A fusion method that performs better than best sensor
In multiple sensor systems, it is generally known that a good fuser will outperform the best sensor, and on the other hand, an inappropriate fuser can perform worse than the worst sensor. If the error distributions of the sensors are precisely known, an optimal fuser--that performs at least as well as best sensor--can be designed using statistical estimation methods. In engineering and robotic systems, however, it is too difficult and expensive to derive closed form error distributions required by these methods. This problem is further compounded by the variety and complexity of present day sensor systems, wherein a number of sensing hardware units and computing modules could be integrated into a single sensor. In these systems, however, it is possible to collect sensor data by sensing objects with known parameters. Thus, it is very important to have sample-based methods that enable a fuser to perform at least as well as the best sensor. In this paper, the author presents a generic analytical formulation of this problem, and provide a very simple property that yields such fuser.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Energy Research, Washington, DC (United States); Oak Ridge National Lab., TN (United States); Office of Naval Research, Washington, DC (United States)
- DOE Contract Number:
- AC05-96OR22464
- OSTI ID:
- 654156
- Report Number(s):
- ORNL/CP-97085; CONF-980725-; ON: DE98004934; BR: 233001000; KC0401030; WJ07101; TRN: AHC2DT05%%246
- Resource Relation:
- Conference: 1. international conference on multisource-multisensor information fusion, Las Vegas, NV (United States), 6-9 Jul 1998; Other Information: PBD: Mar 1998
- Country of Publication:
- United States
- Language:
- English
Similar Records
Preface to foundations of information/decision fusion with applications to engineering problems
Projective Method for Generic Sensor Fusion Problem