Download presentation
Presentation is loading. Please wait.
1
Decentralised Coordination of Mobile Sensors using the Max-Sum Algorithm Ruben Stranders, Alex Rogers, Nick Jennings School of Electronics and Computer Science University of Southampton {rs06r, acr, nrj}@ecs.soton.ac.uk
2
2 This presentation focuses on the use of Max-Sum to coordinate mobile sensors Sensor Architecture & Max-Sum Empirical Evaluation Speeding up Max-Sum Model Value Coordinate
3
This work can be applied to improve situational awareness in dynamic scenarios Disaster Response Military Surveillance Climate Research
4
Our contribution is a coordination mechanism for a team of autonomous mobile sensors
5
These mobile sensors continuously monitor spatial phenomena
6
The main challenge is to coordinate the sensors in order to the state of these spatial phenomena
8
Limited Communication
9
The main challenge is to coordinate the sensors in order to the state of these spatial phenomena No centralised control
10
The main challenge is to coordinate the sensors in order to the state of these spatial phenomena No centralised control
11
To solve this coordination problem, we had to address three challenges 1.How to model the phenomena? 2.How to value potential samples? 3.How to coordinate to gather samples of highest value?
12
The three central challenges are clearly reflected in the architecture of our sensing agents Samples sent to neighbouring agents Samples received from neighbouring agents Information processing Model of Environment Outgoing negotiation messages Incoming negotiation messages Value of potential samples Action Selection Move Samples from own sensor Sensing Agent Raw samples Model Value Coordinate
13
These three challenges are clearly reflected in the architecture of our sensing agents Samples sent to neighbouring agents Samples received from neighbouring agents Information processing Model of Environment Outgoing negotiation messages Incoming negotiation messages Value of potential samples Action Selection Move Samples from own sensor Sensing Agent Raw samples Model
14
The sensors model the spatial phenomenon using the Gaussian Process Weak Strong Spatial Correlations
15
The sensors model the spatial phenomenon using the Gaussian Process Areas of Rapid Change
16
The sensors model the spatial phenomenon using the Gaussian Process Weak Strong Temporal Correlations
17
The value of a sample is determined how much it reduces uncertainty Samples sent to neighbouring agents Samples received from neighbouring agents Information processing Model of Environment Outgoing negotiation messages Incoming negotiation messages Value of potential samples Action Selection Move Samples from own sensor Sensing Agent Raw samples Value
18
The value of a sample is based on how much it reduces uncertainty But how to determine uncertainty reduction before collecting a sample?
19
The value of a sample is based on how much it reduces uncertainty But how to determine uncertainty reduction before collecting a sample? Prediction Confidence Interval Collected Sample Gaussian Process not only gives predictions, but also confidence intervals
20
The value of a sample is based on how much it reduces uncertainty But how to determine uncertainty reduction before collecting a sample? Prediction Confidence Interval Collected Sample Gaussian Process not only gives predictions, but also confidence intervals Potential Sample Location
21
The value of a sample is based on how much it reduces uncertainty But how to determine uncertainty reduction before collecting a sample? Prediction Confidence Interval Collected Sample Gaussian Process not only gives predictions, but also confidence intervals Measure of uncertainty
22
The value of a sample is based on how much it reduces uncertainty Prediction Confidence Interval Collected Sample Specifically, we can use information metrics such as Entropy, or Mutual Information
23
The sensor agents coordinate using the Max-Sum algorithm Samples sent to neighbouring agents Samples received from neighbouring agents Information processing Model of Environment Outgoing negotiation messages Incoming negotiation messages Value of potential samples Action Selection Move Samples from own sensor Sensing Agent Raw samples Coordinate
24
24 Max-Sum is a powerful algorithm for solving DCOPs Complete Algorithms DPOP OptAPO ADOPT Communication Cost Iterative Algorithms Best Response (BR) Distributed Stochastic Algorithm (DSA) Fictitious Play (FP) Max-Sum Algorithm Optimality
25
Max-Sum solves the social welfare maximisation problem in a decentralised way Mobile Sensors
26
Max-Sum solves the social welfare maximisation problem in a decentralised way Movement Parameters
27
Max-Sum solves the social welfare maximisation problem in a decentralised way Utility Functions
28
Max-Sum solves the social welfare maximisation problem in a decentralised way Localised Interaction
29
Max-Sum solves the social welfare maximisation problem in a decentralised way Social welfare: Mobile Sensors
30
The input for the Max-Sum algorithm is a graphical representation of the problem: a Factor Graph Variable nodes Function nodes Agent 1 Agent 2 Agent 3
31
Max-Sum solves the social welfare maximisation problem by message passing Variable nodes Function nodes Agent 1 Agent 2 Agent 3
32
Max-Sum solves the social welfare maximisation problem by message passing From variable i to function j From function j to variable i
33
To use Max-Sum, we encode the mobile sensor coordination problem as a factor graph Sensor 1 Sensor 2 Sensor 3 Sensor 1 Sensor 2 Sensor 3
34
Variables represent the sensors’ movements Sensor 1 Sensor 2 Sensor 3
35
Functions represent the uncertainty reduction that results from collecting a sample Sensor 1 Sensor 2 Sensor 3
36
Functions represent the uncertainty reduction that results from collecting a sample Sensor 1 Sensor 2 Sensor 3
37
Functions represent the uncertainty reduction that results from collecting a sample Sensor 1 Sensor 2 Sensor 3
38
Functions represent the uncertainty reduction that results from collecting a sample Sensor 1 Sensor 2 Sensor 3
39
Functions represent the uncertainty reduction that results from collecting a sample Sensor 1 Sensor 2 Sensor 3
40
Unfortunately, the straightforward application of Max-Sum is too computationally expensive From variable i to function j From function j to variable i
41
Unfortunately, the straightforward application of Max-Sum is too computationally expensive From variable i to function j From function j to variable i Bottleneck!
