Scalable nonparametric Bayesian learning for heterogeneous and dynamic velocity fields
This work addresses the need for scalable analysis of complex traffic patterns for applications like autonomous driving, though it is incremental as it builds on existing nonparametric Bayesian methods.
The paper tackles the problem of learning heterogeneous and dynamic velocity fields from spatio-temporal data, such as for autonomous vehicle navigation, by developing a nonparametric Bayesian model and a scalable approximate inference method, demonstrating effectiveness on simulated data and the NGSIM dataset.
Analysis of heterogeneous patterns in complex spatio-temporal data finds usage across various domains in applied science and engineering, including training autonomous vehicles to navigate in complex traffic scenarios. Motivated by applications arising in the transportation domain, in this paper we develop a model for learning heterogeneous and dynamic patterns of velocity field data. We draw from basic nonparameric Bayesian modeling elements such as hierarchical Dirichlet process and infinite hidden Markov model, while the smoothness of each homogeneous velocity field element is captured with a Gaussian process prior. Of particular focus is a scalable approximate inference method for the proposed model; this is achieved by employing sequential MAP estimates from the infinite HMM model and an efficient sequential GP posterior computation technique, which is shown to work effectively on simulated data sets. Finally, we demonstrate the effectiveness of our techniques to the NGSIM dataset of complex multi-vehicle interactions.