Timeline
-
January 2015
All starts from a problem
Everything started with a question. A problem occurred to an acquaintance, with the streams; he drifted and lost the boat once he surfaced. No sight of it, the waves were too high… So he asked “Can you imagine a system which would allow me finding the boat during my decompression stops?”
-
June 2015
Research Phase
We decided to get together and start this project as soon as possible. For the school to accept it, we had to do research and to prove the project is theoretically doable. We found a solution, did the mathematical study and submitted it to our professor.
-
September 2015
Starting the Project
We started working on the project in September 2015. We started with an Arduino Uno and all the other devices (i.e. TFT screen, sensors, etc.) to develop a first prototype. Unfortunately, we quickly realised that the Uno wasn’t powerful enough and did not meet our requirements. We decided to switch to an Arduino M0 Pro. Based on an Atmel 32 bits microcontroller, it is much more powerful and futureproof. We mainly learned how to use this new device, to set it up and use the Graphics’ library. We also completed the mathematical study and wrote the Heading Determination code.
-
June 2016
Autorization to continue in 5th year
There are no mandatory school projects in the final year however; there is the possibility to replace a course by a project. The school allowed us to pursue our project which they find promising! However, Clement Fontaine will not be part of our project anymore, as he is not specialising in Embedded Systems.
-
July 2016
Back to work
A lot has been done in July! Schematics, PCB design are complete. The GUI and the programme for the diver are almost ready as well. We still need to do the FSK tests to validate this part of the project.
-
September 2016
Test of Ultrasonic FSK (reported)
The first version of the PCB has been manufactured mid-September. We faced some issues. Finally, our calculation method seemed impossible to achieve. After some reflexion days, an all new way to achieve it has been found, more precise, easier to set up and much less CPU intensive. With some trigonometry and with storing our values in a table in E2PROM instead of calculate them, it allows us to constantly refresh our GUI and make it smoother. Without a FPU, it is the simplest method to process data quickly.
-
October 2016
Test of Ultrasonic FSK
Test have been done ! FSK is okay. But another huge problem came to us. The way to get signals before processing them is impossible to achieve because of ambiant noise undersea between 20KHz and 100KHz. Some locations in south of France are polluated by companies that produces noises in the water. After some researches, a new method have been found, nearly insensible to noise. We will not get the different times of reception anymore, but we will work their phases, out of each PLL with a phase comparator. Three phase comparators to determine the differents distances between each mic depending of the angle of arrival of the signal.
-
November 2016
Test of mesurement with Phase Detector and 3 microphones
As soon as phase detectors will be working, tests with full circuit will begin. Mics + AB Class Amp + PLL + Phase Detector + µC + Screen
-
KickStarter
in
2017/2018 ?
