Autonomous underwater vehicles (AUV) perform various applications such as seafloor mapping and underwater structure health monitoring. Commonly, an inertial navigation system aided by a Doppler velocity log (DVL) is used to provide the vehicle's navigation solution. In such fusion, the DVL provides the velocity vector of the AUV, which determines the navigation solution's accuracy and helps estimate the navigation states. This paper proposes BeamsNet, an end-to-end deep learning framework to regress the estimated DVL velocity vector that improves the accuracy of the velocity vector estimate, and could replace the model-based approach. Two versions of BeamsNet, differing in their input to the network, are suggested. The first uses the current DVL beam measurements and inertial sensors data, while the other utilizes only DVL data, taking the current and past DVL measurements for the regression process. Both simulation and sea experiments were made to validate the proposed learning approach relative to the model-based approach. Sea experiments were made with the Snapir AUV in the Mediterranean Sea, collecting approximately four hours of DVL and inertial sensor data. Our results show that the proposed approach achieved an improvement of more than 60% in estimating the DVL velocity vector.

The Chinese navy, under instruction from President Xi Jinping, has undergone a modernisation and expansion programme that is nothing short of spectacular. Friday's launch of its third and most advanced aircraft carrier, the Fujian, for sea trials underscores just how far it has come, and how fast. The first two carriers, the Liaoning and Shandong, were ex-Soviet designs; the Liaoning initially bought for scrap from Ukraine and refitted. While antiquated, they have been used to train new generations of naval officers and pilots in the complex science and art of aircraft carrier operations. This new design of aircraft carrier is a quantum leap in capabilities from these older models and will greatly enhance China's combat power.

There's nothing more romantic than a hunt for hidden treasure--and when those riches are located in the watery depths of the ocean, it can seem even more exciting. Shipwrecks spark the imagination, prompting dreams of untold riches and swashbuckling adventure. More vessels lie at the bottom of the sea than you might think; the National Oceanic and Atmospheric Administration's database lists over 10,000 known wrecks off of United States shores alone--and that's not a complete list. According to United Nations cultural agency UNESCO, there are at least 3 million such wrecks worldwide, some thousands of years old. And then there's the booty some of those ships carried. Though there's an argument to be made that the treasure aboard now-sunken vessels is priceless, some experts estimate as much as $60 billion in precious metals lies on the ocean floor.

The 50ft (15m) long solar-powered trimaran is capable of speeds of up to 10 knots (20km/h) and was navigated by on-board artificial intelligence (AI) created by IBM with information from six cameras and 50 sensors.

A stunning drone video has emerged showing a Ukrainian warship narrowly escaping a massive Russian artillery fire, some of which lands as close as 200 feet. The footage, allegedly captured by a shooting spotter drone, shows the Ukrainian vessel Yuri Olefirenko, a Polnochny-class landing ship, coming under Russian attack as it sails along the Bugsko-Dneprovsko-Limansky Canal near the port of Ochakov in Mykolaiv region. According to defense analysts, the incident happened on June 3. The warship appears to be heading to Odessa when invaders rain down missiles on it. The artillery attack covers almost the entire area around the ship, some weapons falling dangerously close to the vessel.

A whopping 350 years after it sank off the coast of Norfolk, authorities have revealed on Friday that HMS Gloucester has finally been found. The'outstanding' ship, which sank on May 6, 1682 after hitting the Norfolk sandbanks in the southern North Sea, was uncovered 28 miles off the coast of Great Yarmouth half-buried on the seabed. But HMS Gloucester is just one of thousands of shipwrecks that litter the British coast, the majority of which haven't been seen by the human eye for centuries. It's thought nearly 40,000 wrecks could be waiting to be found off the British coast, according to Historic England, providing snapshots of the UK's rich maritime heritage. But at least 90 are known to exist and experts have pinpointed their location, although many likely won't ever be brought to land and could disintegrate to nothing in the decades to come.

Prism Courage, a 134,000-tonne commercial tanker, recently sailed from the Gulf of Mexico to South Korea while controlled mostly by an artificial intelligence system called HiNAS 2.0. Avikus, a subsidiary of South Korean technology giant Hyundai, recently announced that Prism Courage, a tanker designed to transport natural gas, had become the first large ship to make an ocean passage of over 10,000 km (6,210 miles) autonomously. The key to this incredible achievement was HiNAS 2.0, an AI-powered system capable of analyzing different kinds of sensor readings in real-time and responding to them swiftly, efficiently, and, most importantly, in accordance with the rules of maritime laws. Just like airplanes, ships have very advanced auto-pilots capable of keeping them on a steady course, responding to GPS waypoints and currents, and even bringing them into harbor in case the human crew is no longer present on board or capable of doing it. However, sailing autonomously for tens of thousands of kilometers through the Atlantic is a lot more complex than putting a ship on autopilot. Apart from steering the tanker in real0-time, Avikus' HiNAS 2.0 system is capable of picking the optimal routes and best speeds to reach its destination, by analyzing data collected through advanced sensors.

In a voyage lasting 40 days and conquering approximately 3 500 unmanned miles at sea, the Mayflower Autonomous Ship arrived in North America in Halifax, Nova Scotia on June 5, 2022. Following two years of design, construction and AI model training, the Mayflower Autonomous Ship (MAS) was officially launched in September 2020. Fast forward to 5 June 2022, and the ship completed an historic transatlantic voyage from Plymouth, UK to its North American arrival in Halifax, Nova Scotia. With no human captain or onboard crew, MAS is the first self-directed autonomous ship with technology that is scalable and extendible to traverse the Atlantic Ocean. MAS was designed and built by marine research non-profit ProMare with IBM acting as lead technology and science partner, with IBM automation, AI and edge computing technologies powering the ship's artificial intelligence (AI) captain to guide the vessel and make real-time decisions while at sea.

Physical motion models offer interpretable predictions for the motion of vehicles. However, some model parameters, such as those related to aero- and hydrodynamics, are expensive to measure and are often only roughly approximated reducing prediction accuracy. Recurrent neural networks achieve high prediction accuracy at low cost, as they can use cheap measurements collected during routine operation of the vehicle, but their results are hard to interpret. To precisely predict vehicle states without expensive measurements of physical parameters, we propose a hybrid approach combining deep learning and physical motion models including a novel two-phase training procedure. We achieve interpretability by restricting the output range of the deep neural network as part of the hybrid model, which limits the uncertainty introduced by the neural network to a known quantity. We have evaluated our approach for the use case of ship and quadcopter motion. The results show that our hybrid model can improve model interpretability with no decrease in accuracy compared to existing deep learning approaches.

Recent work suggests that convolutional neural networks of different architectures learn to classify images in the same order. To understand this phenomenon, we revisit the over-parametrized deep linear network model. Our analysis reveals that, when the hidden layers are wide enough, the convergence rate of this model's parameters is exponentially faster along the directions of the larger principal components of the data, at a rate governed by the corresponding singular values. We term this convergence pattern the Principal Components bias (PC-bias). Empirically, we show how the PC-bias streamlines the order of learning of both linear and non-linear networks, more prominently at earlier stages of learning. We then compare our results to the simplicity bias, showing that both biases can be seen independently, and affect the order of learning in different ways. Finally, we discuss how the PC-bias may explain some benefits of early stopping and its connection to PCA, and why deep networks converge more slowly with random labels.