PRE2016 3 Groep5

From Control Systems Technology Group
Jump to navigation Jump to search

Group Members

Student ID Name
0957942 N.S.A. Messaoudi
0958470 J.J.J.B. Verstappen
0955491 C. van Otterlo
0939540 M.J.M. Smits
0956810 W.J.P. Goudriaan
0953119 J.I.A. Spapen


Since 1945 plastic is inseparable from our society, where it brought us great fortune and great use. It is used to carry drinks, package foods, toys, etc. Whereas it has a good side that helps humanity, it also has its downsides. A lot of people throw their garbage, of which most is plastic, very easily on the ground and in rivers. This means the mainland and the oceans get polluted, instead of the plastic being recycled, when it is thrown away. On the mainland, there are a lot of organizations, which focus on preserving the mainland’s environment. For example by sending people to collect waste next to roads and beaches. A number of organizations in the seas and oceans are, however, greatly outnumbered. While the organizations on the mainland can hardly keep the pollution at a steady level, the pollution in the seas and oceans has risen exponentially for several years already. The plastic in the ocean decomposes gradually which results in a highly soiled sea which is more difficult to clean up. Even when the plastic decomposes into small parts, it affects and influences the marine ecosystem as a whole. A fish cannot tell food and plastic apart and slowly gets poisoned and killed by the toxic plastic in its stomach. The same happens to many birds that get their food out of the water.

A number of organizations in the seas and oceans are, however, greatly outnumbered. While the organizations on the mainland can hardly keep the pollution at a steady level, the pollution in the seas and oceans has risen exponentially for several years already. The plastics in the ocean decompose gradually which results in a highly soiled sea which is more difficult to clean up. The small parts of plastics influence the marine ecosystem as a whole. Fish cannot tell food and plastic apart and slowly gets poisoned and killed by the toxic plastic in its stomach. The same happens to many birds that get their food out of the water and even humans can be affected by this. If a fish gets poisoned and a human will eats that fish, then the human will also be affected by the effects.

The plastic that gets in the sea, gathers in so-called ‘gyres’. These are huge area’s to which the plastic floats due to the current. It slowly floats in circles in the same area. There are 5 huge gyres around the world. See Figure 1 below.


          Figure 1, the 5 gyres in the world

Currently, there is a project that is cleaning up these gyres. They put down a giant net in the gyres and let the current of the gyres pour the plastic in the gyres. This way all the plastic gets pushed into the net and collected in one place. However cleaning up the gyres alone is not enough. The plastic that is in the gyres came from somewhere and the gyres are still growing every day. That’s why it is also important to prevent the gyres from getting bigger, by cleaning up the plastic closer to its source.

The thing that has to be accomplished, is preventing the gyre from becoming bigger. Prevention starts with looking at the coast of The Netherlands. The goal that is set for this project is to try to let no plastic/garbage pass through the coast of The Netherlands. The current of the North Sea along The Netherlands is from the bottom to the top or the other way around, which depends on the tides. This is very convenient for setting up places for agents. When a plastic/garbage unit is not cleaned on the beginning of the flow, then it will be cleaned up later by another agent that is in another place at the coast. Also, a lot of plastic comes from the beaches and the harbors. It is, therefore, important to stop the flow of plastic/garbage as close as possible to the source. This is the reason why the agents are placed along the coast of The Netherlands. This will mean that the plastic/garbage will be cleaned up before it comes in the gyres.[1] [2] [3] [4] [5] [6] [7] [8]

Current situation

In order to tackle the problem, a good overview of the situation need to been set. This is done by writing down all the questions about the garbage problem in the North sea and then the searching for answers started. The strategy was as follows; first search for the answer on internet to get a basic intuition and then contact several organizations asking them all the same questions and combine all this information into one final answer.

