New AI helps underwater robots get their bearings
To overcome the problem of overconfidence in AI-piloted autonomous underwater vehicles (AUVs), a new AI navigation system has been developed that enables an AUV to determine its coordinates using visual clues and evaluate the accuracy of those coordinates probabilistically. This can lead to a higher level of reliability for tasks like environmental monitoring or infrastructure inspection.
The ocean is such a complex environment – fluctuating temperature, current and visibility due to shifting sunlight, bubbles and silt – that, regardless of the number or sophistication of sensors added, an AI-piloted AUV can lose its bearings and veer off boldly in the wrong direction. GPS can’t help, since water blocks its signals. Sound waves travel rapidly through water, yet they too have their limits, requiring a network of transponders that compromise an AUV’s autonomy.
3D Gaussian Splatting
To address these significant challenges, Mengxue Hou, assistant professor of electrical engineering at the University of Notre Dame, and her lab have developed an artificial intelligence-based navigation system that enables an AUV to determine its coordinates using visual clues and evaluate the accuracy of those coordinates probabilistically. To achieve this, Hou’s team pairs a 3D scene-rendering technique known as 3D Gaussian Splatting (3DGS) with a Bayesian framework, a mathematical approach that allows the system to quantify uncertainty.
Before the AUV is launched on its mission, its ‘brain’ is loaded with a 3DGS map with underwater shapes – e.g. the pillars of a pier, a coral reef, a rocky outcrop – represented as ‘fuzzy’ ellipsoids (Gaussians). Since 3DGS renders objects as clouds of data rather than as rigid, visual points, it more accurately reflects the inherent murkiness of underwater environments.
Avoiding navigation errors
When the team tested an AUV in the HoloOcean simulator, they demonstrated that in chaotic environments – full of view-obscuring piers and poles – or ones that simply failed to provide sufficient data, the vehicle dialled back its confidence. Using a probabilistic Bayesian model, the vehicle recognized when it was appropriate to pause or revisit an area rather than make a navigation error.
“We essentially built a system that allows an underwater robot to say, ‘I think I’m here, but I’m only 60% sure, so I should probably go back and double-check,’” said Hou, who – together with first author and doctoral student Yu Zhou – published the results in OCEANS. “This system provides a massive leap in confidence for navigating complex structures like piers and reefs.”
Field experiments
The team is now trying to take the technology out of the simulator and into the field. Hou’s lab deployed a miniature underwater vehicle, equipped with an onboard oxygen sensor, for field experiments at the University of Notre Dame Environmental Research Center (UNDERC) facility in Wisconsin, USA, to carry out an environmental monitoring mission.
“Our framework ensures a higher level of reliability for tasks like environmental monitoring or infrastructure inspection,” said Hou. Karla Cruise of Notre Dame Engineering adds: “We’re moving beyond pre-programmed routes. By teaching robots to evaluate uncertainty in their knowledge of the surroundings, we are enabling active perception – where the vehicle realizes it lacks data and actively repositions itself to get a clearer picture of its surroundings.”
