In a multisensor tracking system, sensors often operate asynchronously and provide data at different rates with different communication delays. In this case the sequential processing of the sensor data may be computationally intensive. In addition, due to the inherent delay associated with some sensors, such as a multi-tasking radar, it may not be possible to process the data real time. The present paper presents an approach to deal with asynchronous sensors' data without performing sequential processing, i.e., using a data fusion approach. The result of the paper applies to an arbitrary number of dissimilar sensors.
展开▼