“…The set of states can be seen as a set of labels for the regions in the partition, and each transition corresponds to a controller driving the vehicle between two adjacent regions. By partitioning the environment into simplicial or rectangular regions, continuous feedback controllers that drive a robotic system from any point inside a region to a desired facet of an adjacent region have been developed for linear ( Kloetzer & Belta, 2008a), multi-affine (Habets, Collins, & van Schuppen, 2006), piecewise-affine (Desai, Ostrowski, & Kumar, 1998;Habets, Kloetzer, & Belta, 2006;Wongpiromsarn, Topcu, & Murray, 2009), and non-holonomic (unicycle) (Belta, Isler, & Pappas, 2005;Lindemann, Hussein, & LaValle, 2007) dynamical models. By relating the initial continuous dynamical system and the abstract discrete finite system through simulation or bisimulation relations (Milner, 1989), the abstraction process allows one to replace the original more complex continuous system with the ''equivalent'' abstract system when solving a control synthesis problem.…”