Introduction: "Scoutbot Kuching" Robotic Boat

About: Education: - Sailing robots: Protei Inc & Scoutbots CEO Personal:

"Scoutbot Kuching" is a semi-autonomous boat that is targeted to monitor marine protected areas, or map shallow coral reefs. It has both remote control (RC) capacity as well as route planning using the Pixhawk microcontroller.

  • Footprint: 1.2m x 1.0m x 0.4m
  • Weight: <10kg
  • Speed: 0.5m/s for Auto mission (2.0m/s max)
  • Runtime: 4S (14.8V) 10000mAH
  • 1 hour Coverage = 7000m2/hour; ~340 pictures (assuming 5m depth water)
  • Image Resolution: 0.3 - 2cm/pixel (GoPro). Can upgrade better camera or multi-spectral
  • Depth of observation: Observe from 0.5m to 10m depending on visibility
  • Control/Telemetry Range: 2km (Recommend to add a front facing real time image transmission system if moving beyond line of sight), Manual Override, Auto Return Home
  • Optional Features: Obstacle Avoidance, Environmental Sensors, Real time telemetry sync to cloud
  • Servicing: Ship back defective module to be replaced
  • Hardware Cost: <1000 USD
  • Time to build: 4 days with 3 experienced fabricators + 1 young assistant

This project was built by the Scoutbots team ( Cesar Jung-Harada , Ken & Joe Chew, Eddie Yung) for the Make4ThePlanet Hackathon by Conservation X Lab , during the IMCC5 (5th International Marine Conservation Congress) in Kuching, Sarawak, Malaysia between June 24-28 2018. More prototypes by the Scoutbots team here .

Step 1: Why? Monitoring Marine Protected Areas, Mapping Coral Reefs

All life as we know it comes from the ocean, yet the oceans are under threat. Marine Protected Areas (MPAs) are one of the most effective and low-cost ways to protect our oceans while making them more productive and financially profitable. Currently only about 3% of our ocean has strongly protected waters while scientists agree that 30% are necessary for a sustainable ocean. We have a long way to go to create more MPA, monitor and enforce them. The brief was presented by former NOAA Administrator Jane Lubchenco below.

"There is broad acceptance of “green parks” on land, but demonstrating positive impacts that ultimately effect nearby communities with “blue parks” remains a challenge.

MPAs are constrained by three major factors in public opinion:

  1. Lack of awareness about the low area coverage of MPAs;
  2. Lack of knowledge and awareness of massive conservation benefits of fully protected MPAs;
  3. Lack of knowledge and understanding that those benefits spill out to adjacent areas and help make the oceans more resilient.

MPAs are further constrained by forces that impact their existence and efficacy:

  1. Commercial interests and small-scale fishers and communities who use the ocean now see MPAs as a negative, as a loss, and not an investment in the future;
  2. In some communities, community-based management projects offer some respite for depleted ecosystems and populations, but they require clearly delineated rights, and may not scale beyond the community;
  3. Real-time monitoring and enforcement remains a critical challenge that while technology exists, action is much less frequent and politically

How might we create tools and systems to not only raise local awareness of the (ecological, cultural, and financial) benefits of MPAs, but also make sure that the protected areas maintain a viable ecosystem with healthy populations of marine life?

What kinds of scalable tools and systems can accurately monitor and enforce the protection of MPAs in new ways, so that “enforcement” is not perceived as yet another reason for powerful lobbies to view MPAs as negative? What kinds of functions can these technologies help augment and improve the functions of policy makers and managers of MPAs?

What are the tools and systems needed formulti- stakeholder sustainable use of MPAs (cultural, economic, and environmental)? How might we create tools to design MPAs like how terrestrial multi-use protected areas are designed – what are the ecologically relevant spaces, what are the spaces for multi-use, how do you design for both economic outcomes & protection of marine life?"

