graceful_controller: implement iterative selection of control points by mikeferguson · Pull Request #4795 · ros-navigation/navigation2

mikeferguson · 2024-12-13T00:30:28Z

Basic Info

Info	Please fill out this column
Ticket(s) this addresses	#4115
Primary OS tested on	Ubuntu
Robotic platform tested on	UBR-1
Does this PR contain AI generated software?	No, only real intelligence was used

Description of contribution in a few bullet points

Iteratively selects a valid target pose to use with the control law. This is done by forward simulating the control law towards the target pose and rejecting the target pose if there is a collision detected.
Adds collision checking to the "final_rotation". If this fails, we will attempt to roll out further movement via the control law (this is especially important for non-circular robots!). With this in mind, the parameter final_rotation has been renamed to prefer_final_rotation.
Modifies how final_rotation is done a bit - this is closer to how the graceful_controller did it. When prefer_final_rotation is true, we ignore the rotation of the final pose in the path and drive straight towards (and then do an in-place rotation). Previously, this code was simply using final_rotation as "do we latch and rotate", which is only half the battle if you aren't using a kinematically aware planner.

Description of documentation updates required from your changes

Replaces motion_target_dist parameter with two new parameters: min_lookahead and max_lookahead.
Renames final_rotation to prefer_final_rotation (to match graceful_controller and also indicate that this doesn't actually ensure there is a final rotation, just that the controller will prefer to generate one if it is collision free)
Replace boolean initial_rotation and double initial_rotation_min_angle with initial_rotation_tolerance which is a better name for what this parameter does (and avoids needing two parameters to do one thing). This also involved a number of tests updates to ensure that initial_rotation_tolerance cannot conflict with allow_backward.
Adds parameter v_angular_min_in_place to make sure that rotation commands actually work.
Fixed a test that was erroneously looking at twist.angular.x

Future work that may be required in bullet points

These will be done before this converts from DRAFT:

This still needs to be tested on robot (I'm simply opening for visibility and to see where we are in terms of code coverage)
~~There is a TODO around returning cmd_vel when TF fails - this seems like the wrong thing to do, need to look at it closer~~ - updated this to throw, similar to no command found
~~Review parameter max_robot_pose_search_dist - seems to be a potentially VERY large number?~~ This is consistent with other controllers in nav2
Port orientation filter (at least NavfnPlanner doesn't set orientations)

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

mikeferguson · 2024-12-13T00:50:06Z

A few other things noted when porting to UBR-1:

need to add a "min_in_place_vel_theta" equivalent - most robots can't generate arbitrarily low angular rotations
Got my answer - yeah, not all planners output headings - 100% incompatible with the NavfnPlanner - I'm going to port over the orientation filter stuff as well then.

codecov · 2024-12-13T01:03:50Z

Codecov Report

Attention: Patch coverage is 95.28302% with 5 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...v2_graceful_controller/src/graceful_controller.cpp	93.24%	5 Missing ⚠️

Files with missing lines	Coverage Δ
...e/nav2_graceful_controller/graceful_controller.hpp	`100.00% <ø> (ø)`
...ude/nav2_graceful_controller/parameter_handler.hpp	`100.00% <ø> (ø)`
nav2_graceful_controller/src/parameter_handler.cpp	`100.00% <100.00%> (ø)`
...v2_graceful_controller/src/graceful_controller.cpp	`89.69% <93.24%> (-0.82%)`	⬇️

... and 5 files with indirect coverage changes

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson · 2024-12-13T18:31:29Z

This was the updated config for testing this on the UBR-1 (moving from the version of graceful controller in my repo): mikeferguson/ubr_reloaded@a24ad85

mikeferguson · 2024-12-13T18:35:39Z

Note for posterity: I did also test "allow_backward" - and it still works

mikeferguson · 2024-12-13T18:40:35Z

Notes on things that are still different versus my controller (that maybe we'd want to do in the future):

In the control law - the slowdown_radius is a poor approximation for actual deceleration limits. Really, you want to cap the velocity based on knowing you can slow to a stop by the time you arrive at the pose.
The collision checking in the iterative loop assumes you are starting at max speed (not at your current speed). This is probably not that big of a deal, but if you're going very slowly in a tight area, the controller might wobble a bit more.
add_orientations just adds the orientations but doesn't do any more sophisticated filtering - I'm not actually convinced that additional filtering is really needed though (and, it always had a few odd edge cases - if we were to implement something here, it should really be a bit more sophisticated than what was in my controller).

