KAUST Grant NumberKUS-C1-016-04
MetadataShow full item record
AbstractSampling-based solutions to the motion planning problem, such as the probabilistic roadmap method (PRM), have become commonplace in robotics applications. These solutions are the norm as the dimensionality of the planning space grows, i.e., d > 5. An important primitive of these methods is the local planner, which is used for validation of simple paths between two configurations. The most common is the straight-line local planner which interpolates along the straight line between the two configurations. In this paper, we introduce a new local planner, Toggle Local Planner (Toggle LP), which extends local planning to a two-dimensional subspace of the overall planning space. If no path exists between the two configurations in the subspace, then Toggle LP is guaranteed to correctly return false. Intuitively, more connections could be found by Toggle LP than by the straight-line planner, resulting in better connected roadmaps. As shown in our results, this is the case, and additionally, the extra cost, in terms of time or storage, for Toggle LP is minimal. Additionally, our experimental analysis of the planner shows the benefit for a wide array of robots, with DOF as high as 70. © 2012 IEEE.
CitationDenny J, Amato NM (2012) The Toggle Local Planner for sampling-based motion planning. 2012 IEEE International Conference on Robotics and Automation. Available: http://dx.doi.org/10.1109/ICRA.2012.6225212.
SponsorsThis research supported in part by NSF awards This research supportedin part by NSF Grants CRI-0551685, CCF-0833199, CCF-0830753, IIS-096053, IIS-0917266 by THECB NHARP award 000512-0097-2009, byChevron, IBM, Intel, Oracle/Sun and by Award KUS-C1-016-04, made byKing Abdullah University of Science and Technology (KAUST).