Seafloor Robotic Investigations: Advancing Technologies for Deep Ocean Exploration

The ocean floor remains one of Earth's least explored frontiers, with less than 10% of it adequately mapped despite covering over 70% of our planet's surface14. In contrast, we have complete maps of the Moon and Mars13. This disparity highlights the technical challenges of deep-sea exploration and the critical importance of robotic systems designed specifically for seafloor investigations. These technological marvels are transforming our ability to study, monitor, and interact with the ocean depths, revealing geological structures, biological communities, and environmental processes previously hidden from scientific inquiry. Recent advancements in artificial intelligence, materials science, and autonomous navigation have dramatically enhanced the capabilities of seafloor robots, enabling unprecedented access to and understanding of the marine environment.

Underwater robotic systems have evolved significantly since their early applications primarily in defense, where they served as remotely controlled missiles or recovery vehicles for sunken artillery13. Today's seafloor robots generally fall into two primary categories: Remotely Operated Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). ROVs are tethered to surface vessels via cables that provide power and enable real-time communication with human operators, while AUVs operate independently using pre-programmed instructions without physical connections to surface vessels139.

The remotely operated vehicle segment currently dominates the market, estimated to hold 52% of the global underwater robotics market by 20379. This dominance stems largely from the deep-sea oil drilling industry's growing reliance on these vehicles for tasks previously performed by human divers, such as surveying, inspection, and sampling9. The global underwater robotics market, valued at approximately $4.11 billion in 2025, is projected to expand at a compound annual growth rate of 16.3% to reach $25.92 billion by 2037, indicating substantial growth potential in this technological sector9.

Beyond the traditional ROV/AUV dichotomy, specialized seafloor robots have emerged to address specific operational needs. Seafloor crawlers represent an innovative approach that reverses the evolutionary pattern from sea to land observed in natural history. While tracked robots initially developed for land-based military applications, technological advances in energy storage, autonomy, and telemetry have enabled their adaptation for underwater use12. Companies like C-2 Innovations have developed amphibious bottom crawlers capable of operating at depths up to 100 meters through high water current regimes and transitioning onto land, offering unique capabilities for nearshore and surf zone operations12.

Hybrid systems that combine elements of different robot types represent another important development. The Mesobot, for instance, addresses specific unmet needs for observing and sampling phenomena in the ocean's midwaters, with capabilities for both autonomous operation and teleoperation through a lightweight fiber optic tether7. This versatility makes it particularly valuable for behavioral studies of marine organisms, allowing unobtrusive observation over extended periods with substantial autonomy7.

The extreme conditions of the deep ocean present formidable challenges for robotic systems. At depths of 6,000 meters (3.7 miles), water pressure reaches 596 atmospheres—equivalent to the weight of an elephant concentrated on a quarter6. These conditions necessitate specialized technologies and design approaches for effective operation.

Imaging and mapping capabilities represent a cornerstone of seafloor robotics. Synthetic Aperture Sonar (SAS) technology, such as Kraken Robotics' MINSAS system, provides high-coverage, high-resolution seabed imagery and 3D bathymetry rated for depths up to 6,000 meters17. Recent innovations in visual seafloor mapping employ navigation-aided hierarchical reconstruction, combining the benefits of simultaneous localization and mapping with global structure from motion techniques19. This approach addresses challenges specific to underwater visual mapping while optimizing for the limited dive times available to AUVs19.

Sampling technologies have also advanced significantly, with robots now capable of autonomously collecting seafloor samples. The Woods Hole Oceanographic Institution's Nereid Under Ice (NUI) vehicle achieved a milestone in 2020 by taking the first known automated sample from the ocean floor—specifically from the mineral-rich floor of Kolumbo volcano off Santorini Island, Greece10. This achievement marked a significant step toward eliminating direct human control in underwater sampling operations, with researcher Rich Camilli noting: "For a vehicle to take a sample without a pilot driving it was a huge step forward"10.

