Сортировочная ячейка Isaac Sim: CV-пайплайн и меши товаров

Замкнутый контур "поток -> CV -> механика": товары идут по конвейеру с шагом 700 мм, класс определяется стереопайплайном во время движения, пушер и плуг реагируют физически. Состав: * control_test/ - ячейка и CV. run_sorting_cv.py + cv_worker.py (два процесса, потому что torch внутри Isaac роняет сцену), cell.py (физика лент, плуга, пушера), measure_plane.py (замер габаритов), README.md и .memory.md с замерами, проблемами и ловушками * robozon_sorter/ - модули симуляции, scripts/ - утилиты, scene/ - сцены * assets/ - меши товаров, плуг, объекты Objaverse Бейзлайн CV: DEFOM-Stereo vitl, вход 480, iters 24, кроп зоны осмотра, без сегментации. На потоке 700 мм - классы 8/9, габариты MAE 32.8 мм, 469 мс на товар при такте 700 мс. Веса моделей (4.5 ГБ) и пропсы конвейера NVIDIA (274 МБ) не включены - источники и команды скачивания в MODELS.md. Выход прогонов (captures/, runtime/) не включён: воспроизводится. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-01 13:07:24 +00:00
parent 6ce460378a
commit 0d32f32db0
342 changed files with 18000 additions and 0 deletions
@@ -0,0 +1,98 @@
+"""SUPERSEDED - kept for history only.
+
+SUPERSEDED by cam_configs.apply_config(stage, cam_configs.DEFAULT).
+Hard-codes the old 600 mm standoff.
+"""
+
+"""Reposition the camera rigs:
+  * RealSense D435: Left and Right on OPPOSITE sides of the belt, facing each other.
+  * All three rigs: working distance exactly 600 mm to the inspection point.
+Gemini 305/345 keep their azimuth and stay rectified (shared orientation, right eye
+offset along the camera's local +X by the baseline).
+"""
+import json, math
+import omni.usd
+from pxr import Gf, Usd, UsdGeom
+
+stage = omni.usd.get_context().get_stage()
+TARGET = Gf.Vec3d(-0.750, 0.0, 1.781)      # inspection point on the belt
+STANDOFF = 0.60                             # requested working distance, metres
+SIDE_ELEV_DEG = 20.0                        # D435 sits low, at the side
+UP = Gf.Vec3d(0, 0, 1)
+
+def find(name):
+    for p in stage.Traverse():
+        if p.IsA(UsdGeom.Camera) and p.GetName() == name:
+            return p
+    raise KeyError(name)
+
+def look_at_matrix(pos, target):
+    """USD camera convention: looks down local -Z, local +Y is up."""
+    fwd = (target - pos)
+    fwd = fwd / (fwd.GetLength() or 1.0)
+    zax = -fwd
+    up = UP
+    if abs(Gf.Dot(up, zax)) > 0.999:
+        up = Gf.Vec3d(0, 1, 0)
+    xax = Gf.Cross(up, zax); xax = xax / (xax.GetLength() or 1.0)
+    yax = Gf.Cross(zax, xax)
+    M = Gf.Matrix4d(1.0)
+    M.SetRow3(0, xax); M.SetRow3(1, yax); M.SetRow3(2, zax)
+    M.SetTranslateOnly(pos)
+    return M, xax, fwd
+
+def set_pose(prim, M):
+    xf = UsdGeom.Xformable(prim)
+    xf.ClearXformOpOrder()
+    xf.AddTransformOp().Set(M)
+
+def current_dir_from_target(prim):
+    p = UsdGeom.XformCache().GetLocalToWorldTransform(prim).ExtractTranslation()
+    d = Gf.Vec3d(p) - TARGET
+    return d / (d.GetLength() or 1.0)
+
+report = {}
+
+# ---- D435: opposing pair, one each side, both aimed at the belt --------------------
+th = math.radians(SIDE_ELEV_DEG)
+for side_name, sgn in (("Left", +1.0), ("Right", -1.0)):
+    prim = find(f"RealSense_D435_{side_name}")
+    pos = TARGET + Gf.Vec3d(0.0, sgn * STANDOFF * math.cos(th), STANDOFF * math.sin(th))
+    M, _, fwd = look_at_matrix(pos, TARGET)
+    set_pose(prim, M)
+    report[f"RealSense_D435_{side_name}"] = dict(
+        pos=[round(v, 4) for v in pos], fwd=[round(v, 4) for v in fwd],
+        dist_mm=round((TARGET - pos).GetLength() * 1000, 1))
+
+# ---- Gemini pairs: same azimuth, distance forced to 600 mm, stay RECTIFIED ---------
+for rig, baseline_m in (("Orbbec_Gemini305", 0.0265), ("Orbbec_Gemini345", 0.1294)):
+    left, right = find(f"{rig}_Left"), find(f"{rig}_Right")
+    # keep the direction the pair currently views from, measured at its midpoint
+    xc = UsdGeom.XformCache()
+    pl = Gf.Vec3d(xc.GetLocalToWorldTransform(left).ExtractTranslation())
+    pr = Gf.Vec3d(xc.GetLocalToWorldTransform(right).ExtractTranslation())
+    mid = (pl + pr) * 0.5
+    d = mid - TARGET
+    d = d / (d.GetLength() or 1.0)
+    centre = TARGET + d * STANDOFF
+    M, xax, fwd = look_at_matrix(centre, TARGET)
+    # both eyes share ONE orientation - that is what keeps the pair rectified; the right
+    # eye is displaced along the camera's own +X by the baseline
+    for prim, off in ((left, -baseline_m / 2.0), (right, +baseline_m / 2.0)):
+        Mi = Gf.Matrix4d(M)
+        Mi.SetTranslateOnly(centre + xax * off)
+        set_pose(prim, Mi)
+        report[prim.GetName()] = dict(
+            pos=[round(v, 4) for v in (centre + xax * off)],
+            fwd=[round(v, 4) for v in fwd],
+            dist_mm=round((TARGET - (centre + xax * off)).GetLength() * 1000, 1))
+
+print(f"{'camera':>26} {'position':>26} {'dist to belt':>13}")
+for k, v in report.items():
+    print(f"{k:>26} ({v['pos'][0]:+6.3f},{v['pos'][1]:+6.3f},{v['pos'][2]:+6.3f}) {v['dist_mm']:>10.1f} mm")
+
+out = "/home/dasha/robozon-sorter/control_test/calib_rs_side.json"
+json.dump(dict(target=[round(v, 4) for v in TARGET], standoff_m=STANDOFF,
+               side_elevation_deg=SIDE_ELEV_DEG, cameras=report),
+          open(out, "w"), indent=2)
+print(f"\nextrinsics -> {out}")