SeaClear – The Usage of surface and underwaterdrones to fight plastic waste in oceans


Photo by Javardh on Unsplash

When we talk about sustainability, we often hear about 9 planetary boundaries. One of the best-known boundaries is climate change. This has inevitably gained notoriety in recent decades. However, experts often talk about the fact that there is at least one other plantar boundary: plastic waste in our oceans. More and more tons of plastic waste end up in our world’s oceans. The consequences of this continuous littering are serious. For this reason, it makes us all the more proud to be able to introduce Claudia Hertel-ten Eikelder. She is a project manager at the Hamburg Port Authority AöR, where she coordinates the SeaClear project. As the name suggests, it is about “cleaning up” coastal regions, some of which are struggling with devastating waste problems.

What exactly is her solution to the problem, who is working with her on this challenge and what else is possible through the use of drones on and under the water; that’s what we talked to her about.

homePORT: Claudia, can you explain what the project is about?

C: Sure. SeaClear stands for SEarch, IdentificAtion and Collection of Marine LittEr with Autonomous Robots and addresses one of today´s biggest topics namely ocean pollution. We will create a mixed team of Unmanned Underwater, Surface, and Aerial Vehicles — UUV, USV, UAV — to find and collect litter from the seabed and from the water column, focusing on coastal areas since that is where waste inflow concentrates. The UAV and one or several inspection UUVs map the litter, aiming to establish correlations between surface and underwater litter. One or multiple collection UUVs then classify and collect litter, using a combined suction-gripper manipulator for both small and large waste. The UUVs are tethered to offload power and computation to the USV. Our objective is to operate the robots autonomously, without remote human intervention, and to that end we plan novel developments in debris mapping, classification, and robot control. Today’s oceans contain 26-66 million tons of waste, with approximately 94% located on the seafloor. With an increase of the worldwide plastic production, we are estimating an increase of inflow into our oceans. SeaClear is like no other initiative dedicated to collect litter from the seabed.

homePORT: Who are the project partners and what is the role of HPA?

C: HPA is especially involved as potential end user and infrastructure supplier for tests and demonstration phases. At this point it is important to mention, that Hamburg Port doesn’t have a litter problem, that asks for a clean-up system like SeaClear. Nevertheless, there are metallic and magnetic items on the sea floor like rivets, that found their way into the water during historic ship building activities in Hamburg. Those are identified as anomalies during ordnance surveys prior to construction work and need to be removed. Classifying those with special sensory equipment, collecting harmless objects and marking questionable findings on a map for further investigation by e.g., divers is a huge benefit of SeaClear. Besides that, turbid water, tidal flows and currents as well as heavy ship traffic are challenges, that also exist in other parts of the world facing an actual litter problem.

Creating awareness for the topic in the population and tourists is part of the challenge in our second test area in Dubrovnik in Croatia. The city itself, the coastal region in general and the protected nature reserves are suffering from the waste inflow due to the adriatic currents. The perfectly clear water just makes the litter so much more visible.
In order to operate successfully in both settings, we are developing the system accordingly. The mammoth tasks in developing all technical components lies with our partners at TU Munich from Germany, TU Delft from the Netherlands, Fraunhofer CML from Germany, TU Cluj-Napoca in Romania and University Dubrovnik in Croatia and Subsea Tech from France.

homePORT: Why is this project so important?

C: The idea initially came from our HPA-Network and was quickly discussed with our colleagues from the environment and sustainability department. We decided to get engaged in the project, because it covers so many important aspects. My head is literally spinning, while deciding on how to answer the second part of your question first. Looking at it from a societal point of view, the project clearly appeals to a more responsible treatment of nature in general and our oceans specifically. It also asks us to rethink our consumer behavior. Existing initiatives range from unpacked shops and secondhand markets to repair and refurbishing services for technical devices. Politically the topic is found in the Chinese import ban on German plastic waste, the existing packaging law and the upcoming single-use-plastic ban valid from 03.07.2021. The technical innovations are extensive:

• The cooperation of various components – aerial drone, surface vehicle and underwater robots – in such a complex task is outstanding. Integrated underwater and surface mapping is crucial to task completion.

• We are using an artificial intelligence to distinguish waste and marine life, that is capable to recognize characteristic movements and shapes of sea creatures and litter fractions.

• An additional challenge is the autonomous collection of litter on the sea ground. No solution exists that exploits autonomous robots for underwater litter collection; the SeaClear project will develop the first.

• We are aiming for a transition in human operator control from initially direct control to assistance to finally only supervision of the robotic system.

homePORT: How is the importance of the project perceived by externals outside the project?

C: Whenever I have the chance to talk –private or business- about the issue and our project, I meet great interest. Everybody understands what the project is about, can relate to the idea and gets excited. At the same time there is a lot of respect regarding the complexity. To me, the most rewarding moment is reached, when the potential is recognized, and further ideas evolve. Examples to that could be taking water and sediment samples, monitoring in nature and species conservation, recovery of large items and boat equipment, maintenance of submarine infrastructure.

homePORT: Are there other R&D-projects focusing on the same topic?

