#robotics Logs
Jan 11 2019
#robotics Calendar
07:18 AM SleepyTurtle: hi! I have a question regarding robot locomotion and slipping
07:22 AM SleepyTurtle: We have to control a 4 wheeled robot made up of 2 differential drives, both can rotate wrt the chasis
07:22 AM SleepyTurtle: apparently this is a common configuration for AGVs
07:22 AM SleepyTurtle: According to a coworker, we should only actuate the front 2 wheels so that we can use the rear ones to check if the robot is slipping
07:27 AM SleepyTurtle: it kind of makes sense, but it seems like a huge waste
07:28 AM SleepyTurtle: and having a free-turning base in the rear seems like a bad idea
07:30 AM deshipu: SleepyTurtle: actuating all 4 wheels is tricky
07:35 AM SleepyTurtle: deshipu: why would you say that?
07:48 AM deshipu: because it's true
07:50 AM SleepyTurtle: What are the major problems that can arise?
07:51 AM SleepyTurtle: Is it trickier than having a freely rotating base with two non steerable wheels in the back?
07:52 AM deshipu: with just 2 wheels, the speeds of the wheels are independent, and you can control them to control the vehicle
07:53 AM deshipu: with 4 wheels, the speeds of all the wheels depend on each other, and you have to solve some tricky equations to figure out which wheel should have what speed
07:54 AM deshipu: that's assuming no slipping
07:54 AM veverak: "some tricky quetions" ? those are basic no? :/
07:57 AM deshipu: I suppose they are basic to you
07:58 AM SleepyTurtle: Are you referring to rolling constraints?
08:00 AM SleepyTurtle: I'm sorry, I should have asked this earlier, I have to go now. Thanks you for your replies, bye!
08:14 AM veverak: deshipu: thought that that may be a reason
08:14 AM veverak: oh fod
08:15 AM veverak: *god
08:15 AM veverak: I am slowly converting to the mindset of our professors
08:15 AM veverak: "well, students, thit is simple!" ... rolls out dozens of differential equations on board without breaking a sweat
08:26 AM deshipu: yeah
08:26 AM deshipu: I wish I paid attention to that at the uni
08:28 AM veverak: oh, speaking of differential equations I did not
08:29 AM * veverak payed attention at graphs, algorithms, data strucutres, algo complexity, provabillity of correctnes, automatons...
08:29 AM deshipu: then you try to say "it's not rocket science" and stop, because you realize this is exactly the same thing as you do in rocket science
08:30 AM veverak: lol
08:30 AM veverak: :D
11:28 AM deshipu: veverak: https://www.youtube.com/watch?v=SZvQtlh6ueM
11:29 AM deshipu: veverak: remember how we said they don't explore hybrid wheel/legs very well?
02:25 PM deshipu: well, they do have a prototype as well
02:25 PM deshipu: https://www.youtube.com/watch?v=z4yQyWf5PeY
02:26 PM deshipu: considerably underwhelming, though
03:02 PM veverak: cute one!