Neural Net Sensor Fusion
Abstract
A generic architecture for neural net multisensor data fusion is introduced and analyzed. The architecture consists of a set of independent sensor neural nets, one for each sensor, coupled to a fusion net. Each sensor is trained (from a representative data set of the particular sensor) to map to a hypothesis space output. The decision outputs from the sensor nets are used to train the fusion net to an overall decision. In this report the sensor fusion architecture is applied to the stochastic exclusive-or problem for a benchmark comparison with classical hypothesis testing. The architecture is also applied to a data fusion experiment involving the multisensor observation of object deployments during the recent Firefly launches. The deployments were measured simultaneously by X and L band and CO laser radars. The range Doppler images from the X band and CO2 laser radars were combined with a passive IR spectral simulation of the deployment to form the data inputs to the neural sensor fusion system. The network was trained to distinguish predeployment, deployment, and postdeployment phases of the launch based on the fusion of these sensors. The success of the system in utilizing sensor synergism for an enhanced deployment detection is clearly demonstrated.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 05, 1991
- Accession Number
- ADA241765
Entities
People
- R. Y. Levine
- T. S. Khuon
Organizations
- Massachusetts Institute of Technology