Goal-Driven Autonomous Exploration Through Deep Reinforcement Learning

TL;DR

This paper introduces a goal-driven autonomous exploration system using deep reinforcement learning, enabling robots to navigate unknown environments efficiently without prior maps or information.

Contribution

The paper presents a novel DRL-based navigation system that integrates learned local policies with global waypoint planning for autonomous exploration.

Findings

Outperforms similar exploration methods in complex environments

Operates without prior maps or environment knowledge

Effective in both static and dynamic settings

Abstract

In this paper, we present an autonomous navigation system for goal-driven exploration of unknown environments through deep reinforcement learning (DRL). Points of interest (POI) for possible navigation directions are obtained from the environment and an optimal waypoint is selected, based on the available data. Following the waypoints, the robot is guided towards the global goal and the local optimum problem of reactive navigation is mitigated. Then, a motion policy for local navigation is learned through a DRL framework in a simulation. We develop a navigation system where this learned policy is integrated into a motion planning stack as the local navigation layer to move the robot between waypoints towards a global goal. The fully autonomous navigation is performed without any prior knowledge while a map is recorded as the robot moves through the environment. Experiments show that the proposed method has an advantage over similar exploration methods, without reliance on a map or prior information in complex static as well as dynamic environments.