Distributed Estimation, Inferencing and Multi-Sensor Data Fusion for Real Time Supervisory Control

Abstract Fully autonomous or supervisory controlled guided vehicles that utilise on-board intelligent sensing to determine the vehicle's state, the external world, correlate real time events/objects with mapped knowledge, monitor the vehicle's own system health, and compute dynamically its own control strategy, require the use of a wide range of sensors, and the means to fuse or integrate disparate sensor databases when they refer to the same object. In this review paper we consider a multilevel approach to sensory integration for AGVs; level 1 local positional estimation, level 2 sensory consensus, level 3 sensor fusion, level 4 situation assessment. And consider for statistically rich data sources (such as radar, sonar) Bayesian distributed data fusion, whereas for substantialy uncertain events we review fuzzy logic, Dempster-Shafer evidential theories, and finally consider non-monotonic Al methods such as endorsements or explanation based reasoning for dealing with symbolic fusion.