DYNAMIC BELIEF NETWORKS FOR DISCRETE MONITORING

We describe the development of a monitoring system which uses sensor o bservation data about discrete events to construct dynamically a proba bilistic model of the world. This model is a Bayesian network incorpor ating temporal aspects, which we call a Dynamic Belief Network; it is used to reason under uncertainty about both the causes and consequence s of the events being monitored. The basic dynamic construction of the network is data-driven. However the model construction process combin es sensor data about events with externally provided information about agents' behavior, and knowledge already contained within the model, t o control the size and complexity of the network. This means that both the network structure within a time interval, and the amount of histo ry and detail maintained, can vary over time. We illustrate the system with the example domain of monitoring robot vehicles and people in a restricted dynamic environment using light-beam sensor data. In additi on to presenting a generic network structure for monitoring domains, w e describe the use of more complex network structures which address tw o specific monitoring problems, sensor validation and the Data Associa tion Problem.