Abstract | ||
---|---|---|
In a practical multi-sensor fusion target tracking system, the measurement noise of different sensors is often correlated. By using the Cholesky factorization and inverse calculation method for unit lower triangular matrix, multi-sensor measurements with correlated measurement noises are transformed to equivalent pseudo ones with uncorrelated measurement noises; then based on the Kalman filtering, a new multisensor centralized fusion target tracking algorithm with correlated measurement noises is proposed. Compared with the existing centralized fusion algorithm and the centralized fusion algorithm which uses the measurements of original sensors directly, they are equivalent in computational accuracy, but the new one reduces the computational complexity greatly. Monte Carlo simulation results are provided to demonstrate the validity of the new algorithm further. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1109/ICSMC.2004.1399800 | SMC (2) |
Keywords | Field | DocType |
kalman filters,computational complexity reduction,monte carlo simulation,target tracking,matrix algebra,unit lower triangular matrix,data fusion,computational complexity,cholesky factorization,monte carlo methods,inverse calculation method,optimal multisensor fusion target tracking,centralized fusion algorithm,correlated measurement noises,kalman filtering,sensor fusion,kalman filter | Monte Carlo method,Mathematical optimization,Computer science,Control theory,Matrix decomposition,Sensor array,Algorithm,Tracking system,Kalman filter,Sensor fusion,Cholesky decomposition,Computational complexity theory | Conference |
Volume | ISSN | ISBN |
2 | 1062-922X | 0-7803-8566-7 |
Citations | PageRank | References |
4 | 0.48 | 1 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zhansheng Duan | 1 | 187 | 28.65 |
Chongzhao Han | 2 | 446 | 71.68 |
Tangfei Tao | 3 | 24 | 5.99 |