Rapid multi-robot exploration with topometric maps

Abstract

Abstract-Multi-robot map building has advanced to the point where high quality occupancy grid data may be collected by multiple robots collaborating with only intermittent connectivity. However, the tasking of these agents to most efficiently build the map is a problem that has seen less attention. Unfamiliar, highly cluttered environments can confound exploration strategies that rely solely on occupancy grid frontier identification or even semantic classification methods keyed on geometric features. To reason about partial maps of novel, highly cluttered locations, hypotheses about significant structure in the disposition of free space may be used to guide exploration task assignment. A parsing of map data into places with semantic significance to the exploration task provides a foundation from which one may infer an efficient exploration strategy.

Decentralized waypoint-based multi-robot coordination

“…Voronoi-based multi-robot exploration received an increasing attention from the research community in the last decade [14] [15] [16]. We implemented this method for creating the original waypoints in the experiments.…”

Mapping with depth panoramas

“…Once this gravity vector has been recovered, the second phase seeks to estimate the dominant rectilinear structure of the scene. This can be done by employing the entropy compass idea described in [7]. To do this we consider all of the points in the image whose normals are approximately perpendicular to the gravity direction, that is the points that correspond to vertical surfaces.…”

Simultaneous Localization and Layout Model Selection in Manhattan Worlds

“…The first step in this process involves projecting the depth points into the plane defined by the measured gravity vector and then rotating the resulting 2D point set in one degree increments to find a yaw orientation that minimizes the entropy of the resulting point distribution. This algorithm is described in a number of previous works including [21], [22], and [23] where it is referred to as an entropy compass. Upon completion, this procedure recovers the orientation of the frame with respect to the prevailing Manhattan structure.…”