Efficiency of dynamic local area strategy for frontier-based exploration in indoor environments

Abstract

Exploration in unknown environments is a fundamental problem for autonomous robotic systems. The most of the existing exploration algorithms are proposed for indoor environments and aim to minimize the overall exploration time and total travelled distance. In this paper, a modified version of frontier-based exploration approach is presented to decrease exploration time and total distance. This approach introduces two more parameters to the Exploration Transform (ET) algorithm, which evaluates all detected frontiers to select next target point. On the other hand, the proposed approach considers locally observed frontiers, which might be changed with the parameter of dynamic distance. The proposed algorithm is individually tested and compared to ET algorithm with five random starting points in three different environments. Experimental results show that the proposed algorithm provides superior performance over conventional ET algorithm.