This cutting-edge underwater robot can tackle ocean trash
The ocean is in trouble. Marine litter, from abandoned containers to fishing gear, poses a grave threat to marine life. Plastic waste breaks down into microplastics, releasing chemicals that enter the food chain, potentially disrupting plankton's carbon storage abilities and impacting climate change (https://www.cnn.com/2024/04/22/health/plastics-food-wellness-scn). Between 1950 and 2020, an estimated 32 million metric tons of plastic accumulated in the ocean, weighing as much as 200,000 blue whales (https://www.oecd.org/en/data/insights/data-explainers/2025/05/stemming-plastic-pollution-to-protect-the-ocean.html). Without intervention, that figure could reach 76 million tons by 2040.
A dedicated team of researchers and students is fighting back with an autonomous diving robot designed to detect and retrieve litter from the seabed. Meet the Smart Grapple, developed at the Technical University of Munich (TUM) in Germany.
This innovative robot utilizes AI to identify trash through onboard cameras or sonar imaging in murky waters. Once it spots an item, it employs a four-fingered hand capable of lifting objects up to a meter long and weighing 250 kilograms. It then brings the litter to the surface, where it's deposited on a small autonomous boat for recycling.
The Smart Grapple measures 115 centimeters tall and 70 centimeters in diameter, weighing 120 kilograms. It's connected to the boat by a cable for power, though it can also move underwater using its own thrusters. Stefan Sosnowski, the research team leader at TUM, explains that the cable acts like a crane's cable, allowing the winch to pull up the system efficiently.
The robot's AI displays the trash in 3D, ensuring precise grasping. A complex sensor system prevents the robot from breaking items into smaller pieces. Nicolas Hoischen, a doctoral student at TUM who's worked on the project for two years, finds the robot's capabilities awe-inspiring.
Hoischen's motivation stems from finding practical applications for his robotic systems research. He emphasizes the project's potential impact beyond the university, aiding ports and countries in tackling waste.
The Smart Grapple is part of the EU-funded SeaClear 2.0 project, aiming to map and collect marine litter using unmanned robots. Thirteen European organizations, including universities and underwater tech companies, contribute to this initiative. Bart De Schutter, the project coordinator, believes automated robotic technology could revolutionize marine litter cleanup.
De Schutter highlights the limitations of human divers in underwater litter retrieval, emphasizing the potential of robotic solutions to extend cleanup efforts. He anticipates widespread deployment of SeaClear's robots between 2030 and 2033.
However, Sosnowski acknowledges that the Smart Grapple isn't a panacea. He stresses the importance of reducing ocean littering rather than relying solely on robots. He envisions robotic systems being particularly helpful in specific areas like harbors and natural reserves under ecological pressure.
The Smart Grapple has demonstrated its capabilities at the Ports of Marseille and Hamburg, successfully retrieving a diverse range of litter, from car seats to cloth handkerchiefs. While some trials are still remotely controlled, the robot can operate autonomously.
Developing the Smart Grapple presented challenges, particularly distinguishing between waste and non-waste. Sosnowski explains that the robot was trained on over 7,500 labeled images of underwater objects, enabling it to recognize litter. This training created the largest underwater dataset available, crucial for the robot's accurate identification.
