Probabilistic Trajectory Segmentation by Means of Hierarchical Dirichlet Process Switching Linear Dynamical Systems
This work addresses the need for automated trajectory segmentation in robotics to simplify movement library creation, though it appears incremental as it builds on existing switching models with a Bayesian approach.
The paper tackled the problem of automatically segmenting robot demonstration trajectories to build movement primitive libraries, by modeling trajectories with Switching Linear Dynamical Systems and using a nonparametric Bayesian Gibbs sampler, achieving meaningful segmentation without prior segmentation.
Using movement primitive libraries is an effective means to enable robots to solve more complex tasks. In order to build these movement libraries, current algorithms require a prior segmentation of the demonstration trajectories. A promising approach is to model the trajectory as being generated by a set of Switching Linear Dynamical Systems and inferring a meaningful segmentation by inspecting the transition points characterized by the switching dynamics. With respect to the learning, a nonparametric Bayesian approach is employed utilizing a Gibbs sampler.