Vision-Based Goal-Conditioned Policies for Underwater Navigation in the Presence of Obstacles

TL;DR

This paper introduces Nav2Goal, a data-efficient, end-to-end goal-conditioned visual navigation method enabling underwater robots to autonomously reach waypoints, avoid obstacles, and collect scientific data in complex ocean environments.

Contribution

The paper presents Nav2Goal, a novel goal-conditioned visual navigation approach that operates without prior maps and is validated in real-world underwater deployments.

Findings

Successfully navigated over a kilometer in open ocean.

Reached approximately 40 waypoints autonomously.

Operated within 0.5 m of sensitive coral reefs.

Abstract

We present Nav2Goal, a data-efficient and end-to-end learning method for goal-conditioned visual navigation. Our technique is used to train a navigation policy that enables a robot to navigate close to sparse geographic waypoints provided by a user without any prior map, all while avoiding obstacles and choosing paths that cover user-informed regions of interest. Our approach is based on recent advances in conditional imitation learning. General-purpose, safe and informative actions are demonstrated by a human expert. The learned policy is subsequently extended to be goal-conditioned by training with hindsight relabelling, guided by the robot's relative localization system, which requires no additional manual annotation. We deployed our method on an underwater vehicle in the open ocean to collect scientifically relevant data of coral reefs, which allowed our robot to operate safely and autonomously, even at very close proximity to the coral. Our field deployments have demonstrated over a kilometer of autonomous visual navigation, where the robot reaches on the order of 40 waypoints, while collecting scientifically relevant data. This is done while travelling within 0.5 m altitude from sensitive corals and exhibiting significant learned agility to overcome turbulent ocean conditions and to actively avoid collisions.