often dependent on other information • Difficult to calculate during execution, even

with complete real-time knowledge– Need to know final state of team

Page 18: Information Sharing in Large Heterogeneous Teams

18

Objective

• Utility: the change in team performance when an agent gets a piece of information

• Communication cost: the cost of sending a piece of information to a specific agent

FRC Seminar - August 13, 2009

Page 19: Information Sharing in Large Heterogeneous Teams

19

Objective

• Maximize team performance:

FRC Seminar - August 13, 2009

utility communication

agentsinfo. source

dissemination tree

In actual systems, this solution must be formed through local decisions!

Page 20: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 20

Distributions of Utility

• For large amounts of information, consider the distribution of utility– May be conditioned on known data, or just

independently sampled• Characterize domains as having specific

distributions of utility• Estimate performance of various algorithms as

function of this distribution

Page 21: Information Sharing in Large Heterogeneous Teams

21

Back to the simple example

FRC Seminar - August 13, 2009

Freq

uenc

y

Utility (Δ path cost)

Maze Utility Distribution

Page 22: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 22

Abstract Problem

• Suppose we are given some metric for team performance in a domain:– How much information sharing complexity and

communication is necessary to achieve good performance in a large team?

– How can we characterize the effects of information sharing on performance in large teams?

Page 23: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 23

Approach

• Useful information sharing algorithms fall between two extremes:– Full knowledge/high complexity (omniscient)– No knowledge/low complexity (blind)

• Observe performance of two extremes of information sharing algorithms– Learn when it is useful to use complex algorithms– If blind policies do well, other low complexity

algorithms will also work well

Page 24: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 24

Utility vs. Communication

Team

Util

ity

Communication Cost

Distributional upper bound

Omniscient policy

Blind policy

Efficient policies

Page 25: Information Sharing in Large Heterogeneous Teams

25

Expected Upper Bound

• Order statistic: expectation of k-th highest value over n samples– Computable for many common distributions

• Expected best case performance – What values of utility would we expect to see in a

team of n agents?– Sum of k highest order statistics

FRC Seminar - August 13, 2009

Page 26: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 26

Utility vs. Communication

Team

Util

ity

Communication Cost

Distributional upper bound

Omniscient policy

Blind policy

Efficient policies

Page 27: Information Sharing in Large Heterogeneous Teams

27

Omniscient Policy

• Lookahead policy1. Assume we are given estimate of utility for every

other node (possibly with noise)2. Exhaustively search all n-length paths from current

node3. Send information along best path4. Repeat until TTL reaches 0

– Approximation of best omniscient policy – Full exhaustive search is intractable

FRC Seminar - August 13, 2009

Page 28: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 28

Utility vs. Communication

Team

Util

ity

Communication Cost

Distributional upper bound

Omniscient policy

Blind policy

Efficient policies

Page 29: Information Sharing in Large Heterogeneous Teams

29

Blind policies• Random: “Gossip” to randomly chosen neighbor

• Random Self-Avoiding– Keep history of agents visited– O(lifetime of piece)

• Random Trail– Keep history of links used– O(# of pieces/time step)

FRC Seminar - August 13, 2009

Page 30: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 30

Questions

• How well does the lookahead policy approximate omniscient policy performance?

• How wide is the performance gap between the omniscient policy and blind policies?

• How does team size affect performance?• Is omniscient policy performance better

because it knows where to route, or where not to route?

Page 31: Information Sharing in Large Heterogeneous Teams

31

Experiment

• Network of agents with utility sampled from distribution

• Single piece of information shared each trial• Average-case performance recorded

FRC Seminar - August 13, 2009

Distributions:• Normal• Exponential• Uniform

Networks:• Small-Worlds (Watts-Beta)• Scale-free (Preferential attachment)• Lattice (2D grid)• Hierarchy (Spanning tree)

Page 32: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 32

Questions

• How well does the lookahead policy approximate omniscient policy performance?

• How wide is the performance gap between the omniscient policy and blind policies?

• How does team size affect performance?• Is omniscient policy performance better

because it knows where to route, or where not to route?

Page 33: Information Sharing in Large Heterogeneous Teams

33

Lookahead convergence

FRC Seminar - August 13, 2009

2-step lookahead: pathological case?

Page 34: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 34

Questions

• How well does the lookahead policy approximate omniscient policy performance?

• How wide is the performance gap between the omniscient policy and blind policies?

• How does team size affect performance?• Is omniscient policy performance better

because it knows where to route, or where not to route?

Page 35: Information Sharing in Large Heterogeneous Teams

35

Performance Results

FRC Seminar - August 13, 2009

Normal Distribution Exponential Distribution

Page 36: Information Sharing in Large Heterogeneous Teams

36

Policy Performance

FRC Seminar - August 13, 2009

(Utility sampled from Exponential distribution)

Page 37: Information Sharing in Large Heterogeneous Teams

37

Utility of knowledge

FRC Seminar - August 13, 2009

~120 communications

Page 38: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 38

Questions

• How well does the lookahead policy approximate omniscient policy performance?

