Hybrid hierarchical motion planning
- View All
In this paper, the problem of motion planning of an autonomous agent in an uncertain environment is solved in a hierarchical fashion. At the higher level of the hierarchy, the state-space of the agent is reduced to a set of "landmarks" through the use of suitably defined control policies at the lower level, called "options", which results in a semi-Markov Decision problem (SMDP). Existing algorithms for path planning such as PRMs/RRTs are utlized as the lower level options. Two model-based approaches (based on the model of the uncertain map): one based on simulations and the other based on heuristics, are proposed to estimate the parameters of the higher level planner, i.e., the SMDP. The methodology is applied to an unmanned ground vehicle, with non-trivial dynamics, navigating a cluttered and uncertain urban environment. © 2008 by the American Institute of Aeronautics and Astronautics, Inc.
author list (cited authors)
Chakravorty, S., & Saha, R.