Integrated Joint Probabilistic Data Association and Interactive Multiple Model Filter
Nagavenkat Adurthi, Taewook Lee · AIAA Scitech 2021 Forum · 2021
View Video Presentation: https://doi.org/10.2514/6.2021-1756.vid In this paper we develop the integrated Joint Probabilistic Data Association (JPDA) and Interactive Multiple Model (IMM) Filter. This combined filter is developed for applications where the targets can change their dynamical behavior and the measurements have association ambiguity. The potential applications involve airplane tracking, satellite tracking, vehicle and pedestrian tracking for autonomous vehicles. We provide the complete derivation with intuitive insights and the algorithmic implementation details. Further, to improve accuracy, the filters are developed to use sigma or quadrature points. Finally, multi-target tracking examples are used to illustrate the working and efficacy of the integrated JPDA and IMM filter. An example of tracking airplanes with measurement association ambiguity is used to demonstrate the combined filter using the Extended Kalman filter, Unscented Kalman Filter and the recently developed Conjugate Unscented Transform points.