Week 3
Serial Communication with Mark
On Monday, Mark gave us a short serial communication lecture, and how serial communication works on an Arduino Uno, when connected a laptop. From there, I learned how to connect it with MATLAB, which Annie has been able to do successfully for the CERVA project. Using the knowledge used in this mini lecture and other online resources, I was able to then connect MATLAB with an Arduino Mega, which allows for more serial communication since it has three pairs of TX and RX pornts. We hope to use this microcontroller for the following weeks to develop the CERVA project. To the right is a image of me trying to move the connections between the Arduino Uno to the Arduino Mega.
e-VaC with Xinyi
This week, I've mainly been focused on the PCB design for the e-VaC self-sensing circuit using off-shelf components available in Mouser. I hypothesize I should be done with the design by next week. Some of my challenges right now have been finding suitable parts and software migration. The literature that presented the circuit design, in Ly et al., provides only a small amount of specific parts for the board, so I'm having to look through mouser to find suitable parts that will function well with the circuit. Though it is a tedious task, it is imperative that all components work well together in order to make the best circuit possible. It's also been a challenge designing the board due to software contraints. Intiially, I was using EAGLE but I didn't like the software layout, so now I've merged the .SCH and .PCBDOC documents over to Altium, since it is better in my opinion. it will also allow us to have a 3D view of the board, which is a design component that EAGLE lacks. I'm looking forward to continuing this project and wrapping it up next week to have it be built before Xinyi leaves for her time off, July 26th. I figure I should have plenty of time by then.
Link to Bill of Materials:
Literature mentioned:
Week 2
CERVA with Annie
This week (and a little at the end of last week), I was able to successfully drive a stepper motor to displace its load by 10cm. I edited Annie's program and added functions to wait for commands to move the motor. By doing this, the program is able to control a JKM 28 Linear Stepper Motor with a step angle of 1.8 degrees. Other variables were considered, including the screw pitch, total displacement, and linear speed. The system responds to commands to move the motor forward or backward, with a kill command to stop the motor immediately. The image to the right is a view of part of the code, the function that waits for the commands to move the motor.
Code: steppercode.rtf
How it works:
- "f" to move forward 10 cm. (in relation to the motor being on the right, so it's moving left ->).
- "b" to move backward 10 cm. (in relation to the motor being on the right, so it's moving right <-).
- "x" to kill movement, but the program does not stop, it is able to detect "f" or "b" at any time.
e-VaC with Xinyi
I started the PCB design layout this week with Xinyi, since I was mainly focused on helping Annie last week. The proposed PCB design will be made in KiCad EDA. Some steps we've taken to progress the project are:
- Choosing the board layout we want (which is pictured to the right) from the literature we read, which will be modeled after the board presented in Ly et al (to the right), subject to changes.
- Choosing components to design the board, since the literature only mentioned specific part #s for the main components of the board, but not all of them.
- Devise new design considerations, such as implementing a control switch that would individually control the pouches, which would allow for different types of individual movements.
Literature mentioned:
Week 1
CERVA with Annie
This week I learned the fundamental background of the CERVA project I will be working on with Annie, and how we aim to build the prototype. We hope our project will be miniaturized later on, such as already done in various papers (Do et al., Yuan et al.), in order to fulfill clinical needs.
The images to the right are presented in Do et al, "DenseTact 2.0: Optical Tactile Sensor for Shape and Force Reconstruction", and they are the frameworks we will be working around.
Our goal is to create a tactile sensor that can visualize the deformed surface, and use a neural network to perform shape reconstruction.
How it works:
- By applying the gel with the pattern on the end of the reflective surface, we can track changes as the sensor moves around a surface using neural networks, but I'm not sure we'll get that far during this summer.
- The reflective surface with an LED light will allow us to visualize the different forces applied to the sensor, in the X, Y, and Z directions.
Literature mentioned: