Timeline for Trouble interpreting joint velocity profiles
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 15 at 12:31 | comment | added | Chuck♦ | What tells the motors to move, and what tells them how fast to move? Is this just an on/off control where the trapezoidal profile is set by a combination of the velocity and acceleration scaling factors, or is there some speed profile issued by a controller? Whatever the case, something must be telling the motors to move, so can you plot that motion command (on/off, or some other command speed profile) along with the speed feedback? | |
| Jul 15 at 9:12 | comment | added | Ten | Could you please clarify what you mean by command speed profiles? I'm giving my robot a MaxVelocityScalingFactor() and a MaxAccelerationScalingFactor() and plot the resulting joint speed profiles via plotjuggler. Could Motoros2 have a built in sway control that is conflicting with the Moveit trajectory execution? | |
| Jul 11 at 12:44 | comment | added | Chuck♦ | Welcome to Robotics, Ten. Do you have plots which also show the command speed profiles? I've done a lot of work on commercial sway control systems and the speed profiles here, with the "knee" on the way up and down, looks like what I would expect with a sway control algorithm. This leaves me wondering if what you're seeing is a flaw with a standard trapezoidal speed profile or if there's some other controller. Apart from the rising transient, the first plot looks good. Images 2 and 3 show acceleration changing during what looks like a pretty smooth transient. | |
| Jul 11 at 11:54 | history | edited | Ten | CC BY-SA 4.0 | deleted 486 characters in body |
| S Jul 11 at 9:15 | review | First questions | |||
| Jul 25 at 9:20 | |||||
| S Jul 11 at 9:15 | history | asked | Ten | CC BY-SA 4.0 |