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

nav2_graceful_controller/src/graceful_controller.cpp

SteveMacenski

I'll admit, this was reasonably hard for me to review with the diffs and reordering of the main code, so I'm not 100% confident in not having missed something on an initial look. I'll probably need another complete read through after changes to sanity check.

@ajtudela since this touches the docking code and I know you're thinking about these areas - want to review / test?

nav2_graceful_controller/include/nav2_graceful_controller/graceful_controller.hpp

nav2_graceful_controller/include/nav2_graceful_controller/parameter_handler.hpp

nav2_graceful_controller/src/graceful_controller.cpp

SteveMacenski · 2024-12-14T00:32:01Z

I'm curious about your thoughts from the ticket brainstorming #4115 (comment) (or is the wobbling here more or less gone with this method recomputing at the full rate?)

mikeferguson · 2024-12-14T00:40:39Z

I'm curious about your thoughts from the ticket brainstorming #4115 (comment) (or is the wobbling here more or less gone with this method recomputing at the full rate?)

I expect most of the wobbling will be gone since our motion_target can be so much farther away in most cases

mikeferguson · 2024-12-14T01:26:28Z

@ajtudela since this touches the docking code and I know you're thinking about these areas - want to review / test?

I don't think this affects the docking code at all, because that is only using the SmoothControlLaw - which is unmodified

mikeferguson · 2024-12-14T02:25:05Z

@SteveMacenski all review comments have been addressed - this is ready for another look when you get a chance

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

ajtudela · 2024-12-14T21:14:30Z

I'll admit, this was reasonably hard for me to review with the diffs and reordering of the main code, so I'm not 100% confident in not having missed something on an initial look. I'll probably need another complete read through after changes to sanity check.

@ajtudela since this touches the docking code and I know you're thinking about these areas - want to review / test?

Of course I'll test it on our robots on Monday morning.

This part doesn't change the controller of the docking but this is exactly the behaviour I need for the new following server. I'll have to figure out how to integrate this part into the server reusing most of the code.

ajtudela · 2024-12-16T18:59:57Z

I tested the improvements with the UBR-1 this morning. Although I don't really like the new order of the "steps" (rotation before movement, control, final rotation, ...) because it's harder (for me) to read, I have to admit that these changes take care of the wobbling much better than before.

Tomorrow I'll test all the combinations for the new parameters to make sure I haven't missed anything.

I recorded a little video if you want to see the changes in motion:

new_graceful.mov

mikeferguson · 2024-12-16T20:08:33Z

I have to admit that these changes take care of the wobbling much better than before.

What are you running for the k_phi/k_delta parameters with the controller? I found that the defaults in the controller right now are pretty aggressive on heading correction (which causes both wobbling, and divergence from the path - which appears to be also causing lots of replanning for you).

I went with lower values that were actually the defaults in https://github.com/mikeferguson/graceful_controller:

GracefulController:
      plugin: "nav2_graceful_controller::GracefulController"
      v_linear_min: 0.1
      v_linear_max: 1.0
      v_angular_max: 2.8
      v_angular_min_in_place: 0.6
      max_lookahead: 1.25
      initial_rotation: true
      initial_rotation_tolerance: 0.25
      prefer_final_rotation: true
      # The defaults aren't as good as this (more wiggling)
      k_phi: 2.0
      k_delta: 1.0
      beta: 0.4
      lambda: 2.0
      # This isn't quite comparable to using actual acceleration limits
      # (but works well in practice)
      slowdown_radius: 0.75

I didn't update these defaults - at first thinking it might impact some people's docking servers - but upon further review, it looks like that isn't the case (the docking server has it's own parameter set). @SteveMacenski thoughts? Should we update the defaults here?