The main sources that were used are either reports or the organizations that were contacted. The following are the reports used:

  • Wat spoelt er aan op het strand – Stichting De Noordzee
  • Mariene Strategie voor het Nederlandse deel van de Noordzee 2012-2020 – Rijksoverheid
  • Jaarverslag 2015 – Stichting De Noordzee
  • Guideline for Monitoring Marine Litter on the Beaches in the OSPAR Martime Area – OSPAR commission
  • The vertical distribution of buoyant plastics at sea: an observational study in the North Atlantic Gyre – J. Reisser, B. Slat, K. Noble, K. du Plessis, M. Epp, M. Proietti, J. de Sonneville, T. Becker and C. Pattiaratchi

The following organizations were contacted:

  • CBS
  • Rijkswaterstaat
  • Kustwacht
  • Clean up
  • Greenpeace
  • Stichting de Noordzee

Of course, not all organizations were as helpful, but there was enough response to get a complete answer to all the questions there were. Here are the questions that were asked, followed by the answer in italic.

  • Is the problem significant enough for organizations like the government to invest money in?

There are several organizations searching for solutions for this problem, which it is significant enough to be solved. Not just the pollution issue in itself is a problem for these organizations, but also the health issues when the garbage breaks down and ends up in the food chain. The government is the biggest investor since it is a societal matter and it is hard for a profit-seeking company to make money off. So if we can show the government that our project is the solution to this problem, our project could be worth investing in.

  • Which kinds of garbage objects are most common in the North sea?

There are multiple organizations (Rijkswaterstaat, Stichting Noordzee en NIOZ) that monitor the garbage on the beaches and in the North sea. With this data a list was composed, summing up the top 10 most found garbage objects in and around the sea.

    • Ropes and nets
    • Plastic pieces
    • Plastic bags
    • lids
    • Candy wrappers
    • Balloons
    • Plastic bottles
    • Wood
    • Plastic food packaging
    • Industrial plastic contents
  • Who are responsible for the garbage in the North sea and where are they located at the coastline?

More than half of the waste comes from the maritime sector. So logically the major port cities are geographical hot-spots with regard to the dumping of plastic waste in the North sea. Other pollution comes mainly from the people visiting the beaches, so crowded beaches are also potential hot-spots. Also, the rivers are highly polluted which makes the places where the rivers goes into the ocean as a dumping hot-spot

  • Are there hot-spots were the garbage gathers (under influence of the current), are these hot-spots on the coastline and/or in the sea? If they exist, where can they be found?

Generally speaking, the plastic gets smaller in terms of size the further one distances itself from the coast. Our focus is mainly on the larger pieces of plastic, so in combination with the previous question, the hot-spots can be recorded around the mouths of rivers and harbors and along the coastline crowded beaches.

  • What is the order of magnitude of the garbage that is in the North sea, do you have exact numbers?

Regarding soil pollution of the North Sea, we are talking about 110 pieces of plastic per square kilometer in the North sea. Pollution along the coastline is 380 pieces per debris 100 meters beach.

  • At what depth is can the garbage be found?

The greatest concentration of debris is around or on the surface. There has been measured up to a depth of 5 meters, and from this, it was found that 80% of the waste was in the first 2-3 meters. Which will be further explained in the chapter garbage distribution in the Appendix.

  • Are you familiar with a comparable project like ours, if so could u give us some information about the projects?

A similar project is the ocean clean up. The idea is to be achieved on a large scale waste in the North Pacific gyre using two long arms. However, this project is on a much larger scale with a higher concentration of waste per km ^ 2 so not really applicable to our problem.

Another project was undertaken by water network. It has created an autonomous agent, named Nautonomous, which navigates through the water and gathers waste on a treadmill, this agent can be seen in Figure 2.[9] [10]


     Figure 2, Nautonomous

Another similar project was set up by the recycled Island Foundation. they make a sort of trial and place it on the coast of Rotterdam in strategic places and capture that way plastic. [11]

Environmental aspects

Work space of agents

In the North sea, there are some areas that cannot be traversed or are very busy with traffic. Along the coast of The Netherlands, there are areas that are protected by law. This is however mostly because of fisherman. That means that the agent will not be restricted, when it comes to those areas since it does not fish or harvest any other natural products. The problem that should be taken into account is the ship’s navigational routes. There are areas where there is a lot of boat traffic, especially at the ports. Along the coast, this is not an issue. (Appendix Environment)[12][13][14][15][16]

Waves and wind issues

When an agent is at sea it is not safe for the agent when there are high waves. It could break because it gets destroyed by a big wave. Looking at how waves are created, it is found that the wind is playing a big part in it. The harder and longer the wind blows, the bigger the waves will be. It is, therefore, hard to say when the agent should go back at a certain wind speed. However, some things can be determined, like at what wind speed it is dangerous and how big waves will get at a certain wind speed. For small boats, for example, canoes, it is recommended not to go on open sea if the wind speed is bigger or equal to 6Bft. Which means a medium wind. In table 2 is shown how high the waves will get at a certain wind speed.

     Table 2, Influence wind on wave height


When the waves get higher than 1,7-2 meters it is too high for the agents. Before the agents will go on the water, the wind speed is measured. Also, while they are working the wind speed should be updated. This way the height of the current waves can be determined. When this calculation is done a decision can be made whether the agents should go to work or not. Also when they are working and the wind speed gets too high, it can be determined after how much time they should be called back to the coast. [17] [18]



An often overlooked problem are birds sitting on the boat. This is not convenient since a part of the boat might break or can get dirty. This problem is not hard to solve because there are some small tips and tricks that can prevent them from getting on a boat. For example, getting some netting and putting it above places they can land. Putting a fake crow on the boat or a string with cd’s attached to it will most likely scare the most of those birds away.


There are a lot of fish in the sea, which poses a problem. It would not be the intention to accidentally catch fish instead of garbage on the way. However, it is assumed, that the agents will make a certain noise and they won’t go too fast. This means the fish have plenty of time to see it coming and have enough time to get away.


Different concepts

During the project, a few different designs and possible solutions were created. These vary on a wide spectrum. Following are some of the concepts created.

The retrieval arm

The first design was to construct a boat with a number of arms on it to pick up each piece of garbage separately. While this would be a good option since it could grab pieces that are two to three meters deep. The programming of an arm is a whole project in and of itself. also, this design would make the boat too unstable when picking the garbage out of the water due to a large amount of mass moved. Nets on either side The second design was to make a boat with nets on either side that would rotate whenever a piece of garbage is caught. However, this could also, catch fish which is unwanted. Also, the design is not energy efficient because of the fact that it needs to move the nets in a rotational way which would consume a lot of energy.

Conveyor belts

The third design was to make a boat with conveyor belts to guide the garbage to a center flow area where they would be scooped up by another conveyor into a compartment. This design also was not very energy efficient. A lot of energy would be lost in moving the conveyors and the non-hydrodynamic design. It also could only pick up garbage that floats on the surface of the water.

Ramp to catch garbage

The final design was to make a boat with a ramp in the water. This way the garbage drifting two to three meters deep in the water can be pulled to the surface where they will be transported via conveyor belts on either side of the boat into a compartment for storage. This design will generate a lot of downwards force due to the giant ramp in the water, this can, however, be prevented by putting a floater or engine in the front to compensate for the downwards force.

Requirements, preferences and constraints

To make sure a good design is made there is a large list of RPCs constructed. These are split up into the following categories to make it easy to read:


  • The boat needs to be 2 by 3 by 3 meters. This is so that the boat can still hold a lot of garbage. It also makes sure the boat is stable and it provides a large frontal area to pick up garbage.
  • The compartment needs to be 1.5 by 2 by 2 meters, this way it can carry a lot of garbage without having to be emptied.
  • The compartment needs to be separable in order to decrease downtime when emptying.
  • The clips holding the compartment in place need to be easily removable for easy separation when emptying.
  • The vehicle needs to have two floaters on each side (0.25 by 2 by 2.5 meters) in order to keep the boat floating in the water. Via Archimedes’ law, this means that the floaters will stay in the water for 80 percent with a full compartment.
  • The boat needs one ball-shaped floater (radius 0.6 meters) to compensate for downwards drag when moving through the water.
  • Space needs to be available for the conveyor belts on either side of the boat. These are needed to pull up the garbage once it is out of the water and transfer them to the compartment. This space would equate to 0.25 by 1 by 0.5 meters.
  • There must be a guiding slide to transfer the garbage pulled up by the conveyors. The space would equate to 0.25 by 1 by 0.2 meters.


  • The boat and the compartment are both made of polyester. This material is used in the construction of sailboats and is both strong and light.
  • The floaters need to be made of poly-ether since it is the most used material for floaters in general.
  • The whole vehicle and all of its parts need to be waterproof to prevent it from sinking. This can be done by using a coating over the polyester. According to the sources used this holds for about 25 years. [19] [20]


  • The vehicle needs to have a motor to move forward. In this project we have chosen for a ring motor, this motor should be strong enough for small boats up to 8 meters and therefore should be sufficient for our agent. [21]
  • It needs to have a fin located in the water to steer.
  • It needs to have a large enough battery in order to fill its compartment and return to a charging station.The battery life should be at least 12 hours, however, to determine the exact battery life the size of the agent, the speed of the agent, the engine distribution, the conveyor belts and the fin has to be looked at in more detail.


  • The vehicle needs to have a laser sensor in order to check if its compartment is full and needs to emptied. This laser gives the "full" signal when its trajectory is interrupted. However, this means that when the plastic is tossed into the container, it gives that signal. Therefore there need to be some kind of counter, which makes sure the signal gets send if and only if a certain time is past.
  • It needs a GPS system to locate its position and to make it back to the nearest pickup point when a storm is inbound.
  • The agent needs an inertial measurement unit in order to make sure it is still on the right path and the right speed when picking up garbage.[22]
  • It needs to have an internet connection to check the weather for conditions in which the agent should not be on the water and to communicate with the control station if there is a part that has broken down or the agent needs to be called back to a pickup point in event of an emergency situation.


  • The agent needs to keep working for 25 years, to minimize costs and to make sure it can keep cleaning the ocean, which unfortunately is never finished.


For the agents to gather plastics as efficiently as possible a gathering strategy has to be made. To cover the whole area, a strategy in which the agents move vertically would be a logical approach. When this strategy is used, it seems that almost all of the plastic on top of the surface of the water will be gathered. However, due to the fact that most of the plastics in the North sea are found at a depth of 1-3 meters, this strategy will not be sufficient enough. Therefore a new strategy has to be considered.

To gather the plastic that is 1-3 meters deep in the water, 2 slides are designed in front of the agent. These slides are 1-3 meters deep and push the plastics to the surface of the water. To stimulate this upwards thrust, the strategy will be adapted to the figure below:


     Figure 3, Strategy

In this strategy, the agent will navigate in the opposite direction of the current of the water. In this way, the plastics will collide to the slides and will be pushed due to the cooperation of the current and the speed of the agent. The best way to stimulate the thrust is to navigate the agents horizontally, parallel to the coast. However when the total area has to be covered there will be a lot of agents needed to gather all the plastics, therefore the strategy in the figure is the final strategy. In this strategy, the following are presumed. The area which the agents will clean has a width of Y meters and a height of X meter. When tides shift, the current reverse direction. The tides shift every six hours [23]. Furthermore, the average speed of the current is equal to 1.15 meters per second. According to the RPC’s the agents need to navigate with a speed of at least 3 meters per seconds, this means that in 6 hours an agent can travel a total amount of 64800 meters. To estimate the size of the area it will be presumed that the agent finishes n spikes at the total width of Y meters. De X and Y can be calculated by the following Equation 1:

     Equation 1

64800=2n*√(x^2+(y/2n)^2 )

When the Y varies between 100 and 300 meters the height X can be calculated with the following Equation 2:

     Equation 2

x= √((64800/2n)^2-(y/2n)^2 ) (2)

In the table below the different Y and X are shown, together with an approximation of n.

     table 1, the size of the area.


The X was estimated to be around 50 meters and therefore the size of the last row is chosen, with an approximate value of n.

Furthermore, there is assumed that at least 10 agents should be behind each other to gather plastics when the agents in front of them miss them. Furthermore, these extra agents are necessary to cover the area when the other agent is emptying its container.

In the model part, the optimal amount of agents and the optimal angle in which the agents gather the plastic in the area is chosen by simulating different agent numbers and different angles.

The question remaining now is where the agents have to be placed to clean up the North sea, while not being disturbed by boats, birds, sailors and so on. It is very difficult to take the birds that rest on the agents into account, therefore the focus lies on human disturb factors. According to the information provided in the environment chapter, the following locations are proposed:


     Figure 4, Locations

These locations are based on tourist locations, harbors and places where rivers go into the sea. These locations are according to state of the art research the hot-spots where plastics goes into the North sea. As can be seen in the figure above, the direction shift of the currents is also taken into account by placing agents on both sides of the river.


Model design in CAD

During the project two models were made, the first one functioned as a basis to make decisions upon, such as what sensors to use, pickup strategy and efficiency among others. The second model created is the final design, this design solved the practical problems and limitations that came up during the rest of the project.

the first design

The first thing that came to mind when designing the agent was the fact that it needs to pick up garbage efficiently. A lot of options came up in the initial designing phase. Arms to pick up the garbage, giant nets to trap garbage in etcetera. However these options were not efficient enough, it needed a lot of programming to move the different parts and a lot of energy would be lost in picking up something. So the decision was made to design an agent that could pick up garbage by moving through the water. The main inspiration for this was the nautonomous project and some concept designs made. [5] [24]

And so the first design was made, it uses conveyor belts on the outside to move the garbage horizontally towards the center in front of the agent and then uses a second conveyor to move it from the water in a compartment. The agent would be kept afloat using two floaters on either side and can move forward and steer using two motors at the end of each floater.

However, the first design had a few problems. First of the conveyor belts were fragile and would get a lot of pressure put on them when moving through the water against the flow. The outside conveyors were also held in place by relatively thin beams which would break easily. The second problem with the original design was the fact that the front area was big and was not hydrodynamic which would lead to a lot of water friction when moving. And lastly, most of the garbage which would be floating in the seas is up to two through three meters deep in the water. With all the limitations mentioned before, the agent would not be able to pick up all of the garbage. Since the friction forces would get too large when trying to get garbage out of the water.


     Figure 5, First design

The second design

With all the limitations in mind, the second design was made. The new design picks up the garbage in the front and guides it to the surface without a large friction force. From there the blue conveyors will pick it up and guide it upwards. The conveyer belts will drop it off on the green panels which guide it further to the storage compartment. This way the design has fewer dangers in breaking, compared to the first design. It no longer has belts kept in place by thin beams, but a big front plate integrated into the front. The only conveyor belts left are located above water and therefore do not face such large forces as the first design. Secondly, the new design generates less friction due to the front plate and can pick up garbage which is located as deep as two to three meters. All together this will greatly reduce maintenance and power usage. And lastly, it has a compartment that is easy to be taken out and emptied.


     Figure 6, Second design

Considerations for Cad model

Detecting Waste in the ocean

In this section, three types of agents will be defined. The first are agents that clean the ocean via brute force, this means they keep cleaning the same designated area over and over until they have run out of battery to the point their program tells them to move to a charging station. The second type is agents that use sensors to recognize the shape of plastic and other pollutions and move to the location to clean it up. After they’ve cleaned up the garbage they start scanning again. And lastly, the third type are agents that use ultrasound to scan the ocean for garbage and use the same cleaning procedure as agent prototype two. Below follows a description of the different types, good and bad aspects and both construction and operating cost.

Brute force option


This prototype will have all the basic necessities, this means storage, motors, the ability to float on water, GPS and a mechanism for retrieving garbage from the ocean. Its way of working is via brute force: from the central communication center, it will get allocated to an area of the ocean. Once the agent arrives at its location it will start moving over the whole area, sweeping up every piece of garbage on its way. Once it is done it will start over again until the battery level hits a certain threshold or the compartment is full, after which it will make its way over to the charging station and get its compartment emptied as well.

Good and bad points


  • Simple design
  • When the path of cleaning is well designed, it should have few pieces of garbage seeping through its area
  • Compartments can be smaller due to the fact that more agents need to be used for cleaning


  • Needs a lot more agents than the other two options due to the fact that it is using brute force instead of targeted cleaning, or the cleaning can take much longer than when using the other options
  • Even with good cleaning paths the garbage can get through
  • More agents means more charging options and more substitution agents to fill up the gaps from traveling agents, which can, in turn, cost more

The costs of this option will be the least of all the options listed. Due to the fact that this option uses only the features that are absolutely necessary for this agent. But the fact that more agents could be necessary (see above), the costs could go up.

