crocoddyl merge requestshttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests2021-05-14T11:15:15Zhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/159WIP: Run examples as unit tests and check their logs2021-05-14T11:15:15ZGuilhem SaurelWIP: Run examples as unit tests and check their logsHi,
I'm reopenning https://gepgitlab.laas.fr/gsaurel/crocoddyl/merge_requests/2 here.
It is still a WIP because all checks are not passing.Hi,
I'm reopenning https://gepgitlab.laas.fr/gsaurel/crocoddyl/merge_requests/2 here.
It is still a WIP because all checks are not passing.https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/241WIP: Use exact equation for Vxx if regularisation is used2021-05-14T11:14:58ZWolfgang MerktWIP: Use exact equation for Vxx if regularisation is usedThis continues from an email discussion with @nmansard and @cmastall and adds the exact formula for `Vxx` if regularisation is used. At present, the exact formula for `Vx` is used when control regularisation is used, while the simplified...This continues from an email discussion with @nmansard and @cmastall and adds the exact formula for `Vxx` if regularisation is used. At present, the exact formula for `Vx` is used when control regularisation is used, while the simplified formula is used when no regularisation is applied. However, for `Vxx` only the simplified equation is implemented.
Two notes:
1) This passes unit tests for me in Python. Vxx may not be covered in Python(?)
2) The convergence in C++ did not show any difference between using the short and the exact formula (tested with `quadrupedal_gaits.py` and the walk script with unlimited iteration limit)
3) In Python, the `test_contacts` unit test fails for me on `devel`
@nmansard and @cmastall, can you please have a look whether this corresponds to our email conversation or whether I have missed something?
Many thanks,
Wolfhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/247Fixed Box-QP + removed unnecessary virtual functions in SolverDDP and SolverB...2019-11-07T09:45:38ZCarlos MastalliFixed Box-QP + removed unnecessary virtual functions in SolverDDP and SolverBoxDDPThe BoxDDP was not running in computer because we need to defined the dimension of `H_inv`.
This is a quick fix, however we need to work properly on the Box-QP and to add an abstract for QP solvers.
@wxmerkt could you review this PR?The BoxDDP was not running in computer because we need to defined the dimension of `H_inv`.
This is a quick fix, however we need to work properly on the Box-QP and to add an abstract for QP solvers.
@wxmerkt could you review this PR?Crocoddyl c++ implementationhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/240WIP: quadruped benchmark2019-10-07T14:41:42ZRohan BudhirajaWIP: quadruped benchmarkConcerns #261 Concerns #261 https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/219Delprete focchi2019-09-30T17:21:22ZMichele FocchiDelprete focchiadded friction cones in the WF
added heightMap Class
added CostModelHeightMapDistance for obstacle avoidanceadded friction cones in the WF
added heightMap Class
added CostModelHeightMapDistance for obstacle avoidancehttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/225WIP: Resolve "c++ unittest of action models"2019-09-26T13:51:55ZMaximilien NaveauWIP: Resolve "c++ unittest of action models"Closes #170Closes #170Crocoddyl c++ implementationhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/222WIP: Mpi/crocoddyl topic/kkt solver2019-09-25T17:14:55ZMaximilien NaveauWIP: Mpi/crocoddyl topic/kkt solverImplement the KKT solver
Create unit-tests around the KKT solver + the shooting problemImplement the KKT solver
Create unit-tests around the KKT solver + the shooting problemCrocoddyl c++ implementationCarlos MastalliCarlos Mastallihttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/217Topic/omp2019-09-09T12:14:45ZRohan BudhirajaTopic/omphttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/165Momentum Cost and some other minor modifications2019-09-09T11:17:02ZRohan BudhirajaMomentum Cost and some other minor modificationsThis PR:
1) Creates a Momentum Cost, with the unittest
2) Puts computation of centroidal dynamics inside DAMs
3) Fixes activation of ActivationModelInequality at the point of equality
4) Moves dependency to Pinocchio 2.1.3This PR:
