PHD Filter for Object Tracking in Road Traffic Applications Considering Varying Detectability

This paper considers the object detection and tracking problem in a road traffic situation from participant’s perspective. The information source is an automotive radar which attached to ego vehicle. scenario characteristics are varying visibility due occlusion multiple detections of vehicle during scanning interval. goal maintain report state undetected though possibly present objects. proposed algorithm based on multi-object Probability Hypothesis Density filter. Because PHD filter has no memory, estimate number objects can change abruptly erroneous detections. To reduce this effect, we model calculate state-dependent probability. Thus, unnoticed but probably valid hypotheses for more extended period. We use sequential Monte Carlo method with clustering implementing distinguish between detected, undetected, hidden particles within our framework, whose purpose track likely performance demonstrated using highway measurements.