Shape recognition option


This prototype will have all the basic necessities as described in the previous prototype, but it will also be outfitted with a camera and shape recognition software. From the central communication center, it will get allocated to an area of the ocean. Once arrived it will start scanning the area for shapes it deems worthy of cleaning up. When it sees a shape it perceives as plastic, it will move over to the location and clean it up, storing it in the compartment. When done it will move on with scanning and the cycle will start over again until the battery level hits a certain threshold or the compartment is full, after which it will make its way over to the charging station and get its compartment emptied as well.

Good and bad points


  • Targeted cleaning means fewer agents than brute force cleaning
  • Fewer agents means less charging options and substitute agents which means lower construction and operation cost


  • Technology is not yet advanced to the point that it is a feasible option due to the recognition range
  • Pixel distortion and image distortion can lead to recognition of non-existent garbage
  • In scanning and cleaning garbage can seep through if the software is not correctly calibrated.
  • Cameras have no depth perception when using a single camera since it only tracks pixel data and no distance
  • Some things that are garbage might not be received as such

The costs of this option would rise severely owing to the fact that shape recognizing technology is fairly new to the market, and new technologies tend to cost more than technologies already longer existing. The implementing of this technology could also cause a rise in the costs. The size of the sensor would cause an increase in the dimensions and weight of the agent, among other things.

Ultrasound option


This prototype will have all the basic necessities as described in the first prototype, but it will also be outfitted with an ultrasound device. From the central communication center, it will get allocated to an area of the ocean. When it gets there it will start scanning the ocean for pieces of plastic and other garbage. When it sees something it perceives as garbage the same procedure as the second prototype is initiated.

Good and bad points


  • Targeted cleaning means fewer agents than brute force cleaning
  • Fewer agents means less charging options and substitute agents which means lower construction and operation cost
  • Wide scanning area
  • Tracks depth
  • Can recognize garbage that is underwater


  • Small objects might not be recognized, as they are seen as noise or get lost in continuous error and noise adjustments.
  • Maritime life can be seen as garbage and therefore cleaned up.
  • Presumably very sensitive to adjustments
  • Might have low accuracy compared to prototype two

The costs of this option will lie in between the costs of the other two options. The ultrasound technology has been on the market for a longer time, so it is further developed, and also cheaper and less than shape recognizing sensors. Another advantage is that waterproof ultrasound sensors already exist and could easily be bought. Ultrasound sensors sell at prices between $10 and $250. The scanning range has to be determined, in order to choose one type of ultrasound sensor, and with that its price.

Our choice

We chose the brute force option because the agents work in teams on different spots in the North Sea, which lowers the importance of picking up every piece of waste that passes the spot in which the agent works. If one piece passes one agent, another agent can pick it up. For this agent, there is no need to add the extra features mentioned above, and this will only increase the costs of the agent.

Picking up waste from the ocean

There are different options from which can be chosen regarding picking up waste from the ocean. A simple fishing net, a sort of arm with a basket attached to it, and a slide which would force the waste to the surface of the ocean, where it could be picked up by a conveyor belt, were all considered. Below, the different (dis-)advantages and costs of every option are discussed.

Fishing net


This option would be about the same as the fishing nets used by the fishing industry (see Figure below), but smaller and used only two meters below the surface of the water.


     Figure 6, Fishing industry boat.
Good and bad points


  • Can cover large patches of ocean
  • Many different net sizes to choose from
  • Many options for the size of the holes in the nets
  • Waste of all sizes can be picked up


  • Hard to determine the right size of the holes in the net, too large and waste may not be trapped, too small and other things than waste may be trapped in the net
  • Fish that swim close to the surface of the water could also be trapped by the net
  • How to empty the net into the onboard storage container without humans on board?
  • Takes up a lot of space

Recreational fishing nets aren’t that expensive, but when looking at more professional fishing nets, the costs can rise pretty quick, they can cost between 100 and 200 euros.