• How wide is the performance gap between the omniscient policy and blind policies?

• How does team size affect performance?• Is omniscient policy performance better

because it knows where to route, or where not to route?

Page 39: Information Sharing in Large Heterogeneous Teams

39

Scaling effects

FRC Seminar - August 13, 2009

The costs of maintaining utility estimates for Lookahead increase with team size, but the costs of Random policy do not.

Page 40: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 40

Questions

• How well does the lookahead policy approximate omniscient policy performance?

• How wide is the performance gap between the omniscient policy and blind policies?

• How does team size affect performance?• Is omniscient policy performance better

because it knows where to route, or where not to route?

Page 41: Information Sharing in Large Heterogeneous Teams

41

Noisy estimation

• How does the omniscient policy degrade as its estimates of utility become noisy?

• As noise increases, the omniscient policy approaches an ideal blind policy

• Gaussian noise scaled by network distance:

FRC Seminar - August 13, 2009

Page 42: Information Sharing in Large Heterogeneous Teams

42

Noisy estimation

FRC Seminar - August 13, 2009

Page 43: Information Sharing in Large Heterogeneous Teams

43

Modeling maze navigation

FRC Seminar - August 13, 2009

Freq

uenc

y

Utility (Δ path cost)

Page 44: Information Sharing in Large Heterogeneous Teams

44

Modeling maze navigation

FRC Seminar - August 13, 2009

Page 45: Information Sharing in Large Heterogeneous Teams

45

Summary of Results

• Omniscient policy approaches optimal routing on many graphs (not hierarchies)

• Gap between omniscient and blind policies is small when:– Network is conducive (Small Worlds, Lattice)– Maintaining shared knowledge is expensive– Network is massive– Estimation of value is poor

FRC Seminar - August 13, 2009

Page 46: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 46

Improving the model

• Current work on validating this model– USARSim (Search and Rescue)– VBS2 (Military C2)– TREMOR (POMDP)

• Predictive utility estimation and dynamics• Better solution for optimal policy:

– Prize-collecting Steiner Tree [Ljubić 2007]

Page 47: Information Sharing in Large Heterogeneous Teams

47

Conclusions

• Utility distributions: a mechanism to test information sharing performance– Computable from real-world data– Can be conditional/joint/marginal to encode

domain dependencies• Simple random policies: surprisingly

competitive in many cases– No structural or computational overhead– No expensive costs to maintain utility estimates

FRC Seminar - August 13, 2009

Page 48: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 48

Questions?

Page 49: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 49

Page 50: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 50

Outline

• What we mean by large heterogeneous teams• The common assumptions in our domains• What we mean by utility utility distributions• The experiment• The results• Conclusions• Future work/validation

Page 51: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 51

We need information

• Information generated all over network• Information consumed all over network• Team performance is improved by additional

information– More data = better decisions

• However, information loss degrades performance gracefully– Less data = alright decisions

Page 52: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 52

Scalability of Large Teams

• As size increases, amount of information exchanged grows faster than amount of available bandwidth– Constant network density: O(n)

Page 53: Information Sharing in Large Heterogeneous Teams

53

Motivation

• Large, heterogeneous teams of agents– 100s to 1000s of robots, agents, and people– Must collaborate to complete complex tasks– Decentralized algorithms

FRC Seminar - August 13, 2009

Page 54: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 54

Motivation• Agents need to share information about objects and

uncertainty in the environment to perform roles– Individual sensor readings unreliable– Used to reason about appropriate actions– Maintenance of mutual beliefs is key

• Need effective means to propagate information– Agent needs for information change dynamically– Highly redundant data

Page 55: Information Sharing in Large Heterogeneous Teams

55

Utility of Information

• A given piece of data can improve a given agent’s performance by a certain amount– Need to determine which pieces are useful to

deliver to which agents– Need to determine how a piece of information will

affect team performance

FRC Seminar - August 13, 2009

Page 56: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 56

Utility of Information

• In our domains, we want to maximize the utility of what we are sending around while minimizing the cost of communication

• There are many possible information sharing strategies, how can we estimate or predict their performance?

Page 57: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 57

USARSim

• In search and rescue/disaster response, network communication is very limited, while information generated must be shipped elsewhere to be processed.

• Video and map information can be compressed, but compression is limited because data must be streamed to operators

• Also, as more autonomous vehicles are added, it becomes impossible for single operators to handle all the information anyway

Page 58: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 58

VBS2

• In military C2, high-level decisions must be made based on available information from a large number of units.

• However, military communications are especially limited, and further constrained by hierarchical organization and classification

• Can we intelligently guarantee that information will get between units and to command units?

Page 59: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 59

TREMOR

• Varakantham et al. present a multiagent POMDP solver that uses reward shaping to decompose joint POMDPs into local POMDPs in situations where most interaction occurs at a small number of “coordination locales”.

• The reward shaping component can be described as an intelligent information sharing problem, and as such, we can create a distributed variant capable of solving much larger multi-agent POMDPs

Page 60: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 60

Page 61: Information Sharing in Large Heterogeneous Teams

FRC Seminar - August 13, 2009 61