608_finalproj

USEFUL STUFF:

project proposal: https://docs.google.com/document/d/1MmS5aJmTf7h9zif0waNb0SbHrCHV8-rOJm8-NlFZWDQ/edit milestones: https://docs.google.com/spreadsheets/d/1dAaHvU9l2HsWnL9xeRB0oCax8zR0Zb-WK1GE9FGS7pE/edit#gid=0

ADNS 9800 Sensor: https://github.com/mrjohnk/ADNS-9800 //useful github code https://www.tindie.com/products/jkicklighter/adns-9800-laser-motion-sensor/ //useful information about sensor https://datasheet.octopart.com/ADNS-9800-Avago-datasheet-10666463.pdf //the datasheet https://github.com/pepijndevos/Dwergmuis https://github.com/joshuajnoble/ArduinoADNS-9800.git //arduino code Circuitry: MOT- not plugged in SS- RXD- 1O3 DG- GND AG - GND MO- 1O23 SC- IO18 MI- IO19 VI- 3v3

General Workflow:

1. PULL BEFORE YOU START WORKING!!
2. Message out the files you'll be working on and the changes you'll make. This way, others know not to work on the same file / section and we'll minimize merge conflicts
3. COMMIT OFTEN!!! and please write descriptive commit statements in case someone has to pick up where you left off. Good commit statements means its easier to follow your train of thought.
4. PUSH WHEN YOU'RE DONE. Push when you've completed something. ie try not to push half-finished broken code. But don't hoard the code you've completed either.
5. COMMUNICATE OFTEN. Message out when you've pushed especially if it's a big / relevant change so that everyone else knows to pull.

UPDATES

5/10/18

1. Added "shake to reset" functionality to .ino how it works:

• need to shake at least three times (both ways) to register as a reset.
• you can see the exact threshold numbers on the script, defined under label /// SHAKE TO RESET FXN ///
• it sends a DELETE_request (new function defined at the bottom) with inputs image_Id, num_entries, and color
• at the moment, num_entries is hardcoded to 1000 (delete 1000 most recent entries) but subject to change
• after deleting, the oled displays "RESET" for three seconds and returns to the "INSTRUCTIONS" state

5/4/18 1.integrated the sensor into the 608_finalproj.ino 2. tested it on the server and it works! 3. TODO: work on the accuracy of the drawings.

5/3/18

1. added a class that returns the delta x and delta y readings from sensor
2. TODO: covert these to position values and scale them for the canvas

5/1/2018

tldr; the state of esp + server ultimately has not changed. However, the server has more capabilities that are functional provided that the ESP code changes to use these capabilities.

1. Server code is functional provided that its given proper get and post requests. the javascript essentially "picks up the pen" when colors have changed.
2. POST requests have changed. Need to specify color.
3. Added Artist class. Purpose of this class is to encapsulate all image related attributes, specifically The image name, the color the artist is using to draw, and the actual data points of the image. Purpose of ths class is to also give us the option to NOT hard-code things such as colors.
4. ESP code has been updated to format correct posts.
5. TODOS:
   • integrate delta x and y values and read these values in the ESP code
   • add states to the ESP: a. one to allow the user to input which image he would like to draw on b. allow the user to pick his color

---

1. Added ESP code that uses the ADNS 9800 sensor and returns delta x and delta y values
2. TODOS:
3. integrate the delta x and y values
4. replace these values with the imu values in the 608_finalproj.ino code
5. Hardware: create better mechanism for drawing

4/27/ 18

1. ESP code uses autoincrementing image ID
2. server code now keys into request with form.
3. TODO: integrate sensor

---

1. The request functions take in the x_coords, y_coords, entry_id
2. The server will display a picture based on the coordinates
3. I created an autoincrement id, but I did not add it to the post request-- should this id replace kerberos?
4. TODOS: -unhardcode the kerberos in the server code-- work on implementing autoincrement id -We should also start hacking out the sensors 4/27/18:
5. ESP POST_request function is defined
   • two new relevant global variables are defined at the top, under label /// POST REQUESTS ///
   • IMPORTANT: you need to change these variables for your local code
6. Major adjustments are made to the modifyReading function
   • now has only one input, reading: float value to be scaled
   • all other inputs are hardcoded at the top, under label /// CANVAS DIMENSIONS ///
   • the function now returns an int (instead of void) because the float pointer is not happy about changing float to int
   • simplified some math based on the assumption that variable lowerbound (left/bottom bound of canvas) is always 0
   • (old function is commented out at the bottom; will delete later)
7. uses of modifyReading and POST_request are included inside loop under case DRAW
8. TODOs:
   • clean up docstring on modifyReading after confirming new fxn with team
   • server needs GET request implemented so the drawing shows up
   • (optional) update modifyReading to enable separate scaling for x and y coords, in case canvas is not square
   • (optional) currently, arrays _xCoords and _yCoords used by ImageCoords is float-based; change to int if necessary
   • package pretty print, mod, and modifyReading into a c++ class

4/20/18:

1. ESP TODOs: post instead of serial print (you can also keep the serial prints for testing purposes) Sample actual sensor data instead of using dummy value Expand upon the states. Keep in mind ESP is currently using ONE button tho we have two.
2. Server TODOs: Finish request_handler so that it properly unpacks request Add more GET options. ie return a list of all image IDs, make the number of images to display variable, etc etc
3. Overall TODOs: Incorporate a z coordinate so that we know when the pen should draw a line, or be lifted and moved for example

4/19/18:

1. Button class is done.
2. ImageCoords class has been made. It stores the last k (x,y) coordinates, where k is user defined. ImageCoords also conveniently returns its stored values as a comma separated string.
3. Rough ESP32 loop is completed. Currently data is being added to our ImageCoords object. TODO format the data into a post request instead of just serial printing it
4. Rough server is done. It currently returns the html and js that displays an image based on variable lists of x and y coordinates.