Navigation in three-dimensional cluttered environments for mobile manipulation

Abstract

Abstract-Collision-free navigation in cluttered environments is essential for any mobile manipulation system. Traditional navigation systems have relied on a 2D grid map projected from a 3D representation for efficiency. This approach, however, prevents navigation close to objects in situations where projected 3D configurations are in collision within the 2D grid map even if actually no collision occurs in the 3D environment. Accordingly, when using such a 2D representation for planning paths of a mobile manipulation robot, the number of planning problems which can be solved is limited and suboptimal robot paths may result. We present a fast, integrated approach to solve path planning in 3D using a combination of an efficient octree-based representation of the 3D world and an anytime search-based motion planner. Our approach utilizes a combination of multi-layered 2D and 3D representations to improve planning speed, allowing the generation of almost realtime plans with bounded sub-optimality. We present extensive experimental results with the two-armed mobile manipulation robot PR2 carrying large objects in a highly cluttered environment. Using our approach, the robot is able to efficiently plan and execute trajectories while transporting objects, thereby often moving through demanding, narrow passageways.

Real-time navigation in 3D environments based on depth camera data

“…As the results show, our approach leads to accurate localization estimates and reliable, collision-free navigation in the acquired 3D map. In the future, we will extend the approach to multi-level collision maps for different parts of the robot as in [18]. Hence, we will lift the circular robot model assumption made in this paper and allow the robot to better pass narrow passages.…”

“…For this process, we maintain a projected 2D map for efficient collision checks as in [18]. Each 3D map update of the local obstacle map also updates the 2D projection.…”

Postural modes and control for dexterous mobile manipulation: the UMass uBot concept

“…Mobile manipulation in unstructured environments is the focus of a great deal of current and ongoing research [2], [3], [4], [5], [6]. Among the most important technical challenges concerns the generation of contingencies for situations that are not necessarily anticipated beforehand and for dealing with generic constraints imposed by different tasks.…”

Vision-Based Grasp Planning and Experiments of a Mobile Manipulator

“…PR2 3 processes sensory data from 3D visual sensors in point clouds. Multilayered 2D costmaps and a layered representation of the robot body are used to reduce the possibility of collision 4 . In the work of Stulp et al…”