chenlin/Mujoco_WASM
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Optimize flexedge velocity computation.

Browse Source 
Skip flexedge velocity calculation for rigid and interpolated flexes, as they are not used in those cases. The computation is now done per-flex, only for the edges within each non-rigid, non-interpolated flex.

PiperOrigin-RevId: 915907396
Change-Id: I0beb1f98f0115c83da13fed184e89101f57563ab
This commit is contained in:
Alessio Quaglino
2026-05-15 03:15:27 -07:00
committed by Copybara-Service
parent 7ca1bc6a56
commit 0ad77f1ddf
1 changed files with 10 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/engine/engine_forward.c
+10 -3
View File
@@ -227,9 +227,16 @@ void mj_fwdVelocity(const mjModel* m, mjData* d) {
d->flg_subtreevel = 0;
d->flg_energyvel = 0;
// flexedge velocity: always sparse
mju_mulMatVecSparse(d->flexedge_velocity, d->flexedge_J, d->qvel, m->nflexedge,
m->flexedge_J_rownnz, m->flexedge_J_rowadr, m->flexedge_J_colind, NULL);
// flexedge velocity: skip interp and rigid flexes (edge Jacobians are zero)
mju_zero(d->flexedge_velocity, m->nflexedge);
for (int f = 0; f < m->nflex; f++) {
if (m->flex_rigid[f] || m->flex_interp[f]) continue;
int adr = m->flex_edgeadr[f];
int num = m->flex_edgenum[f];
mju_mulMatVecSparse(d->flexedge_velocity + adr, d->flexedge_J, d->qvel, num,
m->flexedge_J_rownnz + adr, m->flexedge_J_rowadr + adr,
m->flexedge_J_colind, NULL);
}
// tendon velocity: always sparse
mju_mulMatVecSparse(d->ten_velocity, d->ten_J, d->qvel, m->ntendon,