Marine litter is a major environmental problem around the world. As part of the EU project SeaClear 2.0, a research team at the Technical University of Munich (TUM) has now developed an autonomous diving robot that can detect and retrieve litter.
It uses an AI system to analyze objects with ultrasound and cameras, picks them up and brings them to the surface. The autonomous underwater waste collection system demonstrated its capabilities for the first time in the port of Marseille in France.
In countless ports worldwide, divers regularly retrieve e-scooters, bicycles, lost fishing nets and old tires from the harbor basins. In Dubrovnik, researchers counted over 1,000 pieces of rubbish in an area of 100 square meters.
Autonomous waste disposal will soon provide a remedy. The entire system consists of an unmanned service boat with a dinghy, a drone, a small underwater search robot and the TUM diving robot.
According to Dr Stefan Sosnowski of the Chair of Information Technology Control at TUM, a cost-benefit analysis shows that waste disposal using autonomous underwater waste collection becomes profitable at depths of 16 meters or more.
TUM diving robot as part of an overall concept
Here’s how it works: The service boat supplies the underwater robots with power and data connections via cable. It also sends ultrasonic waves into the depths to generate a rough map of the seabed.
A dedicated search robot about 50 centimeters long quickly and efficiently scans the seabed. Armed with this information, the TUM submarine, powered by eight mini turbines, dives to the locations where rubbish is detected.
It then grabs the objects and uses a winch to load them onto an additional autonomous dinghy that serves as a floating waste container.
Features of the TUM underwater robot
System identifies rubbish objects and displays them in 3D
“Since we first have to identify the rubbish and grasping objects requires a high degree of precision, we have a camera and sonar on board that enable orientation even in murky water,” explains researcher Sosnowski. Identifying rubbish is no trivial matter.
This is because hardly any image material is available for underwater objects that could help to train neural networks.
“That’s why the project partners have so far labelled over 7,000 images as objects that don’t belong on the seabed,” says Sosnowski.
Once the rubbish is identified, the AI converts the images to 3D. “This is important for deciding where the object can be gripped securely,” explains Sosnowski.
Strong and sensitive gripper
The giant four-fingered hand on the autonomous gripper developed by TUM for the diving robot, which has a volume of approximately 1 cubic metre, can squeeze with a force of 4,000 newtons and grasp objects weighing up to 250 kg.
However, special sensors enable it to gauge how much force it can apply without causing damage. This prevents plastic buckets from breaking, for example, or glass bottles from shattering.
A cable connects the robot to the power supply and data network
Even though the TUM boat moves autonomously in the water, the researchers kept it “on a leash”. The reason: a battery on board would only provide power for about two hours.
In addition, the performance of the AI can be slightly increased by a cable connection. The cable also serves as a rope to pull heavy objects from the sea to the surface.
Buoyancy foam keeps the diving robot in place
The 120-kilogram submarine is surrounded by buoyancy foam that keeps it in a kind of suspended state in the water when the mini turbines are not in use.
This allows the underwater robot to move freely and maintain an exact course. “This is important for approaching objects precisely,” says researcher Sosnowski.
