-
Notifications
You must be signed in to change notification settings - Fork 101
Feature/fst footstep constraint #719
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
stefanfasano
commented
Mar 12, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This class already existed and has been tested in ihmc-internal. I am just moving it (and its corresponding unit test) to open-robotics. Let me know if there is another project/package within open-robotics that would be a better home for this class
LuigiPenco93
approved these changes
Mar 13, 2025
Beomyeong
approved these changes
Mar 17, 2025
LuigiPenco93
added a commit
that referenced
this pull request
May 15, 2025
* setup for tufts demo * fixed transform lever * manual addition scne nodes support * ai2rNode functionality with check points * fix list growing to infinity * typo * started to expand streaming * added status of completed behavior * comment * p-control direction * added goto/navigation behavior * fixed bug, multiple failures not resetting * collision checking in goto action node * adding toolbox for footstep streaming * separating foot step streaming in its own toolbox, wip * pushling input from UI to toolbox * completed rework into toolbox * typo * cleanup simulation ui nadia * active step adjustment with mnaual mode * added direction and stride estimation to toolbox * added parameter for stride length scaling * simplified stride length estimation * A method to predict the footstep from the average velocity of the tracker is added. This method is going on work, not yet finished. * fixed bugs crash and commented code for testing * fixed bugs in FST and fixing KST state initialization message * fixed bugs in KSt and fixing KST logic for init message * readd configuration command * clean up logs and parameters * fixed bugs in footstep placer * step direction estimate working really nice * method for stride estimate * stride estimate tuned * typos * automatic registration tracker * load tracker role at init VR * reassign role to tracker when disconnected and reconnected * trackers get the model of vr headset * force one step at a time in estimate * working on yaw estimate * added yaw estimate * fix merging * added constraints for computed step * Fixing reactive update foot step position while robot doesn't finish its swing. * fixing calmping distance * minor changes * revert multi contact ik controller changes * fixed yaw estimate * debugging multiple unwanted steps at the end * fixed timing steps * stride default scaling factor * added footstep landing heuristic * fixed and tuned footstep estimate, working nicely * use heigthmap to snap footstep center * minor edit * height map fix * better vr teleport location * clean up kinmeaticsStreamingMode and KST parameters * height map added * bug footstep adaptation * minor fix * added arm manager IK * fixed bug last estimate sent multiple times * work in progress chest tracker * arm streaming completed * setting up footstep optimization * setting up optimization of footstep * setting up test for footstep optimization * disbale heightmap * fixed merge * params * added check for different options * added check for intiial solution validity * tuning gradient descent, added better intiialization and gradient delta * gradient descent version * simulated annealing * global optimization * increased alhpa for testing * setting up kernel for optimization * prepared cu file and java CUDA loader * propogate gpu function into placer and streaming * releasing gpu memory * Feature/fst footstep constraint (#719) * added step position limiter based on reachability constraints * manual and VR placed footstep constrain to reachable region and snap to height map * constraint future footsteps based on future stance foot and CoM positions * constraint frame for VR and manual footsep constraint aligns with pelvis yaw * did some renaming * cleaned things up, added safety checks, moved step limiter params * made step limiter robust to footstep plans without alternating footstep sides * Revert "Auxiliary commit to revert individual files from 1a87926" This reverts commit e5d0e89555cdbb0ae29efdb2e43f595078cc4cff. * fixed kernels * fixed a bunch of bugs in the kernels * fixed kernels and CUDA runner * bug fix kernel solution retreived correctly * optimized cuda logic for footstep optimization * cleaning cuda code, but getting a crash at successive compute calls * fixed cuda address violation error * minor edits * disabled swing speedup * fst streamign param , test on robot * increased search radius and set yaw range more properly * more aggressive stepping and secodn footstep for consecutive stepping * added print for capturability status * increasing transfer * fixed bugs kst reinit * dynamic transfer time * adding com shift * Clear com xy control before walking * Add yoTransferDuration * removed unencessary code, hacks * removed com shift * testing arm streaming * added yovariables for logging fst * ocu chnages * Feature/body collision using heightmap (#747) * made a bodycollisionfilter and its test class * added FrameCapsule3D * Passing the kinematic collisions to the Rapid Height Map Manager * added functionality * Got u v calculation working * added CUDA Kernel * added CUDA Kernel to manager * fixed the kernel file * fixed depth input * fixed pointer * removed torso and pelvis * fixed size of pointer * made a bodycollisionfilter and its test class * added FrameCapsule3D * Passing the kinematic collisions to the Rapid Height Map Manager * added functionality * Got u v calculation working * added CUDA Kernel * added CUDA Kernel to manager * fixed the kernel file * merged develop + added optimization step * deleted old files * formatted and addressed a very irritating Nicks comments * formatted and addressed a very irritating Nicks final comments * added to height map * removed cylinder * added sphere functionality and tolerance * Tune parameter to avoid cell at sensor origin * Remove check that wouldn't allow for UI button to be pressed when robot is paused walking * Cleanup --------- Co-authored-by: nkitchel <[email protected]> Co-authored-by: Tomasz Bialek <[email protected]> * reduced map * parameters for successful test on robot * back to working state * readded preview footstep and swing duration * reduced footstep margin * testing changes * param heightmap fst * index bindings * working well rough terrain * bug fix typo * transfer update beomyeong fixes * Revert "transfer update beomyeong fixes" This reverts commit 1e72eff. * undo init transfer change * temporary initial transfer * init transder = default trasnfer * transfer update beomyeong fixes * clean up pr * readded missing things from develop * tomasz clean up cuda file * Feature/fst stable footstep data message changes (#809) * changed fields in FootstepDataMessage so they are not specific to QFP. VR FST can use these to continuously update footstep reference * removed updating footstep reference from transfer state * changed some logic for toggling off swing trajectory damping * VR FST footstep message is set to continuously update footstep reference * tracker roles * fixed transform for index trackers * key beindings * clean up reviews * Update ihmc-perception/src/test/java/us/ihmc/perception/cuda/CUDAFootstepOptimizerTest.java Co-authored-by: Tomasz Bialek <[email protected]> * disable test footstep optimizer * remove unecessary stuff * fixed gradle issue --------- Co-authored-by: lpenco <[email protected]> Co-authored-by: bpark <[email protected]> Co-authored-by: ihmc-rosie <[email protected]> Co-authored-by: Stefan <[email protected]> Co-authored-by: Stephen McCrory <[email protected]> Co-authored-by: Nehar Poddar <[email protected]> Co-authored-by: nkitchel <[email protected]> Co-authored-by: Tomasz Bialek <[email protected]> Co-authored-by: Stefan Fasano <[email protected]> Co-authored-by: LuigiPenco <[email protected]> Co-authored-by: Tomasz Bialek <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Addition of elliptical kinematic reachability constraint for manual and VR placed footsteps