In this slideshow, we explain why created this surface marine drone: to monitor Marine Protected Areas (MPAs) and map coral reefs. This is a platform technology and can be used for many purposes, easily extended as it is open source.

if you want to know a lot more about coral reefs and fishery decline around the world, you can read on here:

"Coral reefs could be gone in 30 years" By Laura Parker and Craig Welch PUBLISHED JUNE 23, 2017

"One of the World's Biggest Fisheries Is on the Verge of Collapse Major disputes in the South China Sea are putting critical habitat—and the food supply of millions—at risk."

Step 2: Hulls

We used simple thin sheets of corrugated plastic. When it was not long enough, we connected them with transparent tape. Remember, although our machine could sail in the pool, this is only a hackathon prototype. You could use the same material as a base, and fiberglass it with epoxy resin to make it more durable.

The shape of the hull is a simple triangle at the back, a line at the front, the first 1/4 being the transition from line to the triangle. The hull has a good penetration and displacement so it can carry quite a lot of equipment.

Waterproofing of the hull was done with

  1. narrow transparent tape (for positioning)
  2. medium width duct tape (to smoothen edges)
  3. wide aluminum tape (to waterproof)

This increase in width creates several incremental overlaps that guarantee to waterproof while being a cheap and fast way to prototype.

Step 3: Bridge

Originally we thought about making the bridge out of the same material, corrugated plastic. But the box being quite heavy, and since we would have to add 2 powerful thrusters, we started looking for faster alternatives / hacks.

The event was sponsored by MR DIY just nearby, Malaysia large hardware store chain. We found a metal shoe rack that really fitted well our dimensions!

Step 4: Propellers

We then positioned the 3D printed propellers on the shoe rack frame. To be more accurate, a better way to name this part would be "Underwater Thruster Brushless" each with a 250W capacity. Typically, such ROV parts are super expensive. Even on Alibaba such motor go for more than 1000 USD a piece. But Ken 3D printed his own, so it came out for less than 100 USD a piece.

In the USA, Blue Robotics is Education programs' favorite

Step 5: Dry Box

You could easily use a nice Pelican case for this project. The Pelican Case 1120 or anything bigger would work for this project. This is the official website for Pelican ; but being in Hong Kong / China, there is no shortage of cheaper dry case that we felt comfortable drilling through.

Step 6: Motor Controller

we used 2 HobbyKing 50A Boat ESC 4A UBEC.


  • Constant Current: 50A
  • Burst Current: 70A
  • Battery: 2-4S Lipoly
  • BEC: 5.5v / 3A
  • Motor Type: Sensorless Brushless
  • Cooling: Watercooled
  • Size: 87 x 38 x 22mm
  • Weight: 105g

Programming Functions:

  • Running Mode: Forward with brake / Forward and reverse with brake / Forward and reverse
  • Acceleration: 6% / 9% / 12%
  • Low Voltage Protection: None / 2.6V / 2.8V / 3.0V / 3.2V / 3.4V
  • Start Mode (Punch): Level 1 / Level 2 / Level 3 / Level 4 / Level 5 / Level 6 / Level 7 / Level 8 / Level 9
  • Reverse Force: 25% / 50% / 75% / 100%
  • Timing: 0° / 5° / 10° / 15° / 20° / 25° / 30° / Automatic

We had some overheating issues. Also we did not take the time to wire the water cooling channels. It should be ok if you do.

This is a simple way to wire the ESC:

Step 7: Battery

We run this particular prototype with 10,000mA/h, but there is a variety of batteries that we could have used. This video is a great guide to choosing a battery that will suit your project. Our dry box is big enough to house 3 such batteries giving us a hefty autonomy as we run our thrusters at 25% speed max. Being a boat, weight is not as much as issue as when you are buidling a flying drone. In fact, the heavier, the better, the more stable the images you will get.

Step 8: Sailing Control

We love our Pixhawk 4 flight controller. Normally such microcontroller is used for flying machines, but many people have hacked that controller for terrestrial vehicles like cars and boats. This is a very complete flight controller, so I will let you watch this video to learn how to install and test run it.

