Project plan

Arranged in mesh networks of non-synchronized nodes, Internet of Underwater Things (IoUT) devices can serve as efficient agents for detecting the marine environment. As low-cost devices, IoUTs are designed to scan their marine environment both above and below water and collaborate to make joint decisions. 

One specific IoUT application is search and surveillance, where IoUT floaters use acoustic detection to scan the water column to detect ships or submarines. The IoUT detection swarm works either autonomously or cooperatively via an acoustic underwater modem and reports back via satellite communication.

Acoustic detection of ships by low-cost IoUT faces the challenge that acoustic sensors have low sensitivity and computing hardware is limited by power and memory. Shipping URN includes high power impulsive transient waves generated by the ship during ignition and cavitation noise, which are relatively stationary, continuous wave signals caused by the collapse of vapor bubbles near the propeller. 

Methods for distinguishing between ship and ambient noise generally assume that the ambient noise is diffuse, while its anthropogenic component is directional and can therefore be extracted by array processing. However, in the context of energy-limited and small IoUT devices, detection should be performed with low complexity and using only a few hydrophones. The same limitation applies to the control system of the swarm, which must use reliable but compressed communication to reduce energy consumption, and to the operation of the floaters, which can rely not on thrusters but on the change of buoyancy for low energy consumption. 

Floating devices for monitoring the oceans are currently in use worldwide. The best-known devices are the Argo floaters, energy-efficient systems that continuously profile the water column. However, the loud noise generated by the pump during depth maintenance does not allow continuous acoustic monitoring.

A key factor in the efficient operation of the IoUT swarm is its ability to operate for long periods, preferably months. As energy is critical to IoUT operation, it is believed that harvesting renewable energy from the marine environment can enable the long-term operation of these devices. As noted in the Global Marine Technology Trends 2030 Report, with the extremely limited ability to recharge deployed devices, efficient energy harvesting from the ocean has become the holy grail of underwater sensory development. While IoUT may emerge, reliance on solar panels as an energy source is subject to biofouling, and the same is true for any moving device such as a wave or current turbine. Instead, we see great potential in the energy extracted from ambient acoustic noise.

In this project, we aim to design underwater floats that rely on a sophisticated control mechanism to efficiently change their buoyancy as they traverse the water column. The floaters will be equipped with 4 hydrophones and use a complex neural network that relies on the stability of ship noise to enable reliable low-complexity detection. 

To operate together in a swarm, the floaters will communicate to reach a consensus using two main strategies:

1. spreading out in the water column when no detection cue is found

2. concentrating at a specific depth. 

In both cases, the acoustic properties of the water will be taken into account by using the temperature measurements of the floaters, as well as the question where each drifter should be deployed to maximize coverage. 

To enable operation for months at a time, the floaters will incorporate a renewable circuit to harvest energy from the ambient noise. This is done via polymer-based transducers, with a good acoustic impedance matching for efficient energy conversion into the electrical domain, together with an associated electronic module for energy storage and delivery. 

The results of our project will be both theoretical and practical. We will develop energy harvesting circuits, detection algorithms and swarming strategies and implement them on embedded floats. Several field trials and demonstrations will lead to a TRL6 prototype of a 5-floater swarm. Our research question in this project aligns with the following three NATO SPS key priorities:

• 1a.i (Counter-Terrorism): Methods for the protection of critical infrastructure, supplies and personnel. In particular, addressing terrorist attacks on critical offshore infrastructure such as underwater cables (data and power) and semi-submersible explosive submarines to be used against offshore gas platforms.

• 1a.iii (Counter-Terrorism): Detection technologies against the terrorist threat from explosive devices and other illicit activities. Here we will develop a remote sensing platform to detect vehicle threats aimed at damaging offshore installations through the use of explosives.

• 1b.i (Energy Security): Innovative energy solutions for the military; battlefield energy solutions; renewable energy solutions with military applications. In particular, the use of ambient noise as a renewable energy source can revolutionize underwater sensor technology through small IoUT devices.

Our team has a unique set of complementary areas of expertise, including the development of polymer-based energy harvesting devices (co-director Lampe and co-investigator Cretu), acoustic signal processing and in-situ measurements (PPD Diamant), the design and implementation of control algorithms for submerged floats (co-director Mišković), and the design of flexible hardware modems (co-director Neasham) and underwater networks (NPD Casari). 

Workpackages