Skip to main content
Have a personal or library account? Click to login
Optimal placement of mobile sensors for data assimilations Cover

Optimal placement of mobile sensors for data assimilations

By:  and    
Open Access
|Dec 2012

Figures & Tables

Fig. 1. 

Illustrative graph of nodes in discretisation and sensor locations.

Fig. 2. 

A trajectory with sensor locations.

Fig. 3. 

Sensor locations: equally spaced (blue) and optimal locations (red).

Fig. 4. 

The scatter plot of ρ/ɛ vs. .

Fig. 5. 

The scatter plot of ρ/ɛ vs. ..

Table 1. Summary of Monte Carlo experiments

Unobservability index, ρ/ɛ RMSE of u a (0) RMSE of u a (·) Equally spaced sensors 12.92 0.1647 0.0788 Optimal sensor location 1.75 0.0786 0.0325 Improvement 86% 52% 58%
Fig. 6. 

RMSE as a function of time. Curve: equally spaced sensors. Dash: optimal sensor locations.

Fig. 7. 

The ±2σ interval around the initial state. Curve: equally spaced sensors. Dash: optimal sensor locations.

Table 2. Summary of a robustness study. The column under % represents the improvement in accuracy by using optimal sensor locations

Variation of Variation of R RMSE of u a (·) % Equally spaced 10% 0 0.0938 Optimal locations 10% 0 0.0351 62 Equally spaced 50% 0 0.1191 Optimal locations 50% 0 0.0887 25 Equally spaced 0 100% 0.0646 Optimal locations 0 100% 0.0407 37

Table 3. Summary of Monte Carlo experiments using uniform distributions

Observability, ρ/ɛ RMSE of u a (0) RMSE of u a (·) Equally spaced sensors 12.92 0.1376 0.0645 Optimal sensor location 1.75 0.0893 0.0372 Improvement 86% 35% 42%
Language: English
Page range: 17133 - 17133
Submitted on: Jan 2, 2012
Accepted on: Sep 11, 2012
Published on: Dec 1, 2012
Published by: Stockholm University Press
In partnership with: Paradigm Publishing Services

© 2012 Wei Kang, Liang Xu, published by Stockholm University Press
This work is licensed under the Creative Commons Attribution 4.0 License.