A Bayesian Nonparametric Approach to Modeling Motion Patterns
Author(s)Joseph, Joshua Mason; Doshi-Velez, Finale P.; Huang, Albert S.; Roy, Nicholas
MetadataShow full item record
The most difficult—and often most essential— aspect of many interception and tracking tasks is constructing motion models of the targets to be found. Experts can often provide only partial information, and fitting parameters for complex motion patterns can require large amounts of training data. Specifying how to parameterize complex motion patterns is in itself a difficult task. In contrast, nonparametric models are very flexible and generalize well with relatively little training data. We propose modeling target motion patterns as a mixture of Gaussian processes (GP) with a Dirichlet process (DP) prior over mixture weights. The GP provides a flexible representation for each individual motion pattern, while the DP assigns observed trajectories to particular motion patterns. Both automatically adjust the complexity of the motion model based on the available data. Our approach outperforms several parametric models on a helicopter-based car-tracking task on data collected from the greater Boston area.
DepartmentMassachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory; Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Kluwer Academic Publishers
Joseph, Joshua, et al. "A Bayesian Nonparametric Approach to Modeling Motion Patterns" Autonomous Robots, 2011, 31.4 p.383–400. (From the issue entitled "Special Issue: Search and Pursuit-evasion with Mobile Robots")