Step 9: Telemetry, GPS, Firmware

Our intention is to monitor an MPA. Usually, that means that we would have the GPS coordinates of the area we want to cover. We can use these coordinates to prepare our "Flight Mission" on our maritime drone.

What is really wonderful with the flight controller is that we can repeat the same path later, so we can collect consistent datasets.

Step 10: Radio Control

When your maritime drone is NOT running its mission and it is still in radio range, it is a good idea to make sure the "manual" RC mode works well. Especially for leaving the port, or coming back into the port, that is particularly helpful.

Step 11: Voltage Control

Having a voltage checker is a good idea, so that you don't run out of juice in the middle of the sea, and your maritime drone can "home" safely.

Step 12: Spray Paint

Because it is likely that our maritime done would get out of sight while mapping the MPA, it is a good idea to make sure it is highly visible. Adding a flag and LED that switch on at night in case you lose control of your ocean drone is a good idea. Pick a color that has a high contrast with the environment you are working with.

Step 13: Waterproofing and Testing

We were lucky to be given access to a swimming pool for testing our boat. In the background, you can see the beautiful Sarawak government building and the "S" bridge.

We are happy with our vehicle. It is simple to build, but sails nicely. It can go straight as well as being able to rotate 360º on itself. We don't need it to be very fast as taking pictures underwater can be dark and blurry.

Step 14: Laser Quadrat

In another instructables we experimented with the concept of a laser quadrat for coral reef mapping . This version would not work underwater as all electronics are exposed (not waterproofed). This prototype was more to explain the concept. In the middle of the frame, we attached our GoPro. Every picture we take has the laser projected on them thus help us to:

  • measure remotely object
  • establish the distance to the object
  • establish the angle the object is exposed at
  • establish the transparency of the water

We'll continue to experiment with this.

Step 15: Route Planning, GPS Waypoints

Step 16: 2D Map

For generating the 2D map, you can do it manually on photoshop or illustrator. Or you can automate the work with Agisoft.

There are many other softwares out there.

Step 17: 3D Map

You can transform your pictures from 2D to 3D. We like Pix4D, Agisoft, DroneDeploy.

You can then import the photogrammetric file into a GIS package:

Step 18: Data

You can measure or isolate certain areas of an image to get data out of it. In our case we want to know

  • Bathymetry
  • Coverage
  • Specie Classified
  • Health
  • Biology (vertebrate, invertebrate)
  • Human Impact
  • Oceanographic Metadata (Temperature, Turbidity)

We recommend either Imagej or photoshop to measure area from an image

Step 19: 1.Science 2. Education 3. Informing Local Communities and 4. Influencing Policy

Once your robot is done, that you have documented it thoroughly, you can share the data you have collected. In our case, we want to promote technologies for coral reef mapping, as well as MPA monitoring. We were lucky to win the "Silver Prize Winner at the Make for the Planet at the 5th International Marine Conservation Congress". We met many interesting people that asked us hard questions, gave us new ideas, and maybe some MPAs we might go mapping!

Step 20: Thank You!

We are very grateful to Stephanie Santoso for inviting us to this event. Conservation X Lab for running the event, coordinating with all the amazing partners and sponsors . It's been an amazing experience. We learned a lot from the mentors, the judges, the delegates and the other teams! We are looking forward to:

  1. Mapping MPA - please get in touch if you have a site of interest. We are based in Hong Kong, but happy to travel.
  2. Continue to co-developing the Open Hardware (robot) and Software (navigation, image processing) - we intend to meet weekly at Makerbay in Hong Kong - you can join in person or online.
  3. Please refer grants, philanthropic groups.
  4. Educating us: give us feedback, references to study so we make this technology better

Thank you very much from the Scoutbots team: Eddie Yung, Ken and Joe Chew, Cesar Jung-Harada!!!