1) Creates a Momentum Cost, with the unittest
2) Puts computation of centroidal dynamics inside DAMs
3) Fixes activation of ActivationModelInequality at the point of equality
4) Moves dependency to Pinocchio 2.1.3https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/214WIP: Topic/fddp2019-09-07T12:31:36ZRohan BudhirajaWIP: Topic/fddpunitttest not yet added.unitttest not yet added.https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/208WIP: impulses2019-08-29T17:04:52ZRohan BudhirajaWIP: impulsesImplementation of impulse and impact models.Implementation of impulse and impact models.Crocoddyl c++ implementationhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/170WIP: Mpi/crocoddyl topic/action unittests2019-06-28T12:31:53ZMaximilien NaveauWIP: Mpi/crocoddyl topic/action unittestsThis is what happened:
* add the action-num-diff.cpp
* implement the test-action.cpp for all current action implementation
* rename files to fit the states pattern: unycle.cpp -> action-unycle.cpp, etc.This is what happened:
* add the action-num-diff.cpp
* implement the test-action.cpp for all current action implementation
* rename files to fit the states pattern: unycle.cpp -> action-unycle.cpp, etc.Crocoddyl c++ implementationCarlos MastalliCarlos Mastallihttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/105WIP: Floating contact dynamics and impact dynamics2019-03-27T09:41:40ZRohan BudhirajaWIP: Floating contact dynamics and impact dynamicsThis PR takes care of the bug in locomotion related to the sign of the contact forces.
This PR also takes care of the #102
Please take care that this PR changes the impact dynamics. This PR now implements Mdv=JtF as the impact dynam...This PR takes care of the bug in locomotion related to the sign of the contact forces.
This PR also takes care of the #102
Please take care that this PR changes the impact dynamics. This PR now implements Mdv=JtF as the impact dynamics.
This PR also removes the (-f) notation from floating_in_contact
The changes include changes to the unittest as well.Towards python version 1.0Nicolas MansardNicolas Mansardhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/145Make cost hessian pos def for DAMLQR and AMLQR2019-03-21T09:17:20ZRohan BudhirajaMake cost hessian pos def for DAMLQR and AMLQRThis should deal with #149This should deal with #149https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/135Updating logs, fix #1442019-03-12T15:56:32ZCarlos MastalliUpdating logs, fix #144I run the different examples using the latest version of the devel branch (8005a7e8). However I observed a different behaviour for each example, is this expected @nmansard?
Most of them have lower number of iterations, except the salto ...I run the different examples using the latest version of the devel branch (8005a7e8). However I observed a different behaviour for each example, is this expected @nmansard?
Most of them have lower number of iterations, except the salto one. Indeed the solution isn't the same as before.Towards python version 1.0https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/127Remove DAMActuated from the lib2019-03-06T21:06:57ZNicolas MansardRemove DAMActuated from the libSolve #117Solve #117https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/118Minimal usage of robotwrapper2019-03-01T14:13:52ZNicolas MansardMinimal usage of robotwrapperSolve #82Solve #82Carlos MastalliCarlos Mastallihttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/116Topic/1092019-02-28T17:14:34ZNicolas MansardTopic/109https://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/110[activation] add ActivationLinearInequality defined by a(r) = Ar-bu or a(r) =...2019-02-21T16:59:40ZRohan Budhiraja[activation] add ActivationLinearInequality defined by a(r) = Ar-bu or a(r) = bl-ArIntroduce an activation given by bl<=Ar<=buIntroduce an activation given by bl<=Ar<=buTowards python version 1.0Nicolas MansardNicolas Mansardhttps://gepgitlab.laas.fr/loco-3d/crocoddyl/-/merge_requests/87Stable code run at Memmo winter school2019-02-05T21:20:30ZCarlos MastalliStable code run at Memmo winter schoolI created this PR since the current devel branch isn't working as expected (reported here #94).I created this PR since the current devel branch isn't working as expected (reported here #94).Towards python version 1.0Rohan BudhirajaRohan Budhiraja