Directed Exploration Using a Modified Distance Transform

Abstract

Mobile robots operating in unknown environments need to build maps. To do so they must have an exploration algorithm to plan a path. This algorithm should guarantee that the whole of the environment, or at least some designated area, will be mapped. The path should also be optimal in some sense and not simply a "random walk" which is clearly inefficient. When multiple robots are involved, the algorithm also needs to take advantage of the fact that the robots can share the task. In this paper we discuss a modification to the well-known distance transform that satisfies these requirements.

Simplified Occupancy Grid Indoor Mapping Optimized for Low-Cost Robots

“…In [25], the Next-Best-View strategy is used: the best candidate for a new search is the position that will allow for the gathering of more information. The Distance Transform is used in [26]; in this algorithm, the best candidate is the nearest unknown position. Finally, the algorithm in [27] considers for every candidate the possibility of gathering information and the distance to the current position.…”

“…The proposed algorithm is also one of the slowest solutions compared to [25][26][27]. However, it is important to notice that the mapping process is accomplished only once, so the mapping time should not be critical.…”

A gis-based simulation model for military path planning of unmanned ground robots

“…It is generally defi ned as the search for paths (from certain points to destinations) that satisfy specifi ed objectives under various conditions. To date, a variety of approaches have been developed to solve the problem, such as genetic algorithms [2,3], neural networks [4], distance transformation (DT) [5][6][7][8], and analytical methods [9].…”

“…As such, a weighted travel cost function which comprehensively evaluates friendly UGR's energy consumption (with consideration of terrain slope, surface distance, and friction) and degree of exposure to the enemies is developed here to fi nd the best path. A mathematical formulation that minimizes total travel cost of the UGR under specifi ed constraints can be expressed as eqn (7).…”

“…DT is a simple algorithm that determines a transformed distance between a particular point and target point. It is very popular for solution of path planning with both initially known and unknown fi eld topology, and has been widely used for robot path planning in various environments [5][6][7][8]. In DT, the path is generated by following the steepest gradient of the distance transform values from start to target.…”