🏴

From Noisebridge
Jump to navigation Jump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Hack teh Cabañas here https://community.comma.ai/cabana/?demo=1337

"Cabana" is a tool for deciphering CAN data, which is what cars use for talking to themselves.


Get a 🐼 here: https://shop.comma.ai/products/panda-obd-ii-dongle

Build a 🐼 here: DIY Panda




🏴


"app notes"

some features ONLY work in chrome, and there are some issues on linux with chromium

  • Fix double scroll bars
  • Add better grabbers for bit selection
  • Make segment selection clearer in plot
  • Show multi-signal overlay


dumpster dive data

Messages sorted by highest event count, with a descriptive'ish name and unique ID.


acura_ilx_2016_can.dbc

Identified bit that looks like some kind of watchdog or interupt trigger when vehicle is moving above min speed.

ID     Signal                Note
1:400  SMOOTH_CRUISE_PING    bit 20 some kind of flag/watchdog triggers at regular interval when moving above min speed
       X_COUNT_UP_10_SEC     also contains 10 second incremental count in last byte.


hack'n around

You can use the save log button to get a csv file of the data, this only works with chrome and some versions of chromium.

The following converts the csv file to a sqlite3 database file name 'data.db', with a table named 'can' containing all the records.

import sqlite3
import pandas

conn = sqlite3.connect('data.db')

df = pandas.read_csv('data.csv')
df.to_sql('can', conn, if_exists='append', index=False)

# get and show 100 samples of wheel speed data with a time over 100
cur = conn.cursor()
cur.execute("SELECT * FROM can WHERE addr = 464 AND time > 100 LIMIT 100")
rows = cur.fetchall()
rows

conn.close()