The timestep, the solver tolerances, and impratio span several orders of
magnitude, where a linear +/- step is useless. Add ImGui_LogStepper, whose
buttons walk a "well-tempered" logarithmic ladder of round numbers that
closes each decade on a power of ten (see WellTemperedStep). It lays out
like ImGui's own InputScalar stepper -- field, then the - / + buttons, then
the label.
Apply it to the timestep, Tolerance, the LS/Noslip/CCD/Sleep tolerances,
and Imp Ratio. Strictly-positive quantities for which zero is meaningful
(the tolerances) take a zero_below floor: the ladder's bottom rung is 0, so
stepping down snaps to zero and stepping up off zero returns to the floor.
The timestep and impratio, which never want zero, leave it unset.
PiperOrigin-RevId: 965888812
Change-Id: I823e60913ecb2be2cade96baa5bf7de9e2334d5a
https://youtu.be/17XpwnqyCXs
New transmission type mjTRN_SO3: a relative orientation, targeting a ball
joint or a site+refsite pair. It is the first transmission with more than
one force output: its length is the norm of the expmap vector of the
relative rotation and its moment axes are the 3 rows of the
relative rotational Jacobian, without projecting onto per-actuator gears.
New force law mjGAIN_SO3/mjBIAS_SO3: a geodesic PD servo, force =
kp * log(q_current^-1 * q_target) - kv * velocity, exact for arbitrary axis
combinations with a unique equilibrium at every commanded orientation.
Error, moment rows and velocity all live in the child frame (joint or
site): the right-difference error is the gradient of the geodesic
potential in that frame. The parent-frame (left) error is not: driving
child-frame torques with it pumps energy at large angles, settling into
steady-spinning limit cycles (the SO3LargeAngleConvergence test). The
integrator variant stores the 3D orientation setpoint in act (actnum = 3,
re-anchored to a bounded representative at integration time). Exposed in
MJCF as <orientation joint=|site=+refsite= kp kv|dampratio>, or via
<general gaintype="so3" biastype="so3">.
The setpoint input has two charts: an expmap target (3 controls, default)
or a quaternion target (4 controls) -- <orientation input="quat">, the
first actuator with different input and output widths. The signature is
recorded in a new per-actuator field actuator_ctrlspec (mjtCtrlChart),
whose meaning is scoped by the gain type the way gain/bias parameters are;
ctrlnum is derived from it at compile time and remains the layout
authority. An explicit field rather than width inference or a prm slot:
width-as-chart cannot express same-width signatures (upcoming servo input
subsets), and prm slots are the input_mode pattern this stack retires.
The force law normalizes the commanded quaternion, making it scale- and
antipodally-invariant. The all-zero ctrl still maps to the identity via
mju_normalize4, but it is a degenerate point (a nudge of any component
commands a half-turn), so quat inputs reset to the identity quaternion:
new mj_resetCtrl sets neutral ctrl values (zero, except qw = 1), called
by mj_resetData and the viewers' Clear All. The quat chart is
restricted to dyntype 'none': integrating a quaternion setpoint linearly
is not meaningful on the manifold. New mjsActuator.ctrlspec field carries
the signature through the spec and XML round-trip.
Actuator sensors (actuatorpos/vel/frc) now report one value per force
output; dim = 3 on an SO3 actuator.
As the first actuator with nu != nactuator, this commit also makes the
viewers multi-input aware: the control sliders in simulate and studio,
which indexed per-actuator arrays by control index (out of bounds on
this model class), are generated per control and labeled with the
actuator name plus an input suffix ("orient/qw"), via the new
introspection helper mj_actuatorInputName -- the single source of truth
for input names, extended by each new multi-input type (quaternion
components are w-first: qw, qx, qy, qz). Slider ranges now honor a
defined ctrlrange even when ctrllimited is false: range is the UI hint,
limited is the clamp -- wrapped and expmap setpoints are unbounded but
still want finite sliders, while quat components are truly bounded.
The rotational demo model is orientation.xml under
test/engine/testdata/actuation/, upgraded to a three-way contrast:
per-axis wrapped servos vs an expmap-commanded vs a quat-commanded
orientation actuator, on identical checker-textured boxes. It is loaded
by the mixed-axis contrast and input-name tests, and doubles as the
viewer test model (slider groups of 3 independent, 3 grouped, 4 grouped).
PiperOrigin-RevId: 951607063
Change-Id: If235dba8e2f2ca72672e7c62531a27e967c6a373
--
3439d8a5cd745dfd2776593916eead65cb456afc by Kevin Zakka <kevinarmandzakka@gmail.com>:
Respect an explicit CMAKE_INTERPROCEDURAL_OPTIMIZATION=OFF
The default-LTO guard checked the value rather than whether it was DEFINED, so an
explicit -DCMAKE_INTERPROCEDURAL_OPTIMIZATION=OFF (used in CI to cut build time)
was silently flipped back ON. Check DEFINED instead; default-on for Release is
preserved when the caller makes no choice.
--
bd15d1b264e5ab9594669b58fa1e0388d0a16f6b by Kevin Zakka <kevinarmandzakka@gmail.com>:
Speed up CI
~10x faster on the POSIX jobs (~45m -> ~4-5m) and ~2x overall wall-clock.
- Build Studio/Filament and WASM only in their dedicated canary jobs, not in
every matrix job (WASM uses emcc, which ignores the host compiler).
- Compiler matrix -> build_matrix.json with core/extended tiers: PRs build the
core set, pushes to main run the full sweep.
- ccache across build jobs (studio cache keyed on the Filament pin); fix the
Python-bindings build so ccache hits (CCACHE_BASEDIR).
