Self-Exploration in Complex Unknown Environments using Hybrid Map Representation

TL;DR

This paper introduces a hybrid map representation combining topological and metric maps to improve exploration efficiency in complex environments, utilizing a hierarchical global-local strategy for autonomous robot exploration.

Contribution

It presents a novel hybrid map and a hierarchical exploration algorithm that enhances exploration efficiency in complex environments, verified through simulations and real-world tests.

Findings

Improved exploration efficiency in complex environments.

Effective hierarchical global-local exploration strategy.

Successful validation in simulated and real-world environments.

Abstract

A hybrid map representation, which consists of a modified generalized Voronoi Diagram (GVD)-based topological map and a grid-based metric map, is proposed to facilitate a new frontier-driven exploration strategy. Exploration frontiers are the regions on the boundary between open space and unexplored space. A mobile robot is able to construct its map by adding new space and moving to unvisited frontiers until the entire environment has been explored. The existing exploration methods suffer from low exploration efficiency in complex environments due to the lack of a systematical way to determine and assign optimal exploration command. Leveraging on the abstracted information from the GVD map (global) and the detected frontier in the local sliding window, a global-local exploration strategy is proposed to handle the exploration task in a hierarchical manner. The new exploration algorithm is able to create a modified tree structure to represent the environment while consolidating global frontier information during the self-exploration. The proposed method is verified in simulated environments, and then tested in real-world office environments as well.