ajtudela · 2024-12-16T20:18:34Z

Yes, they might be a bit high. I used these:

    Graceful:
      plugin: nav2_graceful_controller::GracefulController
      transform_tolerance: 0.5
      motion_target_dist: 1.2
      initial_rotation: true
      initial_rotation_min_angle: 0.75
      final_rotation: true
      allow_backward: false
      k_phi: 3.0
      k_delta: 2.0
      beta: 0.4
      lambda: 2.0
      v_linear_min: 0.1
      v_linear_max: 1.0
      v_angular_max: 5.0
      slowdown_radius: 0.5

But I agree with you. I had to lower k_phi/k_delta for the docking server. I'd try with these settings tomorrow.

mikeferguson · 2024-12-16T20:26:01Z

I had to lower k_phi/k_delta for the docking server

I think I might have actually been the one who chose those higher values for the docking server initially - that is an application where you want the robot to come at the charge dock as straight-on as possible, so the values make sense for docking. I think they are too high for general navigation controlling (but it wasn't something I caught back when we originally added the controller here)

ajtudela · 2024-12-17T13:39:55Z

I made a new video this morning. Lowered k_phi/k_delta reduces the wobble more.

new_graceful_02.mov

mikeferguson · 2024-12-17T17:46:42Z

@SteveMacenski bumping this - wasn't sure if you maybe missed my comment that this was ready for another review

SteveMacenski

I found that the defaults in the controller right now are pretty aggressive on heading correction (which causes both wobbling, and divergence from the path - which appears to be also causing lots of replanning for you).

but upon further review, it looks like that isn't the case (the docking server has it's own parameter set). @SteveMacenski thoughts? Should we update the defaults here?

I think I might have actually been the one who chose those higher values for the docking server initially - that is an application where you want the robot to come at the charge dock as straight-on as possible, so the values make sense for docking. I think they are too high for general navigation controlling (but it wasn't something I caught back when we originally added the controller here)

We can change those, definitely! I want all defaults to have a really nice out of the box experience to showcase algorithm's values without the need of heavy tuning for 'most' applications.

nav2_graceful_controller/src/parameter_handler.cpp

nav2_graceful_controller/src/graceful_controller.cpp

* update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

SteveMacenski · 2024-12-19T15:37:54Z

@ajtudela any thoughts here? This is your work too, so I want to make sure you have a chance to review and give thoughts!

ajtudela · 2024-12-19T19:52:18Z

I tested heavily this week and it works pretty well, even in narrow corridors.

I think there's missing coverage in L212-216 (or I'm ready Codecov wrong).

Otherwise, LGTM.

mikeferguson · 2024-12-20T19:00:06Z

@SteveMacenski what else do we need to get this merged?

SteveMacenski · 2024-12-20T19:37:53Z

@mikeferguson a couple of small updates in the docs still needed, but code is good -- thanks for the revisions and help! ros-navigation/docs.nav2.org#624

…os-navigation#4795) * initial pass at iterative Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add v_angular_min_in_place Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add orientation filter, fix remaining TODOs Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * try to increase coverage, fixup minor test issues Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * address review comments Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * review comments * update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * revert change in default Signed-off-by: Michael Ferguson <mfergs7@gmail.com> --------- Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

…4795) * initial pass at iterative Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add v_angular_min_in_place Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add orientation filter, fix remaining TODOs Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * try to increase coverage, fixup minor test issues Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * address review comments Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * review comments * update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * revert change in default Signed-off-by: Michael Ferguson <mfergs7@gmail.com> --------- Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

* nav2_smac_planner: handle corner case where start and goal are on the same cell (#4793) * nav2_smac_planner: handle corner case where start and goal are on the same cell This case was already properly handled in the smac_planner_2d, but it was still leading to an A* backtrace failure in the smac_planner_hybrid and smac_planner_lattice. Let's harmonize the handling of this case. This commit fixes issue #4792. Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> * nav2_smac_planner: use goal orientation when path is made of one point Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> * nav2_smac_planner: publish raw path also when start and goal are on the same cell Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> * nav2_smac_planner: add corner case to unit tests Add a plan where the start and goal are placed on the same cell. Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> --------- Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> * creating auto-transition option for nav2_util::LifecycleNode (#4804) Signed-off-by: Steve Macenski <stevenmacenski@gmail.com> * Fix trajectory generation bug in docking controller (#4807) * Fix trajectory in docking controller Signed-off-by: Alberto Tudela <ajtudela@gmail.com> * Ceil and remove resolution Signed-off-by: Alberto Tudela <ajtudela@gmail.com> * Update nav2_docking/opennav_docking/src/controller.cpp Co-authored-by: Steve Macenski <stevenmacenski@gmail.com> Signed-off-by: Alberto Tudela <ajtudela@gmail.com> * Update nav2_docking/opennav_docking/src/controller.cpp Co-authored-by: Steve Macenski <stevenmacenski@gmail.com> Signed-off-by: Alberto Tudela <ajtudela@gmail.com> --------- Signed-off-by: Alberto Tudela <ajtudela@gmail.com> Co-authored-by: Steve Macenski <stevenmacenski@gmail.com> * graceful_controller: implement iterative selection of control points (#4795) * initial pass at iterative Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add v_angular_min_in_place Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add orientation filter, fix remaining TODOs Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * try to increase coverage, fixup minor test issues Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * address review comments Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * review comments * update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * revert change in default Signed-off-by: Michael Ferguson <mfergs7@gmail.com> --------- Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * fix bug in use of v_angular_min_in_place (#4813) Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * publish motion target as pose (#4839) Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * nav2_behaviors: drive_on_heading: return TIMEOUT error code when exceeding time allowance (#4836) Until now, the NONE error code was returned when exceeding the time allowance. Let's return the more appropriate TIMEOUT error code instead. Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> * fix bug in orientation filtering (#4840) * fix bug in orientation filtering some global planners output all zeros for orientation, however the plan is in the global frame. when transforming to the local frame, the orientation is no longer zero. Instead of comparing to zero, we simply check if all the orientations in the middle of the plan are equal Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * account for floating point error Signed-off-by: Michael Ferguson <mfergs7@gmail.com> --------- Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * Adapt GoalUpdater to update goals as well (#4771) * Add IsStoppedBTNode Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * add topic name + reformat Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix comment Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix abs Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * remove log Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * add getter functions for raw twist Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * remove unused code Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * use odomsmoother Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix formatting Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * update groot Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Add test Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * reset at success Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * FIX velocity_threshold_ Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Fix stopped Node Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Add tests to odometry_utils Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix linting Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Adapt goalUpdater to modify goals as well Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix formatting Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fixes Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * change name of msg Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Make input goals be Goals again for compatibility Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Revert "fix" This reverts commit 8303cdc. Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * refactoring Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * ament Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * ignore if no timestamps Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * facepalm Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * update groot nodes Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * Use PoseStampedArray Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fixes Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * use geometry_msgs Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix import Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * use geometry_msgs Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * more fixes Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * . Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * revert Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * . Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * add common_interfaces Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * use PoseStampedArray Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * reformat Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * revert Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * revert Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix warn msg Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix test Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * improve Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix format Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * change to info Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> --------- Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> * fix simple smoother failing during final approach (#4817) * new test case for end of path approach Signed-off-by: rayferric <63957587+rayferric@users.noreply.github.com> * modify tests to match the more permissive smoother policy Signed-off-by: rayferric <63957587+rayferric@users.noreply.github.com> * implement steve's suggestions Signed-off-by: rayferric <63957587+rayferric@users.noreply.github.com> --------- Signed-off-by: rayferric <63957587+rayferric@users.noreply.github.com> * Add acc limit consideration to avoid overshooting in RotationShimController (#4864) * Add acc limit consideration to avoid overshoot in RotationShimController Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Add acceleration limit tests to RotationShimController Signed-off-by: RBT22 <rozgonyibalint@gmail.com> --------- Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Fix flaky ness of opennav_docking test_docking_server (#4878) (#4879) Call publish() (odom -> base_link tf) at startup to kick things off and spin 10 times(1 second) TF, so that it has a chance to propogate to the docking_server so that it will accept an action request. Previously it was only spinning once, hoping the timer would fire and call publish fast enough for it to propogate to the docking_server so that it is able to accept the first 'dock_robot' action request Signed-off-by: Mike Wake <macwake@gmail.com> * [BtActionNode] [BtServiceNode] clear between calls (#4887) Signed-off-by: Guillaume Doisy <guillaume@dexory.com> Co-authored-by: Guillaume Doisy <guillaume@dexory.com> * dwb_critics flaky test - lineCost coordinates must be within costmap (#4889) (#4884) There is no protection/checks in the pathway from lineCost to costmap_2d::getIndex(mx, my) for grid coordinates that exceed the of bounds of the allocated costmap. (presumably for speed) This test was triggering an off by one error attempting to read the the 2500 byte costmap at byte 2503 costmap size 50x50. getIndex(3, 50) = my * size_x_ + mx; = 50 * 50 + 3; = 2503 Signed-off-by: Mike Wake <macwake@gmail.com> * Add option to use open-loop control with Rotation Shim (#4880) * Initial implementation Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * replace feedback param with closed_loop Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Reset last_angular_vel_ in activate method Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Add closed_loop parameter to dynamicParametersCallback Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Add tests Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Override reset function Signed-off-by: RBT22 <rozgonyibalint@gmail.com> --------- Signed-off-by: RBT22 <rozgonyibalint@gmail.com> * Fix unstable test in nav2 util (#4894) * Fix unstable test in nav2 util Signed-off-by: mini-1235 <mauricepurnawan@gmail.com> * Fix linting Signed-off-by: mini-1235 <mauricepurnawan@gmail.com> * Change 5s to 1s Signed-off-by: mini-1235 <mauricepurnawan@gmail.com> --------- Signed-off-by: mini-1235 <mauricepurnawan@gmail.com> * Update bt2img syntax and bt pics (#4900) Signed-off-by: Maurice-1235 <mauricepurnawan@gmail.com> * bumping to 1.3.5 Signed-off-by: Steve Macenski <stevenmacenski@gmail.com> * Revert "Adapt GoalUpdater to update goals as well (#4771)" This reverts commit 55d7387. --------- Signed-off-by: Dylan De Coeyer <dylan.decoeyer@quimesis.be> Signed-off-by: Steve Macenski <stevenmacenski@gmail.com> Signed-off-by: Alberto Tudela <ajtudela@gmail.com> Signed-off-by: Michael Ferguson <mfergs7@gmail.com> Signed-off-by: Tony Najjar <tony.najjar.1997@gmail.com> Signed-off-by: rayferric <63957587+rayferric@users.noreply.github.com> Signed-off-by: RBT22 <rozgonyibalint@gmail.com> Signed-off-by: Mike Wake <macwake@gmail.com> Signed-off-by: Guillaume Doisy <guillaume@dexory.com> Signed-off-by: mini-1235 <mauricepurnawan@gmail.com> Signed-off-by: Maurice-1235 <mauricepurnawan@gmail.com> Co-authored-by: DylanDeCoeyer-Quimesis <102609575+DylanDeCoeyer-Quimesis@users.noreply.github.com> Co-authored-by: Alberto Tudela <ajtudela@gmail.com> Co-authored-by: Michael Ferguson <mfergs7@gmail.com> Co-authored-by: Tony Najjar <tony.najjar.1997@gmail.com> Co-authored-by: Ray Ferric <63957587+rayferric@users.noreply.github.com> Co-authored-by: Balint Rozgonyi <43723477+RBT22@users.noreply.github.com> Co-authored-by: ewak <ewak@users.noreply.github.com> Co-authored-by: Guillaume Doisy <doisyg@users.noreply.github.com> Co-authored-by: Guillaume Doisy <guillaume@dexory.com> Co-authored-by: mini-1235 <58433591+mini-1235@users.noreply.github.com>

…os-navigation#4795) * initial pass at iterative Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add v_angular_min_in_place Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * add orientation filter, fix remaining TODOs Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * try to increase coverage, fixup minor test issues Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * address review comments Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * review comments * update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com> * revert change in default Signed-off-by: Michael Ferguson <mfergs7@gmail.com> --------- Signed-off-by: Michael Ferguson <mfergs7@gmail.com> Signed-off-by: stevedanomodolor <stevedan.o.omodolor@gmail.com>

mikeferguson mentioned this pull request Dec 13, 2024

update docs for iterative graceful controller ros-navigation/docs.nav2.org#624

Merged

mikeferguson force-pushed the graceful_4115 branch from fb7c29e to 76c3de6 Compare December 13, 2024 17:50

mikeferguson added 2 commits December 13, 2024 12:51

initial pass at iterative

938785c

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

add v_angular_min_in_place

d9ec380

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson force-pushed the graceful_4115 branch from 76c3de6 to d9ec380 Compare December 13, 2024 17:51

add orientation filter, fix remaining TODOs

125a073

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson marked this pull request as ready for review December 13, 2024 18:22

try to increase coverage, fixup minor test issues

6c8cbfe

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson commented Dec 13, 2024

View reviewed changes

nav2_graceful_controller/src/graceful_controller.cpp Outdated Show resolved Hide resolved

SteveMacenski requested changes Dec 14, 2024

View reviewed changes

mikeferguson force-pushed the graceful_4115 branch from 14ee149 to 81a9983 Compare December 14, 2024 02:23

address review comments

11c4b15

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson force-pushed the graceful_4115 branch from 81a9983 to 11c4b15 Compare December 14, 2024 02:27

SteveMacenski reviewed Dec 17, 2024

View reviewed changes

review comments

c35f252

* update defaults * rename to in_place_collision_resolution Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

mikeferguson force-pushed the graceful_4115 branch from f3e9500 to c35f252 Compare December 17, 2024 20:40

revert change in default

e6a90a5

Signed-off-by: Michael Ferguson <mfergs7@gmail.com>

SteveMacenski approved these changes Dec 20, 2024

View reviewed changes

SteveMacenski merged commit 7b73a38 into ros-navigation:main Dec 20, 2024

mikeferguson mentioned this pull request Dec 20, 2024

fix bug in use of v_angular_min_in_place #4813

Merged

7 tasks

mikeferguson deleted the graceful_4115 branch January 18, 2025 01:32

Conversation

mikeferguson commented Dec 13, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Basic Info

Description of contribution in a few bullet points

Description of documentation updates required from your changes

Future work that may be required in bullet points

For Maintainers:

Uh oh!

mikeferguson commented Dec 13, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov bot commented Dec 13, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

mikeferguson commented Dec 13, 2024

Uh oh!

mikeferguson commented Dec 13, 2024

Uh oh!

mikeferguson commented Dec 13, 2024

Uh oh!

Uh oh!

SteveMacenski left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

SteveMacenski commented Dec 14, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mikeferguson commented Dec 14, 2024

Uh oh!

mikeferguson commented Dec 14, 2024

Uh oh!

mikeferguson commented Dec 14, 2024

Uh oh!

ajtudela commented Dec 14, 2024

Uh oh!

ajtudela commented Dec 16, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mikeferguson commented Dec 16, 2024

Uh oh!

ajtudela commented Dec 16, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mikeferguson commented Dec 16, 2024

Uh oh!

ajtudela commented Dec 17, 2024

Uh oh!

mikeferguson commented Dec 17, 2024

Uh oh!

SteveMacenski left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

SteveMacenski commented Dec 19, 2024

Uh oh!

ajtudela commented Dec 19, 2024

Uh oh!

mikeferguson commented Dec 20, 2024

Uh oh!

SteveMacenski commented Dec 20, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

mikeferguson commented Dec 13, 2024 •

edited

Loading

mikeferguson commented Dec 13, 2024 •

edited

Loading

codecov bot commented Dec 13, 2024 •

edited

Loading

SteveMacenski commented Dec 14, 2024 •

edited

Loading

ajtudela commented Dec 16, 2024 •

edited

Loading

ajtudela commented Dec 16, 2024 •

edited

Loading