Most pose (3D position and 3D orientation) tracking methods using vision require a priori knowledge about the environment and correspondences between 3D environment features and 2...
Recurrent neural networks serve as black-box models for nonlinear dynamical systems identification and time series prediction. Training of recurrent networks typically minimizes t...
Abstract. The connectivity of the con guration space has been a valuable concept in the motion planning for single robots in both known and unknown terrains. We show here that n-co...
Abstract. Switching linear dynamic systems (SLDS) attempt to describe a complex nonlinear dynamic system with a succession of linear models indexed by a switching variable. Unfortu...
This paper reports on an on-going project to investigate techniques to diagnose complex dynamical systems that are modeled as hybrid systems. In particular, we examine continuous s...
Sheila A. McIlraith, Gautam Biswas, Dan Clancy, Vi...