Pressure-tolerant systems represent another critical technological domain. Most electronics such as cameras, lights, and computers must operate in one atmosphere of air, necessitating enclosure in housings that can withstand extreme external pressures6. Engineers design these components using finite element analysis to simulate material stresses under pressure, with extensive laboratory testing before deployment6. SeaPower™ pressure-neutral battery technology exemplifies this approach, offering double the AUV endurance with more than twice the power per volume at less weight, rated for depths up to 6,000 meters17.

Soft robotics represents an emerging technological frontier particularly well-suited for handling delicate marine objects. Unlike rigid metallic robots that might damage fragile items during manipulation, soft robotic systems conform to an object's shape with light contact forces while retaining the strength needed for secure gripping15. Researchers at the University of Rhode Island have developed programmable, hybrid manipulators that can actively switch between soft/compliant and hard/rigid states, coupled with electric-drive, low-pressure hydraulic systems compatible with various platforms including inspection-class ROVs15.

Seafloor robotic systems serve diverse applications spanning scientific research, environmental monitoring, commercial operations, and defense applications. In the scientific realm, robots enable detailed study of the ocean floor that would otherwise remain inaccessible due to depth, pressure, or environmental hazards.

The Experimental Sedimentology and Seafloor Robot Laboratory at Kristineberg Center exemplifies the scientific application of these technologies. This 150-square-meter facility equipped with advanced instrumentation and a state-of-the-art seawater flume facilitates cutting-edge studies in marine particle dynamics1. Researchers utilize laboratory experiments and seafloor robotics to investigate sediment processes such as erosion, deposition, and transport, with particular emphasis on the ocean carbon cycle and seabed particulate flows1. This integration of technology and science establishes the lab as a crucial site for oceanic research, advancing our understanding of global change and its societal impacts1.

Environmental monitoring represents another significant application domain. Underwater robots equipped with sensors can detect pollution in real-time, measuring water quality parameters such as temperature, conductivity, and toxicity, while collecting samples for laboratory analysis9. Some robots incorporate specialized technology to identify plastic, rubber, and metal rubbish using artificial intelligence image recognition, helping focus limited clean-up budgets on the most toxic and environmentally damaging materials8.

Commercial applications, particularly in the offshore oil and gas sector, drive substantial demand for underwater robotics. These systems enable safer and more effective operations by inspecting and maintaining offshore pipelines and rigs9. Engineers can examine footage captured by underwater vehicles to identify signs of deterioration, corrosion, or structural problems, reducing the risk of serious accidents or equipment malfunction9. Additionally, underwater robots can perform routine maintenance tasks like cleaning and painting that might otherwise pose hazards to human workers9.

Archaeological investigations have also benefited tremendously from advances in seafloor robotics. The DRASSM (Department of Underwater Archaeological Research) in France has pioneered the application of robotics for deep-water archaeology through its CORSAIRE program, an innovative initiative to develop underwater robotics specifically for salvaging deep-water wrecks11. This work culminated in the development of Ocean One, a humanoid robot capable of diving to 2,000 meters depth and working as an intermediary for the eyes and hands of archaeologists at the surface11. The haptic feedback system of robots like OceanOneK produces realistic sensations that approximate what researchers would experience if physically present at depth, creating the impression that the operator is actually touching historical artifacts on the seafloor5.

Despite significant technological advances, seafloor robotic investigations continue to face substantial challenges imposed by the underwater environment. Communication limitations represent a particularly significant barrier, as underwater communication must rely on sound waves, free space optical fiber, or electromagnetic waves, each with significant drawbacks9. Sound waves are affected by temperature gradients and ambient noise, electromagnetic waves operate poorly in salt water, and optical signals travel extremely limited distances underwater9. Additionally, sophisticated signal processing and underwater communication require substantial power, complicating the implementation of advanced robotic systems9.

Navigation presents another major challenge, as underwater robots must often operate autonomously in complex, unstructured environments without human intervention9. This requires sophisticated sensing and decision-making capabilities, particularly when operating near delicate marine ecosystems or valuable archaeological sites. The underwater robot developed by a team from Norway addressed this challenge with stereo cameras and a spectrometer capable of identifying materials even through meters of murky water, allowing it to autonomously scan large areas of seabed for pollution8.

Physical challenges include the extreme pressure at depth, which increases by approximately one atmosphere for every 10 meters of depth6. At abyssal depths, this pressure can crush inadequately designed equipment, necessitating specialized pressure-resistant housings and components. Low temperatures, typically around 4°C (39°F) below 200 meters depth, create additional design challenges, as dissimilar materials change size at different rates as temperature falls, potentially causing mechanical failures6. The near-total darkness below 1,000 meters requires powerful lighting systems for effective imaging, which must be carefully positioned to avoid backscatter that would reduce visibility6.

Regulatory compliance adds another layer of complexity, as operations in marine environments must navigate a complex landscape of maritime and environmental regulations9. These may include restrictions on activities in protected areas, requirements for environmental impact assessments, and compliance with international maritime law, all of which can affect the design, deployment, and operation of seafloor robotic systems.

Research into seafloor robotics continues to accelerate, with numerous projects pushing the boundaries of what these systems can achieve. The Seafloor Intelligent Robot Exploration and Classification (SIREC) project, for example, seeks to improve the autonomy and cognitive capabilities of underwater robots during seafloor exploration3. This initiative focuses on two groundbreaking developments: intelligent mapping of large seafloor areas to optimize survey times, and intelligent mapping of complex three-dimensional structures previously limited to human pilots3. The project aims to enable robots to identify targets of interest autonomously and adapt their surveying approach accordingly, with applications in mapping Posidonia oceanica seagrass for carbon sequestration studies and surveying complex geological features like submarine rock outcrops and hydrothermal vent chimneys3.

The RESTORE project (Robotic-based invESTigation and mOnitoring Ross sEa) exemplifies another cutting-edge research direction, developing portable robotic technologies for 3D multi-parameter monitoring of marine ecosystems in Antarctica's Ross Sea Marine Protected Area16. This initiative focuses on interface zones including air-sea-ice, air-sea, sea-ice-sea-water-bottom, and sediment-water boundaries, supporting research on Antarctic life, geology, and climate change effects16. The modular, easily transportable system can be assembled in situ to form various configurations, including ROVs deployed through pack ice holes and unmanned semi-submersible vehicles for surface and sub-surface sampling16.

Educational initiatives are also emerging to develop the next generation of marine robotics expertise. The SEA Expedition: Marine Robotics program offers a two-week STEM curriculum for high school students interested in underwater robotics18. Participants work alongside experts from the Sea Education Association, NASA's Jet Propulsion Laboratory, and Woods Hole Oceanographic Institution to gain hands-on experience with ROV deployment while learning about ocean engineering's role in marine exploration and conservation18. Similarly, the 2025 Marine Robotics Summer School at Faial Island in the Azores will provide university students with lectures, workshops, hands-on experience, and time on the water with MIT and Portuguese faculty and other experts in the field20.

The market for underwater robotics shows strong growth prospects, driven by increasing demand for defense applications, technological advancements, and offshore infrastructure maintenance requirements9. The rising need for oceanography research and exploration, alongside the potential for deep-sea mining in response to growing mineral demands, suggests continued expansion of seafloor robotic investigations9. China's allocation of 161,211.2 hectares for deep-sea mining exploration (across 263 separate licenses) since July 2019 illustrates the growing commercial interest in seabed resources9.

Conclusion

Seafloor robotic investigations stand at the frontier of ocean exploration, offering unprecedented capabilities to study, monitor, and interact with the deep marine environment. From the pioneering remotely operated vehicles that first enabled human observation of the abyss to today's sophisticated autonomous systems capable of independent sampling and analysis, these technologies continue to evolve rapidly in response to scientific needs and commercial opportunities. Despite the substantial challenges posed by the underwater environment—extreme pressure, darkness, cold, and communication limitations—innovative engineering approaches have yielded increasingly capable systems that expand our understanding of the ocean floor.

