Adaptive Landmark Color for AUV Docking in Visually Dynamic Environments

TL;DR

The paper addresses DS detection for AUV docking in visually dynamic underwater environments where ambient water color changes rapidly. It introduces a vision-based method that dynamically adapts landmark color using two HSV-based mappings (pure complementary and pure ternary) by estimating the water background color from both the AUV and DS cameras, without inter-vehicle communication. Through pool and lake experiments, the ternary mapping demonstrates superior noise rejection and landmark detection, achieving robust detection at practical ranges and outperforming static color-thresholding methods by roughly an order of magnitude. This approach reduces dependence on expensive acoustic systems and enables reliable, on-board vision-driven docking to extend AUV mission durations.

Abstract

Autonomous Underwater Vehicles (AUVs) conduct missions underwater without the need for human intervention. A docking station (DS) can extend mission times of an AUV by providing a location for the AUV to recharge its batteries and receive updated mission information. Various methods for locating and tracking a DS exist, but most rely on expensive acoustic sensors, or are vision-based, which is significantly affected by water quality. In this \doctype, we present a vision-based method that utilizes adaptive color LED markers and dynamic color filtering to maximize landmark visibility in varying water conditions. Both AUV and DS utilize cameras to determine the water background color in order to calculate the desired marker color. No communication between AUV and DS is needed to determine marker color. Experiments conducted in a pool and lake show our method performs 10 times better than static color thresholding methods as background color varies. DS detection is possible at a range of 5 meters in clear water with minimal false positives.

Figures (9)

• Figure 1: The LoCO AUV (bottom) using adaptive color RGB markers to locate a docking station (top).
• Figure 2: Left - Prototype DS without buoys and detached camera module. Middle - The LoCO AUV successfully docked with the DS in a water flume. Right - A composition of two images taken five seconds apart. A cloud passes overhead, drastically changing water color in Lake Superior.
• Figure 3: The hexcone model of the HSV color space. In this model, chroma is the radial dimension when value is less than one, not saturation, but we are mainly considering colors whose value is one.
• Figure 4: A color $C$, its complement $\widehat{C}$, and pure complement $\widehat{C_{p}}$ in HSV color space.
• Figure 5: A color $C$, its ternary $\acute{C}$, and pure ternary $\acute{C_{p}}$ in HSV color space.