42
Therefore, we developed two general pruning techniques that speed up Max-Sum Goal: Make as small as possible
43
Therefore, we developed two general pruning techniques that speed up Max-Sum Goal: Make as small as possible 1.Try to prune the action spaces of individual sensors 2.Try to prune joint actions
44
The first pruning technique prunes individual actions by identifying dominated actions
45
1. Neighbours send bounds ↑ [2, 2] ↓ [1, 1] ↑ [5, 6] ↓ [0, 1] ↑ [1, 2] ↓ [3, 4]
46
The first pruning technique prunes individual actions by identifying dominated actions 2. Bounds are summed [8, 10] [4, 7]
47
The first pruning technique prunes individual actions by identifying dominated actions 3. Dominated actions are pruned [8, 10] [4, 7]
48
We developed two general pruning techniques that speed up Max-Sum Goal: Make as small as possible 1.Try to prune the action spaces of individual sensors 2.Try to prune joint actions
49
The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Action Sensor 1Sensor 2Sensor 3
50
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
51
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
52
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
53
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
54
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
55
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3
56
Action The second pruning technique reduces the joint action space because exhaustive enumeration is too costly Sensor 1Sensor 2Sensor 3 Etcetera…
57
The second pruning technique prunes the joint action space using branch and bound Sensor 1 Sensor 2 Sensor 3
58
The second pruning technique prunes the joint action space using branch and bound [7, 13][0, 4][2, 6] Sensor 1 Sensor 2 Sensor 3
59
The second pruning technique prunes the joint action space using branch and bound [7, 13][0, 4][2, 6] Sensor 1 Sensor 2 Sensor 3
60
The second pruning technique prunes the joint action space using branch and bound 91078 [7, 13][0, 4][2, 6] Sensor 1 Sensor 2 Sensor 3
61
The second pruning technique prunes the joint action space using branch and bound 91078 [7, 13][0, 4][2, 6] Sensor 1 Sensor 2 Sensor 3
62
This demonstration shows four sensors monitoring a spatial phenomenon
63
Sensors
64
This demonstration shows four sensors monitoring a spatial phenomenon Uncertainty Contours
65
This demonstration shows four sensors monitoring a spatial phenomenon
66
To empirically evaluate our algorithm, we measured speed up and prediction error Uncertainty Contours 91078 [7, 13][0, 4][2, 6]
67
The two pruning techniques combined prune 95% of the action space with 6 neighbouring sensors Number of neighbouring sensors % of joint actions pruned
68
Average RMSE Our Algorithm reduces Root Mean Squared Error of predictions up to 50% compared to Greedy
69
In conclusion, the use of Max-Sum leads to an effective coordination algorithm for mobile sensors 1. Decentralised
70
In conclusion, the use of Max-Sum leads to an effective coordination algorithm for mobile sensors 1. Decentralised 2. Fast % Pruned
71
In conclusion, the use of Max-Sum leads to an effective coordination algorithm for mobile sensors 1. Decentralised 2. Fast 3. Accurate predictions % Pruned Prediction Error
72
For future work, we wish to extend the algorithm to do non-myopic planning
73
References R. Stranders, A. Farinelli, A. Rogers and N.R. Jennings (2009): Decentralised Coordination of Mobile Sensors Using the Max-Sum Algorithm. In: Proc 21st Int. Joint Conf on AI (IJCAI), Pasadena, USA. (In Press) R. Stranders, A. Farinelli, A. Rogers and N.R. Jennings (2009): Decentralised Coordination of Continuously Valued Control Parameters using the Max-Sum Algorithm. 8th Proc. Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS), Budapest. (In Press)
74
In conclusion, the use of Max-Sum leads to an effective coordination algorithm for mobile sensors 1. Decentralised 2. Fast 3. Accurate predictions % Pruned Prediction Error Questions?
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.