[mppi] Don't reset zone-based speed limits

[adivardi]

---

Basic Info

Info	Please fill out this column
Ticket(s) this addresses	-
Primary OS tested on	Ubuntu
Robotic platform tested on	Gazebo sim
Does this PR contain AI generated software?	No
Was this PR description generated by AI software?	No

---

Description of contribution in a few bullet points

Fix 2 bugs:

1. Set the reset() flag correctly to avoid resetting speed limits when other parameters are changed and after a fallback.
2. initialize the motion model in setSpeedLimit() so that the new speed limit is updated in the motion model settings. This way the predict() function is also aware of the new speed limit.

Description of documentation updates required from your changes

None

Description of how this change was tested

Tested in simulation:
Bug 1:

• Cause Controller to fail while inside a speed limit filter (e.g. with an obstacle). then check that the speed limit is still set after the controller is reset
• Changed param with reconfigure, and test the speed limit is still set

Bug 2:
Print the value of one of the constraints params inside predict(). Without this PR, it is never updated. With it, is it updated when entering/leaving a speed limit zone.

---

Future work that may be required in bullet points

None

For Maintainers:

• Check that any new parameters added are updated in docs.nav2.org
• Check that any significant change is added to the migration guide
• Check that any new features OR changes to existing behaviors are reflected in the tuning guide
• Check that any new functions have Doxygen added
• Check that any new features have test coverage
• Check that any new plugins is added to the plugins page
• If BT Node, Additionally: add to BT's XML index of nodes for groot, BT package's readme table, and BT library lists
• Should this be backported to current distributions? If so, tag with backport-*.

[SteveMacenski]

The other 2 lines I agree with

[SteveMacenski]

If the parameters were updated, that can mean that there are new maximum/minimum limits that need to be respected. If for instance the speed limits are set as 50% of 1m/s but the new max controller speed is now 0.3m/s, that would be invalid. That seems like a potential problem on this line

[adivardi]

Good point.
I can see different approaches, each could make sense:

• Run the speed limit conversion again (so in your example, new max speed will be 50% * 0.3 = 0.15m/s)
• Take the lowest between the old speed with speed limit and the new value from reconfigure without speed limit
• Ignore the speed limit and only use the new value from reconfigure

I think (1) is the most strict and consistent, but maybe a bit less intuitive (user sets a new speed, but it's then modified).
What do you think?

[SteveMacenski]

Are you having a situation where you're messing with the parameters in the middle of a mission via dynamic parameters? That feels like it might not be a wise choice if also using speed limit callbacks.

I also just looked at setSpeedLimit and I guess nothing about this technically prevents the ratio or set velocity from being larger than the limits in the controller (i.e. 110% or 100m/s). So... I suppose you're right that we could keep it as false to not mess with the current limits. I don't think I see an issue with keeping that at the end of a control sequence, even if the new limits are not reset here. that's not intuitive to me because then the new limits really wouldn't be looked at until a no speed limit set enum comes into the setSpeedLimit, but I suppose (?) that's the workflow as intended.

Messing around with both velocities in dynamic parameters and speed limits is super duper inadvisable which is what makes this subtle. Maybe the better action is to set boundaries that you can only set one or the other - so that this workflow is simpler to think about and programming logic? If a dynamic parameter comes in but s.constraints != s.base_constraints, reject the update. Then the rest of what you have now would be fine and we wouldn't have to be concerned about this in the future.

[adivardi]

No, not really. The issue I really needed to fix is the other change for the fallback.

I can also restore it to true for the parameter change so the speed limit is reset if someone change the param. Maybe this is the best, at least for manual param changes? (and not some automatic reconfigure).

[SteveMacenski]

I think the best answer is this one:

Maybe the better action is to set boundaries that you can only set one or the other - so that this workflow is simpler to think about and programming logic? If a dynamic parameter comes in but s.constraints != s.base_constraints, reject the update. Then the rest of what you have now would be fine and we wouldn't have to be concerned about this in the future.

Alternatively:

I can also restore it to true for the parameter change so the speed limit is reset if someone change the param. Maybe this is the best, at least for manual param changes? (and not some automatic reconfigure).

I'd accept this for now and kick the can down the road. I'd just want a ticket filed on this topic. It would actually be a good first issue item for a student or junior engineer to work on. If its not something you need, maybe just file this ticket to give another developer the opportunity (and not use your limited cycles on something you don't need)

[adivardi]

Ticket: #5768

and I reverted the change for the dynamic params

[adivardi]

how can I see the CI tests errors? I can't find any error logs.
All I found was

- nav2_navfn_planner.copyright include/nav2_navfn_planner/navfn.hpp
  <<< failure message
    copyright=Willow Garage, Inc. (2008), license=<unknown>
  >>>
- nav2_navfn_planner.copyright src/navfn.cpp
  <<< failure message
    copyright=Willow Garage, Inc. (2008), license=<unknown>
  >>>

[mini-1235]

how can I see the CI tests errors? I can't find any error logs.

@adivardi Can you try pulling in / rebase main? I think this is fixed in main already

[adivardi]

how can I see the CI tests errors? I can't find any error logs.

@adivardi Can you try pulling in / rebase main? I think this is fixed in main already

Done, hope it works

[adivardi]

Now the build has timeout out after 45 minutes :(

[mini-1235]

:( I don't have permission to retrigger this, let's wait for Steve

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
nav2_mppi_controller/src/optimizer.cpp	96.97% <100.00%> (+0.01%)	⬆️

... and 19 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.