GM-PHD Filter Based Sensor Data Fusion for Automotive Frontal Perception System

Advanced driver assistance systems and highly automated driving functions require an enhanced frontal perception system. The requirements of a environment system cannot be satisfied by either the existing automotive sensors. A commonly used sensor cluster for these consists mono-vision smart camera radar. fusion is intended to combine data sensors perform robust perception. Multi-object tracking algorithms have suitable software architecture fusion. Several multi-object algorithms, such as JPDAF or MHT, good performance; however, computational are significant according their combinatorial complexity. GM-PHD filter straightforward algorithm with favorable runtime characteristics that can track unknown time-varying number objects. However, conventional has poor performance in object cardinality estimation. This paper proposes method extends birth model relies on detections extraction module, including Bayesian estimation objects’ existence probability compensate drawbacks algorithm.