#robotics Logs

Dec 23 2019

#robotics Calendar

02:04 AM BitEvil is now known as SpeedEvil
04:38 AM fling_ is now known as fling
04:06 PM chisight: is there opensource gyro based room mapping software that is like off-the-shelf ready to go? something for a small wheeled bot. wheel sensors are probably not hard to add if needed. bumpers or other proximity sensors are assumed.
04:38 PM deshipu: chisight: I would check ROS for such things
04:46 PM chisight: deshipu: thanks for the pointer. looks to have kinda what i want.
06:50 PM logb0t_ is now known as logb0t
08:49 PM rue_mohr: hmm
09:22 PM rxF: trying to map a room using only gyro information is going to give you a really bad estimate due to integration error
09:24 PM rxF: I think a lot of SLAM techniques combine wheel encoder data with some more concrete environmental sampling (sonar, lidar, vision, ...)
09:25 PM rxF: (the latter also provide obstacle avoidance)]
09:44 PM mrdata__ is now known as mrdata-
11:55 PM chisight: lidar is cost prohibitive. ultrasonic is actually something i was thinking about as additional position verification but not seeing any multi-emitter solutions in ROS yet. haven't gotten all that far, my wife got home today...
11:57 PM chisight: also thought about invisible ink 2d barcodes and uv or ir light to read them. scatter them around to get corrections on the other sensors.
11:58 PM chisight: vision seems to be computationally expensive, but i'm again just getting started so maybe not as big a deal as i think it might be.