On June 18, 2023, the Titan submersible was approximately 90 minutes into its planned two-hour descent to the Titanic wreckage, located at the bottom of the Atlantic Ocean, when all communication with its support ship abruptly ceased. This loss of contact ignited a desperate search operation for the tourist submersible and its five passengers, resting around two miles beneath the surface.
Deep-ocean search and recovery missions are critical responsibilities for military organizations, including the U.S. Coast Guard’s Office of Search and Rescue and the U.S. Navy’s Supervisor of Salvage and Diving. In such operations, the bulk of time lost tends to come from shipping search-and-rescue equipment to the remote area of interest and thoroughly surveying that region. The scale of the Titan search — conducted 420 nautical miles from the nearest port, covering 13,000 square kilometers (approximately double the size of Connecticut) — could potentially extend over weeks. Despite being relatively small given the circumstances, focusing on the immediate area around the Titanic, a search in less-known waters could stretch months. Remarkably, a remotely operated underwater vehicle, deployed by a Canadian vessel, discovered the Titan debris field four days after the submersible went missing.
A dedicated team from MIT Lincoln Laboratory and the MIT Department of Mechanical Engineering‘s Ocean Science and Engineering lab is innovating a surface-based sonar system designed to drastically reduce the timeline for both small and large-scale search operations to just days. This groundbreaking system, known as the Autonomous Sparse-Aperture Multibeam Echo Sounder, scans at surface-ship speeds while delivering adequate resolution to identify submerged objects and underwater features without incurring the costs and logistical challenges of using underwater vehicles. The echo sounder utilizes a substantial sonar array coupled with a limited fleet of autonomous surface vehicles (ASVs) that can be deployed via aircraft, boasting the ability to map the ocean floor at 50 times the coverage rate of typical underwater vehicles and 100 times the resolution of conventional surface vessels.
“Our array delivers the best of both worlds: the high resolution of underwater vehicles combined with the rapid coverage capabilities of surface ships,” states co-principal investigator Andrew March, who is the assistant leader of the laboratory’s Advanced Undersea Systems and Technology Group. “Although large surface sonar systems operating at low frequency can analyze the materials and profiles of the seabed, they do so with diminished resolution, particularly at greater ocean depths. Our array is likely able to gather the same information, but with significantly enhanced clarity in the deep ocean.”
Unlocking the Underwater Unknown
The oceans compose 71 percent of our planet’s surface; nevertheless, over 80 percent of this underwater domain remains unexplored and uncharted. In fact, we possess more knowledge about the surfaces of other planets and the moon than we do about the depths of our oceans. High-resolution seabed maps are essential not only for locating lost entities like ships or aircraft but also for various scientific endeavors: grasping Earth’s geological structures, refining ocean current predictions, unveiling archaeological sites, monitoring marine habitats, and pinpointing areas rich in natural resources like minerals and oil.
Globally, scientists and governments acknowledge the urgency of generating a high-resolution, global map of the ocean floor. However, current technologies fall short; they struggle to achieve meter-scale resolution from the ocean surface. The average ocean depth is around 3,700 meters, and technologies capable of detecting human-made objects or person-sized natural features — including sonar, lidar, cameras, and gravitational field mapping — typically operate at ranges below 1,000 meters underwater.
To map the deep ocean, ships equipped with extensive sonar arrays release low-frequency sound waves that reflect off the seabed and return as echoes, but this method sacrifices image clarity. Low-frequency operations are necessary due to water’s absorption of high-frequency sounds, which diminishes resolution further with increasing depth; each pixel in the generated images represents an area the size of a football field. Large mapping vessels deploy sonar arrays confined to hull space, limiting aperture size and image fidelity. On the other hand, autonomous underwater vehicles (AUVs) functioning at higher frequencies can produce detailed images, where each pixel covers one square meter or less. However, the trade-offs are substantial: AUVs are labor-intensive and costly to deploy in deep waters, have a limited operational range of 1,000 meters before their high-frequency sound is absorbed, and proceed slowly to conserve energy, covering just eight square kilometers per hour while surface vessels can expedite mapping at over 50 times that rate.
A Groundbreaking Solution Emerges
The Autonomous Sparse-Aperture Multibeam Echo Sounder proposes a cost-efficient strategy for swift, high-resolution mapping of the deep ocean floor, launching from the water’s surface. An effective fleet of roughly 20 ASVs, each equipped with a compact sonar array, collectively forms an extensive sonar array 100 times larger than conventional systems on ships. This configuration, featuring a large aperture of hundreds of meters, produces a narrow sound beam that facilitates precise steering, enabling the creation of high-resolution maps at low frequencies. The sparse nature of the array, with fewer sonars relative to its size, keeps costs manageable.
That said, this novel arrangement comes with operational hurdles. To generate coherent 3D images, the relative positions of each ASV’s sonar subarray must be accurately tracked amid the dynamic motions of the ocean. Additionally, the array’s design introduces a lower signal-to-noise ratio due to gaps between sonar elements, making it more challenging to filter out unwanted noise. To address these challenges, the team is crafting a low-cost precision relative navigation system, alongside utilizing acoustic signal processing tools and innovative ocean-field estimation algorithms. Collaborators at MIT are focused on developing algorithms for data processing and imaging, aiming to effectively estimate depth-integrated water column parameters. These technologies will help account for the complex physics of the ocean, including temperature variations, dynamic currents and waves, and sound propagation factors like speed.
Control for all required navigation and calculations can occur either remotely or onboard the ASVs. For instance, ASVs launched from a ship or flying boat could be remotely operated from land using satellite links or from a nearby support ship, allowing them to gather seabed data over weeks or months without needing constant ship reinforcements for supplies and crew changes. Preliminary sonar health checks and rough seabed mapping can be performed onboard, while comprehensive, high-resolution seabed reconstructions may require a supercomputing system either on shore or aboard a support ship.
“Deploying these vehicles in a location and allowing them to map freely for extended periods would significantly streamline logistics and operational expenses,” emphasizes co-principal investigator Paul Ryu, a researcher in the Advanced Undersea Systems and Technology Group.
Since the inception of this research in 2018, the team has transformed their vision into a working prototype. Initially, they constructed a scaled-down model of a sparse-aperture sonar array and tested its capabilities in a water tank at the laboratory’s Autonomous Systems Development Facility. Subsequently, they developed an ASV-sized sonar subarray and showcased its effectiveness in Gloucester, Massachusetts. Recent field tests in Boston Harbor involved deploying an 8-meter array featuring multiple subarrays equivalent to 25 ASVs combined. This array successfully generated 3D images of the seafloor and even a shipwreck. Most recently, in collaboration with Woods Hole Oceanographic Institution, they created a first-generation, all-electric ASV prototype, 12 feet in length, housing a sonar array underneath. Preliminary relative navigation tests were conducted in Woods Hole, Massachusetts, and Newport, Rhode Island. The complete ocean mapping vision plans for approximately 20 similar-sized ASVs powered by sustainable energy sources such as wave or solar energy.
This ambitious project is backed by Lincoln Laboratory’s internally managed R&D portfolio focused on autonomous systems. The team is actively seeking external funding to further enhance their deep-ocean mapping technology, which has garnered the prestigious 2024 R&D 100 Award.
Photo credit & article inspired by: Massachusetts Institute of Technology
