Skip to content

Calvin Rioux's Notebook

Jump to bottom Edit New page
Hayden Cook edited this page Aug 20, 2026 · 1 revision

Spring 2026

  • Still on the ros/helics team, I believe our team consists of me, Sabina, Elliot, Esha, Shrayes, and Rick this year
  • We are looking to put everything together and deliver a decent product this semester
  • No new members this semester so no one to catch up on the project
  • Subteam meeting will be on Friday at 11 am.

1/15 - 1/29

  • Kickoff presentation, edited a few slides, deleted some unnecessary ones
  • discussed things to work on with Sabina, decided to look into combining Safiy's Nix environment with the current Docker image to get everything in one image.
  • Kickoff presentation went well
  • Friday meeting we just went over our goals for the week.

1/29 - 2/5

image * not even anything I changed so not sure what is going on. computer is at 100% memory image

2/5 - 2/19

  • found a problem with the docker image where the wrong numpy version (2.4) is being used for the entire image and cannot be changed because of permission problems, while the correct version 1.26 is also installed but is not usable (this is my understanding)
  • sabina suggested some things:

Things to try: move software directory from / to ~ (from os to home directory) and update all install paths in bashrc for helics and girdlabd make sure venv is activated try python 3.10 without venv for global instance

  • I will first look at editing the dockerfile to ensure the correct versions of everything are installed across the image, using a requirements.txt file

  • this would help if some installation installs and upgrades the numpy version as a dependency

  • if this doesnt work i will look into moving the directory and install paths since it does seem to be in the wrong spot

  • docker images just take so long to build on my computer so editing a building a new dockerfile is incredibly inefficient, I will see if I can edit the current image

  • Alright nevermind I made a new docker image with a force reinstall of numpy 1.26.3 at the end and it seemed to get rid of the 2.4 problem, activated venv, installed matplotlib with sudo /software/venv/bin/pip install matplotlib

*now i have this: image

  • after killing helics processes with pkill -9 helics, it now just gets stuck?
image
  • running bash run_example.sh does bring up the gui so that part seems to be working

2/19 - 2/29

  • Ended up making a new Dockerfile with a requirements.txt file to ensure the right versions of everything was installed.
  • Elliot also made a new setup so I will be using that going forwards instead of mine
  • Nguyen managed to get a demo of mission 1 running, I also got it to run by following the steps from discord and his notebook
  • Midterm Presentation went well, I only did some minor editing of fault injection slides and removed some old ones that had been presented multiple times
  • talked to Will about new things to look at, he mentioned Kalman Filtering to put together lidar data, imu, and quadrature encoder data.
  • SLAM toolbox has it built in but I think there are still things to look at.

2/29 - 3/5

3/5 - 3/12

3/12 - 3-19

  • looked through pages for hardware used to get these covariance estimates: BNO055 Covariances:
  • Accel variance: (0.015)² = 2.25e-4 m²/s⁴
  • Gyro variance: (0.00244)² = 5.95e-6 rad²/s²
  • Orientation variance: (2° × π/180)² = (0.0349)² = 1.22e-3 rad²

4/2 - 4/9

image image

4/9 - 4/23

4/23 - 4/30

  • setting up for final presentation, not much to report since I mostly did research but I made a slide
image * still working on putting everything together, unable to get slam connected so there aren't any visual results for right now.

Fall 2025

August - 9/4

  • Unfortunately have a time conflict this semester and will not be able to make the subteam meetings
  • I am on the navigation and fault detection team with Nguyen, Esha, Natalie, Rick, and Sam
  • I am interested in the fault detection side of things so my goal will be to look through Safiy's work and see if I can get some kind of output in the fault detection that can connect to nav2 and ros
  • Edited slides for opening presentation, mostly wrote the ones for fault detection, our goals, and how it connects to our overall project goals.
  • Expanded and edited Safiy's slides mostly
  • Recorded slides since I have a time conflict

9/4 - 9/11

  • Hear presentation went well, good starting point for us

  • Looking through Joseph's documentation of the standardized vm, mine is a bit different right now but as long as it works I don't feel a large need to make the shift over, will always be in mind if I run into issues

  • Looking through fault detection documentation, realize I have never used GridLabD, will look into it

  • Okay nevermind since I need GridLabD I will just use the full cps vm we have.

image
  • off to a good start

9/11 - 9/18

  • fixed problem by just uninstall and reinstalling everything, restarting computer, and getting some commands from stack overflow guy
  • all good now, hopefully this fixes issues I ran into running everything directly on ubuntu on my computer

9/18 - 9/25

  • realized this is old version, tried to get the updated one with everything and failed
image
  • not sure whats going on, for now I might continue using it just on ubuntu on my computer since it worked last year, might try again later
  • install Gridlab-D, not really sure what to make of this but will look
image

9/25 - 10/2

  • paper due, reviewed whole paper to make sure we're all on the same page, made some minor edits
  • looking at some gridlabd tutorials, not sure how much I want to get into it, would rather just familiar before connecting it to other parts
image * currently some issues, will review others' notebooks to see if I should do this a different way

10/2 - 10/9

  • midterm presentation, made some minor changes but not too much to report on, once again recorded voice

10/9 - 10/23

  • running a gridlab-d demo, getting ''"C:\GridLAB-D 5.3.0"\gridlabd.exe' is not recognized as an internal or external command, operable program or batch file," likely just need to have my environment set up correctly.

  • "ERROR [INIT] : demo_full.glm looks like a version 1.x GLM files, please convert this file to new style before loading demo_full.glm(1): load failed at or near 'end of line...' FATAL [INIT] : shutdown after command line rejected"

  • demo_full.glm(14): syntax error at 'start_time "202...' demo_full.glm(14): expected clock block closing } demo_full.glm(12): load failed at or near 'clock {...' FATAL [INIT] : shutdown after command line rejected

  • issue persists, it seems when I get rid of one another one pops up, I will scratch this one and try again with a different one another time

10/23 - 10/30

  • will try looking through safiy's notebook to see what works
  • pivot to checking out the navigation side of things
image * seems my environment is not working as it once did, will troubleshoot * having some trouble with the vm again not sure what I did last time, probably should've recorded it better image * consider going back to ubuntu install since that worked in the past image * kinda works also kinda crashed everything. I'm not sure why everything I used before is breaking. I will have to restart a lot of things

10/30 - 11/6

  • Thinking I can use the built in nav2 waypoint follower tool along with detected faults from gridlab-d in helics to put things together, so looking at demos for that trying to learn more about it.
  • I gave it an attempt and not much was gained, couldn't really get anything to work and was frustrating.
  • Decided to watch this video on it to make better progress
image

11/6 - 11/13

  • ended up just reinstalling nav2 and saving some files, really not sure what the issue was.
  • realize I lost my last attempt at the waypoint follower, but I was probably better off starting fresh anyways
  • got all the nodes running properly
image * seems to be some issue with with amcl node (adaptive monte carlo localization) not publishing the map properly? * ran a different simulation to avoid that issue, now having some issues with my launch files, not sure everything is set up the way it needs to be, it's a bit hard to troubleshoot image

11/13 - 11/20

  • sourced a new turtlebot to use for now I think this fixes the issue
  • ros2 action send_goal /follow_waypoints nav2_msgs/action/FollowWaypoints "{ poses: [ { pose: { position: {x: 1.0, y: 0.0}, orientation: {w: 1.0} }, header: { frame_id: 'map' } }, { pose: { position: {x: 2.0, y: 1.0}, orientation: {w: 1.0} }, header: { frame_id: 'map' } }, { pose: { position: {x: 3.0, y: 0.0}, orientation: {w: 1.0} }, header: { frame_id: 'map' } } ] }" manually sends a bunch of waypoints
  • next steps: somehow connect this to gridlab d with some existing demo, I really need to brush up more on gridlab-d to be able to do this
  • maybe could just try to add using waypoints with anyhting other than just typing it into the cmd

11/20 -

  • Final paper, mostly went through and made sure everything connected well and made sense, did not have anything too significant or new sections to make.

