Expert Systems with Applications: An International Journal
| Hi-index | 0.00 |
Data fusion algorithms have a very wide range of applications in some fields. But, with the growing sensor numbers in multi-sensor target tracking systems, data fusion algorithms using conventional Kalman filter meet problems such as heavy computational burden and poor robustness. Decentralized data fusion algorithms using information filter provide a way of avoiding traditional fusion algorithms' limitations. The work described in this paper aims to develop a decentralized fusion algorithm for multi-sensor target tracking problems. The basic principle of the information filter is introduced. A decentralized data fusion algorithm using information filter is developed. This algorithm is then demonstrated on a multi-senor tracking example.