The Greedy Dirichlet Process Filter - An Online Clustering Multi-Target Tracker
This addresses collision avoidance for autonomous vehicles, though it appears incremental as it builds on existing Dirichlet Processes and SUGS methods.
The paper tackles the problem of real-time state estimation for an unknown number of static and dynamic objects in autonomous driving by proposing the Greedy Dirichlet Process Filter (GDPF), which eliminates the need for prior clustering or data association and outperforms other trackers in accuracy and stability.
Reliable collision avoidance is one of the main requirements for autonomous driving. Hence, it is important to correctly estimate the states of an unknown number of static and dynamic objects in real-time. Here, data association is a major challenge for every multi-target tracker. We propose a novel multi-target tracker called Greedy Dirichlet Process Filter (GDPF) based on the non-parametric Bayesian model called Dirichlet Processes and the fast posterior computation algorithm Sequential Updating and Greedy Search (SUGS). By adding a temporal dependence we get a real-time capable tracking framework without the need of a previous clustering or data association step. Real-world tests show that GDPF outperforms other multi-target tracker in terms of accuracy and stability.