- Disable IPO/LTO on POSIX (it was silently on); keep it on Windows where /GL-off
exposes a latent test bug and build time is not the bottleneck.
- Run ctest in parallel on POSIX (~2x); uv for Python installs (~29s -> ~8s).
- Move MJX (compiler-independent) to a single dedicated job.
- Restrict GITHUB_TOKEN to contents: read; bump actions off deprecated Node20.
--
c0618ab7aef524b238e0a30f2f23c50c4ce0c9b1 by Kevin Zakka <kevinarmandzakka@gmail.com>:
Build the gcc jobs with LTO off too
Restores LTO-off for gcc (recovering the build-time win). That surfaces known
gcc-12 -Wrestrict false positives in libstdc++ <char_traits> at -O3; two clean,
behavior-preserving rewrites in the XML writer avoid them. Confirmed gcc-12 and
gcc-14 build clean with LTO off (PR #3325).
--
2c38da0f48f77635e58e4b7931b86d60f3c253c7 by Kevin Zakka <kevinarmandzakka@gmail.com>:
Respect explicit IPO=OFF in the Simulate and Sample options too
Apply the same DEFINED check as cmake/MujocoOptions.cmake to the simulate/ and
sample/ subprojects (which carry identical copies of the guard) to keep the three
in sync, as the internal import requires. Since the guard now honors =OFF, stop
forcing IPO=OFF on the tiny samples/simulate CI builds so they keep their default
LTO and don't expose the gcc -Werror false positives that -O3-without-LTO triggers.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google-deepmind/mujoco/pull/3315 from google-deepmind:speed-up-ci 2c38da0f48f77635e58e4b7931b86d60f3c253c7
PiperOrigin-RevId: 930008698
Change-Id: Ic41dc87881833c95f2ebfc0602de1ed438db3d4f
Total testspeed runtime for `2humanoids100.xml` reduced by 27.6% (63.4 -> 49.5s) due to early termination on small islands
PiperOrigin-RevId: 906910915
Change-Id: If55ad468c3680ef44eda7000455a77f8003b3122
This is a combination of minor tweaks to the existing Studio WASM application, CMake cleanup to make it buildable, and integration with GitHub Actions for deployment.
The new features added are drag-and-drop, model specification through `?model=` URL parameter, HTTP and HTTPS resource provider, and a user-visible loading message while the model is loading.
PiperOrigin-RevId: 904594610
Change-Id: I3a73b1ca0fcd6fc9ba192469942b2dab010a9308
Add clamping to slider values in both ImGui widgets and mjUI to ensure they remain within the defined min/max ranges, preventing potential floating point inaccuracies from causing values to slightly exceed the bounds.
PiperOrigin-RevId: 902660458
Change-Id: Ia153ae7907f07b890f2b5ac2e3a4be93ecae7bd3
This forces `CMAKE_REQUIRED_FLAGS` to be treated as a string instead of list,
so none of the args get dropped when passed through `check_c_source_compiles`.
PiperOrigin-RevId: 886717366
Change-Id: I6314f9bfabd9d62bdcad5070c8801c453c381b79
The next release will include breaking changes, so this bumps the major
version to reflect that.
PiperOrigin-RevId: 881351265
Change-Id: I970c68a0211651c07fcb8109ebd0b138196cf50a
- Detect linker type (lld-link vs ld) and use the appropriate stack flag: /STACK:16777216 for lld-link, --stack,16777216 for MinGW ld
- Filter out stack-related flags from Python binding modules to prevent crashes when loading MuJoCo as a DLL in Python
Fixes#3011
Toggling visualization flags via keyboard shortcuts (e.g., C for contact points, F for contact forces) causes the SECT_RENDERING event handler to unconditionally re-evaluate the camera selection, which can reset a tracking camera to free mode. This particularly affects users of mujoco.viewer who set tracking cameras programmatically.
Guard the camera update logic so it only runs when the camera selector itself is changed, not when other items in the rendering section are modified.
PiperOrigin-RevId: 871600340
Change-Id: Ia6ef2a1617c6cdc18c7d10b4b099a8f2c620a824
Add usd_decoder to mujoco/plugin.
- Previously was src/experimental/usd/usd_to_mjspec.cc, now that same spec is returned by the plugin.
PiperOrigin-RevId: 854350962
Change-Id: Ia980190a557c50ce8197980922a2594b613859b4
This will fetch and call the build_usd.py build helper for OpenUSD and build within the MuJoCo _deps folder. Simplifying the build process for MuJoCo with USD. Users can still provide their own USD build, but must now do so via pxr_DIR.
PiperOrigin-RevId: 846735578
Change-Id: I0da7eb2da6465e2d518084340e762bf370adb6d5
- When `mjData.nefc` was 0 (solver not run at all), we were plotting the stale diagnostics from the last run; now fixed.
- In the case where islands were used, the semantic of the plotted diagnostic `sqrt_nnz` (square root of Hessian nonzeros) is changed. It used to be the sum of square roots of solver nonzeros per island, it is now the square root of the sum of all nonzeros in all islands. The semantic `sqrt(sum(nnz_i))` was found to be more consistent than `sum(sqrt(nnz_i))`, in the sense that switching islanding on/off leads to smaller changes and is more in line with the interpretation of islanding as [block-diagonalization](https://mujoco.readthedocs.io/en/3.3.7/computation/#constraint-islands).
PiperOrigin-RevId: 827624908
Change-Id: I5fc86364ead585a4752a6d44823fb75cf6c7c318
Michael Moss