search

BCLab-UNM / SC2

Swarmathon Team Code for the NASA Space Challenge 2 Competition
MIT License
2 stars 0 forks source link

Updated Wheel encoders for Round #203

Open gmfricke opened 3 years ago

gmfricke commented 3 years ago

Measured the wheel radius and wheel base and updated the wheel encoder node to match. Noetic displays a timestamp warning for every message that is not later than the last message received. This makes the repl unusable. Added a minimum delta time between publishing to avoid that problem. Still need to figure out the minimum delta. This issue of spamming warnings is under active discussion on the devs github.

Carter90 commented 3 years ago

Numbers look good per https://gitlab.com/scheducation/srcp2-final-public/-/wikis/3.-Simulation/5.-Dimensions-and-Specifications

Carter90 commented 3 years ago

I'm concerned about the timing in the driver it is making decisions at 10hz and I've considered moving that up to 100hz so the odom only being every second will cause issues. might need to agree on timing for the robot with constraints limited by the sensors

gmfricke commented 3 years ago

Thanks for verifying that. I did it by eye, so I’m glad it wasn’t too far off.

On May 5, 2021, at 1:26 PM, Carter Frost @.***> wrote:

Numbers look good per https://gitlab.com/scheducation/srcp2-final-public/-/wikis/3.-Simulation/5.-Dimensions-and-Specifications

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

gmfricke commented 3 years ago

Agreed - we can set the encoder to update at any rate you want.

Sent from my iPad

On May 5, 2021, at 1:41 PM, Carter Frost @.***> wrote:



I'm concerned about the timing in the driver it is making decisions at 10hz and I've considered moving that up to 100hz so the odom only being every second will cause issues. might need to agree on timing for the robot with constraints limited by the sensors

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/BCLab-UNM/SC2/pull/203#issuecomment-832958179, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AA6DSEYEROREVEO5GLIY2I3TMGNPNANCNFSM44FSPSFQ.

Carter90 commented 3 years ago

Agreed - we can set the encoder to update at any rate you want.

Lets try 10hz for this PR and see how it goes

gmfricke commented 3 years ago

Uncomment wheel_encoders in SC2/src/scoot/launch/scoot.launch

abbypribis commented 3 years ago

How should I test this pull request?

gmfricke commented 3 years ago

Hi Abby,

Load up the simulation and the REPL. Then “rosrun localisation wheel_encoders” (.py maybe). Echo the output of //odom and see how the x, y position changes as you drive around. It will get increasingly inaccurate but shouldn’t be horrible to start with. I like to move the robot to 0,0 in the gazebo global frame so it is easy to compare to the odom frame output which will also start at 0,0.

All the best,

Matthew

On Jun 4, 2021, at 11:08 AM, Abby Pribisova @.***> wrote:

How should I test this pull request?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

abbypribis commented 3 years ago

Everything runs well on my end - I don't have a good intuition on how well the odometry should perform. Once Carter's recommendations are implemented, I can approve.