Advanced Sensor Models: Benefits for Target Tracking and Sensor Data Fusion

Modern sensor systems are typically characterized by advanced signal processing techniques which have direct impact on the quantitative and qualitative properties of the sensor data produced. This makes a more advanced modeling of the statistical characteristics of the sensor output inevitable. Via constructing appropriate likelihood functions based on these models the performance of Bayesian tracking and sensor data fusion techniques can be much improved. The proposed paper discusses the benefits by selected examples from various applications