Arm with basket


A visual representation of this option can be seen in the image below. A basket, with small holes (so the water doesn’t end up in the storage container), can move up and down around an axis, picking up waste objects.


     Figure 6, Arm with basket agent.
Good and bad points


  • Can pick up every piece of waste passing
  • The size of the basket can vary, so bigger waste objects can also be picked up


  • The friction force of the water causes the need of a strong motor or servo for turning the axis
  • What material is the basket made from? There could be the need of designing this from scratch, and this could cost a lot of time
  • If the basket is in the process of picking up a piece of waste, and another piece passes, it can’t pick up the other piece of waste, this is inefficient
  • All the electronics need to be at the back of the agent, which can cause the agent to be very heavy in the back and therefore unstable

Since this option will be needing a strong motor or servo, the costs can rise pretty quickly. Also, the material of the basket could be something that has to be designed from scratch, which can cost a lot of time and money. A simple fishing net could be used, of course, this could also solve the problem of the friction force of the water. But using a simple fishing net gives the problems mentioned in the section of the simple fishing net.

Slide with conveyor belt


This option will use the upward force of water. On the front of the agent, a slide is made. When a waste object is in the path of the agent, it will bump against the slide, the slide will force the water and the waste object upwards (toward the surface of the ocean). When at the surface, a conveyor belt, made of plastic, with plastic spikes, will take over from the slide, and bring the plastic object further upwards until it falls into the storage container. The conveyor belt will have holes in it, so that the water, which has been forced upward, falls back into the ocean.

Good and bad points


  • Every piece of waste in the path of the agent is picked up when using this method
  • Using forces outside the agent to pick up waste (water’s upward force)
  • All the agent has to do is sail its path and keep the conveyor belt running
  • No fish and other things than waste will be picked up using this method


  • All waste objects that aren’t in the agent’s path aren’t picked up
  • It may cost a lot of energy to keep the conveyor belt running all the time

The slide and the conveyor belt can both be made out of plastic, which just has to be shaped in a specific form. Since the form of a slide is used, the costs of a children’s slide are used as a base for the costs of the slide. These cost around 30 euros. For a plastic conveyor belt of 60 cm wide, a price of about 150 euros per meter has been found.[25]

Our choice

The choice fell on the slide option. One of the major advantages was the fact that no harm is done to marine wildlife when using this method. The goal is to clean up the ocean, not damage its ecosystem. Furthermore, the disadvantage concerning not picking up every piece of waste can be countered by the same argument used for the brute force navigational option: other agents can pick it up. The simplicity of the solution is also a plus, the agent does not need any extra sensors of knowing when to rotate (like with the basket). The extra costs for the slide and the conveyor belt are well spent if marine wildlife can be spared.

Netlogo model

In this part of the project, a model was made that simulates a certain amount of agents cleaning up an area of the North sea with a width of 300 meters and a length of 54 meters. The purpose of the model was to determine the optimal number of amount of agents by simulation. As said earlier in the chapter about the strategy, the velocity of the agents has to be at least 3 meters per seconds. In the model, this has become 4 meters per seconds because the chance of missing pieces of plastic is smaller. This is determined by simulating the model a few times.

The agents have to empty their container when their containers are full. In the total width of 300 meters, there are 6 stations, where the agents can empty their containers and refill their energy level. The initial energy is chosen very high, so the chance of getting the container full is greater than the chance of running out of energy. These stations are located at 25, 75, 125, 175, 225 and 275 meters. The charge time is estimated at around 8 ticks because the emptying the container will be executed as efficiently as possible.

The rest of the model can be seen by clicking on the link below:

The result of the simulation with the respect to the number of agents are given in the tables located in the chapter data of model in the Appendix. When the date retrieved is taken into account the following can be concluded. According to the model 13 agent are needed with a agent distance of 2.5 meters. This means that 1,64 percent of plastics are missed.


