A very low cost AUV based on development platform for learners
Sayantan Roy Arko, Shemonto Das1 and Elham Robbani1Adil Hossain1
1
BRAC University, Bangladesh
INTRODUCTION on Li-ion battery which does not leave any
This abstract focuses on a very low cost AUV waste in water in any ways.
design based on development platform for
learners and testing in extreme conditions.
Autonomous AUVs are more costly and not
that much suitable for student research Figure 2. Data monitoring interface.
purpose. Especially in the 3rd world countries
students can't afford buying one. As a result of CONNECTIVITY and IMAGING
not getting access to underwater vehicle The system is connected to a wired
development platform students are not getting connectivity based on simpler interfaces. The
more interested in the field and they remain system is based on raspberry pi ssh
behind from discovering all the great resources connectivity. The rover can be on the go
under water. To address this, our prime target programmed and monitored. For the imaging
is to make a low cost AUV that has features we kept a raspberry based optimized video
almost close to industrial AUVs for stream transfer and on board image processing
development purposes. Creating an to guide the whole system. Any underwater
environment to experiment for the students the object detection is possible through open
main purpose of this AUV will be in the field source detection platform or our pre built
of educational research and testing in extreme Tensorflow based detection
conditions. We tested mechanical structures
based on lower cost builds from scratch
reusable plastic and acrylic so that it is easy to
replace and cost effective. Figure 3. System Block Diagram
METHODOLOGY
After deploying into water for safety issue a
Figure 1. Circuit Stack (from left to right: 3D sensor inside the hub tube to ensure that if
Model and Actual Outcome) there is any leakage or not. After ensuring the
hub is fully sealed motors will start to
SYSTEM OVERVIEW calibrate. After calibration rover will start to
In this research we focused on testing all the take sensors reading to reach preprogrammed
cost effective alternatives for underwater target depth which is set by the end user. Then
technologies. Our system includes cheap 3D other tasks will be executed according to user
printed thruster cases with Brushless DC needs and editable program, for which we
motor setup. These motors were made from have also built an easy to use interface and
scratch. And the propellers were also sourced instruction table
from easy to find cooling fans and modified
for underwater efficiency. The rover body
structures was made from recyclable plastic
pipes. The interface system was based on
development platforms so that it is
programmable and ready to deploy in different Figure4. System flowchart
scenario. The depth sensor, compass and
accelerometer used was also easy to source IMPLEMENTATION
models and further engineered to work under
water. The system resides in a waterproof In order to achieve the goal of this study, we
container/tube which has been made from low have designed, build and successfully tested a
cost acrylic structure and tested in extreme prototype in several marine bodies. The water
conditions. The power system was also based proof cavity that we have built was tested in
different water bodies and it was completely
waterproof. Moreover, it can withstand a good
amount of water pressure which makes the
mechanical structure well sustainable and
operable up to a depth of 5-7 meters under
water. We tested our AUV with both wired CONCLUSION
and wireless connection. For wired In this research a low cost and affordable
communication we used cat6 cable and for AUV is built which can be used by students,
wireless communication we have used a single researchers or even hobbyists to explore
antenna System that is expandable up to 7-8 underwater. With this AUV underwater data
meters from the maximum depth our AUV is can be easily collected and stored for further
intended to achieve. We have received analysis. The device developed in this research
efficient data from the sensors, especially from is focused mainly to make it affordable and
our customized depth sensor. As our target is easy for use by students and researchers. To
to come up with a low cost AUV so we have conclude, this AUV will be an alternative to
reversed engineered local Barometric Pressure high cost industrial AUV’s. For learners
Sensors to depth sensor by covering it with a
shield of balloon. The feedback that we Figure 6. Prototype testing
received from it was much satisfactory.
Moreover, the reading that our compass References:
provided was also accurate that helped the [https://ieeexplore.ieee.org/document/830
positioning of our rover perfectly. The 9444]
combined data of the low cost sensors that [https://ieeexplore.ieee.org/document/968
were integrated in our system was satisfactory. 762]
[https://ieeexplore.ieee.org/document/423
Figure 5. Measurement diagram of 1129]
CanSub [https://ieeexplore.ieee.org/document/791
2954]
