Towards Monocular Vision based Obstacle Avoidance through Deep Reinforcement Learning

TL;DR

This paper introduces a deep reinforcement learning approach using a dueling double-Q network for obstacle avoidance with monocular vision, enabling efficient learning and successful transfer from simulation to real robots in dynamic environments.

Contribution

It presents a novel D3QN architecture tailored for monocular vision obstacle avoidance, improving learning speed and transferability to real-world robots.

Findings

D3QN doubles learning speed compared to standard deep Q networks.

Models trained in simulation effectively transfer to real robots.

The approach generalizes well to unseen dynamic environments.

Abstract

Obstacle avoidance is a fundamental requirement for autonomous robots which operate in, and interact with, the real world. When perception is limited to monocular vision avoiding collision becomes significantly more challenging due to the lack of 3D information. Conventional path planners for obstacle avoidance require tuning a number of parameters and do not have the ability to directly benefit from large datasets and continuous use. In this paper, a dueling architecture based deep double-Q network (D3QN) is proposed for obstacle avoidance, using only monocular RGB vision. Based on the dueling and double-Q mechanisms, D3QN can efficiently learn how to avoid obstacles in a simulator even with very noisy depth information predicted from RGB image. Extensive experiments show that D3QN enables twofold acceleration on learning compared with a normal deep Q network and the models trained solely in virtual environments can be directly transferred to real robots, generalizing well to various new environments with previously unseen dynamic objects.