Choosing our Sensors

While designing the second version of RoboGoby we needed to decide what sensors were necessary to have incorporated into the basic design. If you have been following our blog you'll remember when we posted this in October 2013. That post had ideas for sensor we had gotten from different research institutions. After spending the past year and a half working on the sub, we created an updated list of sensors we want to have incorporated on the most basic version of RoboGoby. We decide that these environmental sensors are key to a well functioning submersible and are necessary if we want to have a well working product.

The updated list of sensors we chose is below. There is also a short description with each saying why it is important for our design. While choosing the new sensors we also made sure that they all interface with an I2C bus. This allows us to communicate with multiple different sensors while only using a minimal amount of wiring and only a single micro-controller. 

Breakout board w/ MS5803
We chose to use this sensor because it is waterproof and it can withstand up to 14 Bars of pressure, which is around 200 psi or 450 feet underwater. It allows us to determine the depth of the submersible.

We chose to use current sensors on this version of RoboGoby because we wanted the ability to have current control over the thrusters. As are unable to accurately measure the speed of the ROV in all directions, we will be using current to limit the thrust on each of the motors. This will make it easier to implement autonomous capabilities in the submersible. We are using the ADC in order to use multiple analog current sensors without having to use more analog pins. 

This version of RoboGoby has a watertight compartment. Although this design is useful in many ways, we need to make sure that the compartment gets neither wet or too hot.

We chose to use a LIDAR instead of a ultrasonic sensor for ranging underwater. This sensor is very easy to integrate in an underwater environment and has an extremely large range. We will have to calibrate this for the reflectivity of water, but this sensor works for bathymetry, meaning it should also work for our purpose. Its had two advantages over a sonar sensor. The first is that it doesn't have a minimum range (a very small one), which is important if we end up having to navigate without a camera or with limited vision. Second, it is easier to waterproof as we don't have to find a substance which provides waterproofing without having any sound dampening properties. 

This sensor is very similar to the Razor 9DOF IMU we used last year. The only difference between the two sensors is that this one doesn't have a micro-controller built in. Similar to last year, we will be using this sensor to get a better understanding of the submersible's position in the water. This will not only be useful for human drivers, but also useful in implementing autonomous station-keeping and driving. 


Choosing a Camera – Update

The search for a camera to use in our robot has been long and involved. Throughout the process, we have worked toward finding a camera that satisfied a basic set of requirements. Our camera needs to be capable of outputting high quality video, interfacing with C++ or Java through the Beaglebone black, and be capable of stereoscopic imaging. We believe we have finally found a camera that can satisfy all of those requirements: the DUO MLX.
This camera, while expensive, boasts an impressive array of features and capabilities. It stereoscopic and infrared, allowing it to judge depth in low light environments. It comes equipped with a small array of programmable infrared LEDs and an on-board 9-DOF freedom IMU, allowing it to sense its orientation with a high degree of accuracy. Finally, it comes with a SDK which, just recently, allows it to interface with ARM based micro-controllers like the Beaglebone - although access to this capability requires subscription to DUO's developer program. This SDK is compatible with C++ and C, as well as Java through the Java Native Interface. Another distinct advantage of this camera is its form factor. It is very small, being only 52mm long. This will allow us to fit many more design features into the front portion of the submersible.


Small Thrusters v2.0 – CAD Model

We recently finished the final CAD model for the small thrusters we're using in our submersible. Four of these thrusters, two vertical and two horizontal, helps our submersible attain 4 of our submersibles 5 degrees of freedom.

The design of the small thrusters was focused around designs that were both easy to build and easy to install. We decided to work off of last years design, keeping the same idea, but tweaking it to fit in the new form factor of the sub. Each thruster assembly has only 5 total parts (excluding the motor and prop) making it an easy part to build and assemble. An image of the finished thruster is below:

Thruster Assembly

This design consists of four main parts: the thruster grill, motor mount, PVC mount, and the PVC piping. In the image above the PVC is transparent, the grills are black and both the motor and PVC mounts are green. The thruster is assembled by sliding the PVC mount into the three slots cut in the PVC pipe and then epoxying it place. The motor and motor mount are then able to slide freely into the six slots cut on the opposite side of the PVC pipe. This side is not epoxied and allows us to easily remove the motor by removing three bolts on either side of the thruster and sliding out the PVC pipe, PCV mount, and motor mount all together.

The bolts we are using are 16mm long, 4M bolts that are threaded into tapped Delrin plastic (green). These bolts mostly have to deal with torque and therefore are suitable for the size and power of motor and prop combination we're using. This is also nice because it means the thruster is mounted directly to the aluminum body and the PVC acts only to direct the flow of water. Three of the parts are pictured below:

Thruster Mount
PVC Mount

PVC Pipe

Finally we created the grill for our thrusters. The main purpose of these pieces are to keep any large pieces of material from interfering with the performance of our thrusters. They do, however, also help with the aesthetics of the sub. By using these pieces to slide into the holes cut in the aluminum body we are able to keep a relatively smooth form factor which is important in reducing the drag of the sub. There are also three bolts on either side in order to mount the thruster firmly in the sub.


Both the vertical and horizontal thrusters will be mounted close together and as far from the center of the sub as possible. This gives us finer control over the submersible due to mechanical advantage. An image of the two thrusters mounted in the aluminum body is below:

Mounted Thrusters


Rear Thruster v2.0 – CAD Model

We recently finished the CAD model for our large rear thruster. The thruster from our first version actually worked pretty well, so we kept many of the elements from it and changed only what was necessary.

We started by coming up with a cone to guide water-flow from the rear edge of the submersible to the propeller. The first idea was to have the end near the submersible be square (to make it look like a continuous piece) and have only the end near the propeller be a circle.
Square to Circle Thruster Cone

This shape had two advantages over a circle to circle cone. First, it is more streamline with the rest of the submersible - this helps some with the water flow but is mostly for visual appeal. Second, it would allow us to use the flat part of the surface to bolt fins onto for the thruster nozzle. We would probably have used this design if it did not require remaking the nozzle.

Instead we switched to a circle to circle cone so that we could use the same nozzle as last year.

Circle to Circle Thruster Cone

By using a circular cone and last years nozzle we are able to cut down on the number of pieces we need. The fins that connect the cone to the nozzle are 1/8" plastic and will be epoxied into the grooves that can be found in each piece. While the cone itself is circular we did leave a square part on the larger end allowing for an easy connection with the square submersible.

Cone and Nozzle

Once we had the external portion completed we moved on to the inner parts. First cutting a hole for the shaft and motor so that they can be set into the rear cone and not take any space from the submersible. We then created a mounting bracket for the large thruster motor which would suspend it inside of the thruster cone.

Thruster Cutout
Motor and Bracket

When all of the pieces are put together the result is a pretty weird looking, but highly functional piece of our submersible. For information about the coloring see the general CAD post.


UPDATE (7/13/2015): A New Cone Design 

In our first attempt at creating the cone piece we tried to cut a large block of plastic on a 3-axis router – it did not work well for a couple reasons. Thus we decided to 3D print the piece instead. Although we were skeptical at first, this method turned out to work well. The design however, was changed in a few ways to make printing easier. 

1) The first change we made actually didn't have anything to do with printing: we decided to return the the square to circle lofted cone. This is because it gives us a much better form factor and looks better when attached to the body of the sub. The only down side is that there are three fins going into a four sided object, which just means that we have to make a custom fin for each slot as they are not the same due to a lack of symmetry.
2) Another important change we made is the use of threaded inserts (discussed in this post). Without these inserts a printed cone would not work with out current mounting design because threads do not hold well in printed plastics. 
3) A final change came in the form of moving the mounting bolts. While this may not seem related to printing the cone it is a change that has only come about because we can make multiple versions easier and cheaper using this rapid prototyping method. 

We decided that we want to make as many of the external bolts in one line so that we have the smallest impact on our coefficient of drag. This type of modularity also allows us to mix and match the different sections of the submersible. We decided to move moved the bolts so that all four are on the sides, rather than having one on each of the four sides.

Below is a picture of the final design:


RoboGoby II – General CAD

After deciding to continue with Project RoboGoby this summer we decided that having an in-depth CAD, or Computer Aided Design, of the submersible was the best way to plan Version II.  This post, which is about the initial CAD of sub,  allowed us to flush out all of our ideas and not worry about specifics.  This gives us the freedom of designing the submersible how we want to and then thinking more about specifics later on.  The second round of CAD (which will be done for each of the major pieces) will focus on the specifics before machining parts. 

As part of the initial design phase we broke the submersible up into three distinct pieces — the front section, the middle, and the rear section. Each of the sections are important to the overall design yet still describe the general ideas for the finished sub. Below each of the sections are quickly explained and an images of each section is provided.

We also used the coloring scheme in CAD to color different parts in our design to easily tell what material each piece is going to be made out of. This key will help you in understanding all of our CAD designs for RoboGoby II. 

Printed PLA ----------> Black
Delrin -----------------> Green
PVC -------------------> White
Aluminum--------------> Grey
Polyethylene -----------> Pink
Tapped Holes --------> Yellow
Paint----------------------> Red


UPDATE (7/13/2015): Threaded Inserts

We will also be using RED to represent the holes we are using threaded inserts in. The threaded inserts look like this: 

6010S12-16SS 1
6mm Threaded Insert

And the CAD'ed plastic holes will look like this:

It is designed to be pressed into wood/thermoplastics and is perfect for using in common 3D printed plastics to make use of bolts as printer plastic is notoriously hard to tap. We will be using these in most if not all of the future 3D printed pieces.



The front section houses our camera, lights, sensors, and two of the side thrusters. This section was therefore one of the more complicated of the sections to design as it includes many different components. Below is a transparent image of the front section of the submersible. You can see the horizontal and vertical thrusters which are placed behind the variable pitch camera and lighting system.

Front Section


The middle section of the submersible houses most of our electronics and all of our electronic controllers. This section has been designed to be waterproof in pressure exceeding 150  PSI. Much of the complicated work in this section will be done when mounting our electronics and connectors on the inside. For now the main part of this section is the double o-ring plug you can see in the transparent part of the image below. This plug ensures that everything stays waterproofed while still giving is easy access to the electronics inside.

Middle Section


The rear section is almost completely thrusters (in the future it will have a power connector). Currently it contains side thrusters symmetric to those in the front (vertical and horizontal), but also has our large rear thruster, cone, and nozzle which are extremely important to the submersible. 

Rear Section

Stay tuned for future posts with more detail explanations about the design of the components of each of the sections.