Lock Box Log

thumb Weekly log for Locked Box project

Yuanxiao Gao, Samuel Hoff

Week of September 8

We met with Mo, our TA to discuss how we should proceed with our project.

We determined that we are going to 3-D print a box that will be equipped with a camera, raspberry pi, and a locking mechanism. The purpose of this is to unlock or remain locked based on the state of the user’s eyes. Since it is very difficult to detect a true blink, we will use a three-stage passcode. The way this passcode will be input is by flashing an LED and then taking a picture of the user’s face three times. Each time the picture will be assessed by the raspberry pi and, depending on if the state of the eyes in each state matched the current passcode, the box will unlock or not. We also met with Dr. Feher and discussed what the progression of this project is likely to look like. First, we are to figure out facial detection (if we fail at that we will find a more reasonable form of user input). Once this is done, we will upload this code to a raspberry pi and get it to compile there. After this, we will make the box and wire the LEDs and relay to the raspberry pi so that everything works together properly.

Week of September 15

This week, we met to look into the feasibility of us being able to actually implement some form of facial detection library and we deemed it too difficult, especially considering the other tasks we would have to complete along with facial detection. We are likely to pursue a less complicated method. This is likely to include light sensors and an input consisting of blocking those sensors in a pattern. One idea that we are considering is to have some sort of circular dial with a hole in one side of it to allow light through. This hole would only allow light to shine on one sensor at a time and the dial would rotate, like a combination lock. The user would twist it in a combination that is the existing pass code and the box would unlock if the code is correct. This code would be programmable by the user and stored until the user wants to change it.

Week of September 22

File:Blink UNO123

light blinks faster and faster

This week in class, we learned about Arduino. And we have decided to make a locked box instead of blink box and we met up, talked about how should the locked box be and came up with its basic model. We will have a (four input) code that the user can unlock the box with and reset. Then we would also have a backup code that is much longer (maybe eight inputs) that can be used to reset the temporary combination in case it has been forgotten. We would also have a button that the user presses to indicate what they want to do: one press for entering the short combination to unlock the box, 2 presses to reset the combination (after entering the correct combination first) and third, holding down the button to reset the shorter combination by using the longer backup combination. We will start wring some code this week.

Week of September 29

This week we finalized the objectives of the project, figured out the budget and all of the parts we need, updated our wiki, and planned our progress with a Gantt chart (Both 1.5 hours - Sam did Gantt and budget with parts, Jerry did Wiki). What we will be doing is making a box that has a dial that the user turns to enter a combination that will unlock the box. The position of the dial will now be found by using Hall effect sensors to sense the presence of the magnet located in the dial. We determined that the hall effect sensors are the best option for this situation because it is likely to be fairly difficult to have a freely moving dial with an LED, which requires wiring. Also, it is possible that an LED could burn out or something could go wrong with the connection, since it is a moving part. We met with Mo, our TA and made some changes to the Wiki project page and the log page and also discussed details of progression of the project (Sam 1 hr). We ordered the parts needed to get started (Sam 30 min). We also made our presentation and rehearsed it for this upcoming Friday (Both 1.5 hr). We also looked into using a Hall effect sensor (Jerry 1 hour). We started writing pseudo code for the Arduino (Sam 1 Hour).