Spring 2025

1/9 - 1/23

  • Mostly setup, joined discord and github, went over the goals of the project
  • Looked through general plans for this semester
  • Joined the ros helics subteam

1/16 - 1/23

  • Attended first subgroup meeting
  • Looked through last semester's presentation and plan for this semester
  • Joined the physical environment subtask team

1/23 - 1/30

  • Went through the first part of the ROS tutorial
  • Using the Humble version which I think everyone used last semester although there is a newer version out, Jazzy
  • Turtle for tutorial, rqt, nodes, rest of the CLI tools section

  • Installing Gazebo, version Fortress is apparently compatible with ROS Humble

  • Following tutorial below to make a robot in gazebo

  • https://gazebosim.org/docs/harmonic/building_robot/

  • I am not 100% sure I am getting the right versions of everything, it is a bit confusing

  • Tried a few different installs (below), seemed to work but I got a black window and the weird windows blue screen sound when I run it

https://gazebosim.org/docs/fortress/install_ubuntu/#binary-installation-on-ubuntu

https://gazebosim.org/docs/all/ros_installation/

  • Troubleshooting, ran it with Ogre 1 instead of Ogre 2 and it kind of works but the screen is constantly flashing, so not really usable.

  • ok it doesn't work at all anymore

  • Reinstalling and troubleshooting, could be problem with running out of ram mid install

1/30 - 2/6

  • Trying install again, this time through Docker, using versions Jazzy and Harmonic, and using XLaunch for the visual (although I'm not sure it'll actually be used)

  • Works, although with docker I have to reinstall gazebo every time so I'm not sure what everyone else is doing

  • Try setting up custom image

  • End up reopening same image which has Gazebo installed and keeps everything

2/6 - 2/13

  • Continuing to work on tutorials for ros and gazebo, building the robot
  • Bit annoying editing files in image, surely a better way to do it

2/13 - 2/20

  • using ubuntu 24.04 now since nav2 is easier to use on there
  • trying to get what other members did on my system, Joseph's models

2/20 - 2/27

  • doing tutorials for nav2, goal is to eventually help robot navigate through substation environment

2/27 - 3/6

  • busy week, did a little more tutorials

3/6 - 3/13

3/13 - 3/20

  • spring break, did not do much, aim to use nav2 plugin in substation environment and perhaps write my own

3/20 - 3/27

  • further explore tutorials, mostly looking at navigating physical environment
  • right now just with turtlebot
  • many topics, not sure which are relevant/best to dive deeper into

3/27 - 4/3

  • in meeting, talked with Will about using existing team sdf and bot files instead of tutorial built in ones
  • had to find correct file which we did (nav2_bringup located in opt/ros/jazzy/share)

4/3 - 4/10

  • computer and simple things just like file explorer running abnormally slow, hopefully not due to things from this project
  • trying to put turtle bot into the substation world
  • gazebo and substation file especially run very badly on laptop
  • attempting to make a new launch.py file with the substation sdf so I can use nav2 with it
  • so when I run "ros2 launch nav2_bringup substation_sim_launch2.py headless:=False" it does open the substation in gazebo with the turtlebot and the nav2 view in rviz, but the nav2 view is the one corresponding to the tb3 simulation so something there will have to change, but having the bot in the substation is a win I suppose
    • the substation is very very laggy and I can barely move around and find the turtlebot, much less try to make it move, so i may try more things with other models before modeling more in the substation
* now looking at using slam to map the substation but I think I have to look into change some things in my launch file to run properly. (tried running this ros2 launch slam_toolbox substation_sim_launch.py use_sim_time:=true), which just returns the default nav2 rviz hexagon with dots thingy

4/10 - 4/17

  • meeting - talked with Nguyen about using slam, specifically with existing substation sdf

  • very laggy, removing several things from sdf in hopes of running better

  • aim to run slam, will have to look into changing launch file

  • try headless gazebo

  • on own, attempting to get nav2 demo to work with substation sdf, ran into issue " TF_OLD_DATA ignoring data from the past for frame left_gripper at time"

  • had multiple running things at once so restarted everything, now have new issue " Detected jump back in time. Clearing TF buffer."

  • restarted computer and fixed

  • was previously attempting to use moving_robot model from a previous tutorial to run slam on the substation world, now trying to use turtlebot3 with some issues

  • looking through the specified launch files in the turtlebot's directory for issues

  • added a slide to presentation detailing simulated environment and plans for future

  • presentation went will, next semester should be full of good progress

TODO starting next semester:

  • get slam working on the substation environment * get moving robot working with nav2, lidar sensor * probably rewrite some launch files * coordinate with navigation subteam in getting this to work
Add a custom footer

Home

Active Notebooks
Archive
Tutorials

Clone this wiki locally