C: Luckily there are numerous initiatives that are dealing with the topic in quite different measures. “The Ocean Clean Up” from the Netherlands, “Everwave” (formerly known as Pacific Garbage Screening) from Aachen, Germany, “one earth – one ocean” from Munich, Germany, the “waste shark” from the Netherlands, “elastec” from Chicago, USA, “seabin” from Australia, are just a couple examples. They all developed solutions to remove litter from the water surface, and so SeaClear is a complementary initiative. Some of them are far beyond research and development and in operation all over the world. To my knowledge they are also continuously improving their developments.

homePORT: Why will your idea succeed?

C: SeaClear in its current approach is unique. As a result, our partners at TU Munich, Fraunhofer CML and Subsea Tech will develop entirely new hardware components. Looking at it from a software point of view, the universities in Durbvonik and Romania will contribute to enhanced methodology expertise in broad software development. This left alone is outstanding already. Besides that, there are a lot of opportunities for further development. To me it already is some kind of a tool kit or multipurpose vehicle, capable to meet various tasks. Imagine it to collect litter fractions from a predefined location today, measuring shallow water areas within the port tomorrow and the day after tomorrow the gripper is replaced by a sediment drill and a pipette taking ground and water samples from the Elbe river, while the sonar is traded for a hydrophone capturing noise emissions. And in case the next construction site needs an explosive ordnance survey, SeaClear will support divers in detecting and recovering hazardous substances. Inspection and maintenance of ships and maritime infrastructure may as well be a field of operation. Once you remove the “port-goggles” a system like SeaClear may take a major role in environmental monitoring and protection by documenting the marine creatures, exploring their health state and way of life. It is also suited to supervise commercial, coastal aquaculture farms and offshore wind parks even if it may need an upscaling to fulfill some of those tasks.

homePORT: Can you give us some insights on the current status quo and explain the upcoming milestones?

C: We are standing at the beginning of year two in the project. We developed the use cases and defined the boundary conditions in Hamburg and Dubrovnik as basis to all further developments. The heart and center of SeaClear, the surface vehicle “SeaCat”, has successfully passed its first trials towards navigating autonomously in the waters off Marseille. We completed the sensor selection process within our budget and are planning to use a metal detector besides the sonar and video equipment. The construction of the hybrid gripper design has gone from prototype setup to a construction stage. The collection basket is designed and implemented in small scale. The hardware integration is currently simulated due to Covid-travel restrictions and the Artificial Intelligence training to detect litter is ongoing. We are also working on data collection and processing as well as setting up the communication interface structure. This year is basically all about technical developments and the respective progress. Upcoming milestones are the merging of all components to an overall system and the test and demonstration phases in Hamburg and Dubrovnik towards the end of the project.

homePORT: What are currently the main challenges?

C: Corona indeed is complicating the development process to some extent, because not all our partners can access their workshops. Technical competences are evenly distributed among the partners, which normally requires a lot more of personal exchange which simply cannot be achieved through online calls and conferences. Since we’re aiming to efficiently map the seafloor and detect litter to be picked up later, the project heavily relies on the perception and communication using on-board sensors, which are strongly impacted by the underwater domain.

In terms of mapping the environment, we’re aiming to use multiple reconnaissance vehicles. Herein lies a first challenge, to find a suitable control architecture to have cooperation between multiple vehicles, which also needs to deal with the reduced communication capabilities under water. While it is possible to simulate such an environment, in-situ trials will be very valuable to validate and benchmark our approach.

In terms of detecting and locating litter, the large variety of objects (and their various states of decomposition) raise a challenge. Especially considering that there are few public data sets available for the sensing modalities we are employing. Tests sites, such as those at HPA, are invaluable for gathering more realistic data and more data in general. The development of any type of detection algorithm, likely using a machine-learning approach, will surely improve when subjected to more and increasingly varied situations.

Overall, the reduced visibility and turbid nature of the water make these tasks considerably harder than those of mapping and detection using aerial or even ground-vehicles. Test sites with realistic situations may seem as more of a challenge in the short term, the payoff for deployment in the real world is highly important.

Now the technical challenges are to be solved within the project. But in general, even though autonomous waterborne vehicles are no news, the legal framework is still under construction. The requirements from our harbor master are also substantial to the successful, safe and efficient use of SeaClear in other (port) environments.

homePORT: homePORT has already opened its doors virtually and will also set up a physical campus in the port this spring. What opportunities do you see in the use of homePORT for your project?

C: Besides the above-mentioned aspects, the locations environment perfectly suits our demands, when it comes to proving the functionality under real time conditions, especially because the set up in Hamburg with the turbid water is an additional challenge to the system. The close connection to the city center is just as beneficial as the infrastructural facilities. Port Authorities and related institutions are part of our future client base. Having access to all potential stakeholders interested in such a system – either under the roof of HPA or within the attached network – contributes to a demand driven system development. I hope to gain exclusive knowledge regarding the optimization potential by taking the stakeholder perspective into account.

homePORT: Claudia, thank you very much for these deep insights into a project whose subject matter could not be more topical. We wish you continued success and will continue to follow the SeaClear issue.