See it without touching it: low-cost non-contact sensing for our waterways

Overview

Many of our most important waterbodies, including reservoirs, lakes, lagoons, wetlands, sedimentation basins, and constructed wetlands, are still monitored using sparse in-water sensors and periodic grab sampling. These methods are costly to maintain, hard to scale across many sites, and often miss spatially variable changes in water quality.

Non-contact sensing offers a different approach. Cameras, spectral sensors, radar, thermal imaging, and other sensing modalities can observe water from outside it, reducing fouling, simplifying servicing, improving worker safety, and enabling broader spatial coverage. The harder question is not what we can measure, but what we should: which water-quality targets are reasonable to report directly, which can be inferred through validated surrogates, and which should instead trigger a follow-up sample.

This PhD will develop and validate a low-cost, modular, non-contact multimodal sensing platform built around answering exactly that question.

Research activities

Working as part of the IoT for Water Hub (an ARC Industrial Transformation Research Hub), you will:

• work with industry partners to lock in a three-tier monitoring framework covering directly observable, surrogate-inferable, and trigger-only targets
• design and build a modular sensing node that can carry different combinations of imaging, spectral, radar, and thermal modalities
• run synchronised field campaigns at three contrasting waterbodies, collecting matched non-contact and reference data
• develop confidence-aware multimodal analytics that fuse modalities and produce tier-specific outputs
• Produce modality-selection guidance that partners can use to choose what to deploy where.

You will be supported by a senior Research Fellow and a Research Assistant providing field deployment and modelling capacity.

Outcomes

By the end of the PhD you will have delivered a working multimodal sensing platform, a matched dataset across three contrasting waterbodies, and a clear modality-selection framework that converts the work into industry guidance. You will graduate with skills spanning computer vision, sensor hardware, multimodal data fusion, and field validation, with strong career paths into water utilities, remote sensing, environmental monitoring, and AI-enabled instrumentation.

Skills and experience

We are looking for a student with a Bachelor (Honours) or Masters degree in electrical engineering, mechatronic engineering, computer science, environmental engineering, or related disciplines. Experience or strong interest in imaging, computer vision, signal processing, or sensor hardware is highly desirable. Some familiarity with machine learning is an advantage. The student should be comfortable with field deployment.

Scholarships

You may be eligible to apply for a research scholarship.

Explore our research scholarships

Keywords

• sensing
• water quality
• IoT
• monitoring
• AI

Contact

Contact the supervisor for more information.