As global interest in ocean resources, environmental monitoring, and scientific exploration grows, seafloor robotics will likely play an increasingly central role in our relationship with the deep sea. The integration of artificial intelligence, advanced materials science, and novel propulsion systems promises to further enhance these capabilities, potentially enabling long-duration autonomous missions that could transform our understanding of the least explored region of our planet. Through continued investment in research, development, and education, seafloor robotic investigations may finally illuminate the dark corners of our planetary ocean, revealing secrets that have remained hidden throughout human history.

Citations:

1. https://www.gu.se/en/kristineberg/infrastructure-and-services/labs-and-facilities/experimental-sedimentology-and-seafloor-robot-laboratory
2. https://www.nature.com/articles/s41467-024-54600-8
3. https://www.ceab.csic.es/en/research/seafloor-intelligent-robot-exploration-and-classification-sirec/
4. https://onlinelibrary.wiley.com/doi/10.1002/rob.22433
5. https://news.stanford.edu/stories/2022/07/oceanonek-connects-humans-sight-touch-deep-sea
6. https://oceanexplorer.noaa.gov/explainers/technology.html
7. https://www.science.org/doi/10.1126/scirobotics.abe1901
8. https://www.newscientist.com/article/2331946-underwater-robot-scans-seabed-to-seek-out-the-most-harmful-pollution/
9. https://www.researchnester.com/reports/underwater-robotics-market/1520
10. https://phys.org/news/2020-01-whoi-underwater-robot-first-known-automated.html
11. https://archeologie.culture.gouv.fr/archeo-sous-marine/en/robots-are-future
12. https://www.marinetechnologynews.com/news/seafloor-crawlers-robotic-tools-564662
13. https://builtin.com/articles/underwater-robotics
14. https://www.dal.ca/news/2024/01/26/galapagos-ocean-floor-technology.html
15. https://oceanexplorer.noaa.gov/explorations/22sampling-robotics/welcome.html
16. https://www.pnra.aq/en/project/333/robotic-based-investigation-and-monitoring-ross-sea
17. https://www.krakenrobotics.com
18. https://sea.edu/program/high-school-and-pre-college/sea-expedition-pacific-marine-robotics/
19. https://www.science.org/doi/10.1126/scirobotics.ads9444
20. https://mitportugal.org/events/2025/marine-robotics-summer-school/
21. https://www.designboom.com/technology/underwater-robots-battery-metals-sustainable-impossible-metals-12-12-2022/
22. https://www.whoi.edu/know-your-ocean/did-you-know/how-do-ocean-robots-take-the-pressure/
23. https://www.science.org/doi/10.1126/scirobotics.aaz1012
24. https://oceanexplorer.noaa.gov/edu/themes/underwater-robots/welcome.html
25. https://nauticusrobotics.com
26. https://www.youtube.com/watch?v=vuQ20ZKSWu4
27. https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2022.833617/full
28. https://www.mbari.org/news/mbaris-newest-underwater-robot-seeks-to-make-ocean-exploration-more-accessible/
29. https://collaborate.asce.org/blogs/s-pradeep/2024/09/02/international-conference-on-ai-for-the-oceans-2025
30. https://oceanexplorer.noaa.gov/news/oceanexplorationnews/welcome.html
31. https://www.whoi.edu/know-your-ocean/did-you-know/what-are-ocean-robots/
32. https://stellwagen.noaa.gov/maritime/archaeology.html
33. https://deepblue.lib.umich.edu/bitstream/handle/2027.42/84394/20350_ftp.pdf?sequence=1
34. https://washington.office.cnrs.fr/ocean-one-k-archaeologist-of-the-abyss/
35. https://www.openoceanrobotics.com
36. https://www.pbslearningmedia.org/resource/robotic-sampling-systems-video-gallery/ocean-exploration-inner-space/
37. https://cellula.com
38. https://www.youtube.com/watch?v=8e7AH__G8SI
39. https://news.mit.edu/2019/autonomous-system-sea-sampling-1104
40. https://phys.org/news/2016-04-france-humanoid-underwater-exploration-robot.html
41. https://sensortechcanada.com/2021/03/23/rest-in-greece/