During the project, certain aspects of our progress could have been executed a lot better. Especially at the beginning of the project, the progress went very slow. We could not make decisions on what to focus on during the rest of the project, which resulted in a few shifts of our problem’s composition. When our problem was finally stated in a clear manner, we could finally focus on problem-solving, which resulted in the solution given in this report.

Another problem in our progress was the explanation of choices that we made with respect to the RPCs. Some choices were made by making assumptions, however, it was not clearly stated in our wiki at the time. Another set of choices were based on facts, however, lacked the sources needed. In short, we had to elaborate more on the choices we made, so the tutors could see why those choices were made validly. We have learned a lot about the importance of elaboration of choices and it also led to discussions that have improved our project in the end.

In the group, we had sometimes trouble with meeting up and finding time to work on the project as a group. If we had more time, then we would be exploring more options for strategies and optimizing it more.

In this project, we could have benefited by making a more structural planning because now the state of the art research went slowly. We should have contacted the different organizations earlier in the project. The contact we made with the organizations provided information which we used in the problem-solving part of the project.

If we had contacted them earlier in the project, we could have found a solution much earlier, which would have resulted in more time for optimization and model making.


The goal of our project was to make sure that waste located in the North Sea doesn’t move to the gyres in the open ocean. To do this, we decided that an ocean cleaning agent needed to be designed to sail the North Sea and collect the waste floating there. We started out with researching the state-of-the-art of this subject. This included literature research and contacting organizations which might know more about this subject. Using the information gathered from this research we started the design of the agent. The most important choices were the navigational strategy and the method of picking up the plastic because only then a model for the agent could be made, and a NetLogo environment could be created.

The goal of the agent was to pick up as much waste, if not everything, as possible. A strategy has been developed to reach this goal. The strategy that has been developed, which is stated in the report, has been tested in NetLogo and it only misses 1,64 percent of the waste. This efficiency is accomplished when 13 agents, with a distance of 2,5 meters between them, are working. The workspace has a width of 300 meters and a length of 54 meters. We consider this a strategy that works very well.

Though our agent does pick up most of the plastic it encounters, and when working in teams, 98,36 percent of the plastic is picked up, the North Sea will never be totally clean if dumping of waste into the ocean doesn’t stop. If the dumping keeps going, the ocean will still be polluted with waste, and our agent would need to work forever. In the scenario where the dumping of waste into the ocean stops, our next focus could be to clean up the gyres in the ocean. The ultimate goal is, of course, to make sure that there is no waste at all in the oceans, but we still have a long road ahead before this could ever be achieved.


Garbage distribution

Before the designing of the agent can start an impression of where our target, the garbage, is located is needed. After contacting several organizations a proper estimation could be made of how the garbage is disturbed in the North sea.

At first the depth of the garbage will be elaborated. Research has shown that in a water column of 5 meters, starting at the surface, 80% of it can be found in the top 2 to 3 meters [26] . However, these numbers follow from measurements done in the North Atlantic Gyre, so do they also hold for the situation in the North sea? Without going too much in scientific detail, we will try to explain if the North Atlantic ocean is comparable with the North sea. An object floats if it has a smaller density than the liquid. Density is a ratio of mass and volume and since these quantities are the same in both situations we can assume the results of the study are also applicable for the North sea.

The second variable is the distance of the plastic to the coast. Since we couldn’t find any reports with the necessary data we contacted Rijkswaterstaat. It turned out there isn’t any specific data, but the rule-of-thumb is; the further away from the coast, the smaller and less dense the garbage gets. Combining this information with the location of the hot-spots (outlet of rivers, harbors and beaches), only a few potential locations remain.

data of model

     Table 3, Dataset 1


     Table 4, Dataset 2


     Table 5, Dataset 3




     Figure 7, Fishing areas North sea[14].


     Figure 8, Shipping routes North sea[27].


  5. 5.0 5.1
  14. 14.0 14.1,%20recente%20ontwikkelingen%20en%20toekomstscenario%E2%80%99s_4125.pdf/
  26. The vertical distribution of buoyant plastics at sea: an observational study in the North Atlantic Gyre – J. Reisser, B. Slat, K. Noble, K. du Plessis, M. Epp, M. Proietti, J. de Sonneville, T. Becker and C. Pattiaratchi