(base) shimazueiichi@shimadzueiichinoMacBook-Pro drag_teaching % python3 record_teach_demo.py run --teach drag_v2_20221013_231249.json [RTD ] 到達許容差は本線 run_pick_place.py の stage 値を使います（区間ごと。一律 8.0° は 2026-10-19 にやめた）: [RTD ] HOME 2.5° ← rpp.TOL_HOME（stage1_to_home / HOME 復帰） [RTD ] APPROACH_PICK 2.0° ← rpp.TOL_APPROACH（approach 系） [RTD ] PICK 3.0° ← rpp.TOL_PICK_PLACE（stage4_pick） [RTD ] APPROACH_PICK (lift) 3.0° ← rpp.TOL_LIFT_BACK（lift の残差 2.19° を吸収） [RTD ] APPROACH_PLACE 3.0° ← rpp.TOL_APPROACH_PLACE（stage7） [RTD ] PLACE 3.0° ← rpp.TOL_PICK_PLACE（stage8_place） [RTD ] APPROACH_PLACE (retreat) 3.0° ← rpp.TOL_LIFT_BACK（stage10_lift_back） [RTD ] 到達を待つ最大時間 = clamp(2.0 + max|Δ|/20*3.0, 8.0, 30.0) 秒（本線と同じ wr._compute_arrival_timeout） [RTD ] モード: ドライラン（アームに何も送りません） [RTD ] Pi から教えた記録を取得します: ubuntu@192.168.2.1:r2r2r-scripts/dynamic_tracking_limits/drag_teaching/drag_v2_20221013_231249.json [RTD ] 取得しました → /var/folders/5w/2q2mc64n6yv63gfh31h34w6h0000gn/T/record_teach_demo_vu3bxjcs/drag_v2_20221013_231249.json [RTD ] 入力ファイル: drag_v2_20221013_231249.json [RTD ] 抽出ウェイポイント: APPROACH_PICK <- phase APPROACH_PICK argmax_idx=146 (start=146 end=194, gap=12.04°) angles = [ -28.12, +57.21, +19.86, -3.60, -71.80, -27.50] PICK <- gripper close mark idx=214（PICK phase 範囲内） angles = [ -27.68, +63.10, +13.27, -8.26, -59.76, -27.33] APPROACH_PLACE <- phase APPROACH_PLACE argmax_idx=286 (start=286 end=321, gap=20.83°) angles = [ +41.04, +42.36, +1.66, +2.54, -49.39, -31.90] PLACE <- gripper open mark idx=345（PLACE phase 範囲内） angles = [ +40.69, +61.78, +1.14, +4.13, -70.22, -31.28] [RTD ] JOINT_LIMITS 一括検証 OK（全 endpoint + HOME） [RTD ] 姿勢検査 OK（全 137 姿勢が 有限値 + JOINT_LIMITS + J4_SAFE [-90.0, +60.0] 内） [RTD ] j4 の余裕は最小 18.20°（しきい値 3.0° 以上） ============================================================================== [RTD ] 実行プラン（daemon 経由・区間ごとに 1 回送信 → 到達待ち） ============================================================================== [STEP 1] MOVE → HOME (single send + settle) speed=20 current = 未取得（ドライラン・daemon に接続していない） to HOME = [ +0.00, +0.00, +0.00, +0.00, +0.00, -30.00] [STEP 2] GRIPPER OPEN daemon_set_gripper(0, 50, wait=0) ×1 回（daemon 事前 2.0s + 事後 2.0s + 待ち 2.0s）= 計 6.0s 開いたかは確認しません（gv を使わず・開始前の y/n も無し。判定は終了後の y/n と動画） [STEP 3] MOVE → APPROACH_PICK segment HOME → APPROACH_PICK tol=2.0° [rpp.TOL_APPROACH（approach 系）] from = [ +0.00, +0.00, +0.00, +0.00, +0.00, -30.00] to = [ -28.12, +57.21, +19.86, -3.60, -71.80, -27.50] max |Δ| = 71.80° 1 回送信 speed=20 (~ 3.6s @ 20°/s 目安・到達待ち 最大 12.8s) [STEP 4] MOVE → PICK segment APPROACH_PICK → PICK [critical] tol=3.0° [rpp.TOL_PICK_PLACE（stage4_pick）] from = [ -28.12, +57.21, +19.86, -3.60, -71.80, -27.50] to = [ -27.68, +63.10, +13.27, -8.26, -59.76, -27.33] max |Δ| = 12.04° 1 回送信 speed=20 (~ 0.6s @ 20°/s 目安・到達待ち 最大 8.0s) [STEP 5] GRIPPER CLOSE daemon_set_gripper_value(20, 50) + sleep(2.0s) [STEP 6] MOVE → APPROACH_PICK (lift) segment PICK → APPROACH_PICK (lift) tol=3.0° [rpp.TOL_LIFT_BACK（lift の残差 2.19° を吸収）] from = [ -27.68, +63.10, +13.27, -8.26, -59.76, -27.33] to = [ -28.12, +57.21, +19.86, -3.60, -71.80, -27.50] max |Δ| = 12.04° 1 回送信 speed=20 (~ 0.6s @ 20°/s 目安・到達待ち 最大 8.0s) [STEP 7] MOVE → APPROACH_PLACE segment APPROACH_PICK → APPROACH_PLACE tol=3.0° [rpp.TOL_APPROACH_PLACE（stage7）] from = [ -28.12, +57.21, +19.86, -3.60, -71.80, -27.50] to = [ +41.04, +42.36, +1.66, +2.54, -49.39, -31.90] max |Δ| = 69.16° 1 回送信 speed=20 (~ 3.5s @ 20°/s 目安・到達待ち 最大 12.4s) [STEP 8] MOVE → PLACE segment APPROACH_PLACE → PLACE [critical] tol=3.0° [rpp.TOL_PICK_PLACE（stage8_place）] from = [ +41.04, +42.36, +1.66, +2.54, -49.39, -31.90] to = [ +40.69, +61.78, +1.14, +4.13, -70.22, -31.28] max |Δ| = 20.83° 1 回送信 speed=20 (~ 1.0s @ 20°/s 目安・到達待ち 最大 8.0s) [STEP 9] GRIPPER OPEN daemon_set_gripper(0, 50, wait=0) ×1 回（daemon 事前 2.0s + 事後 2.0s + 待ち 2.0s）= 計 6.0s 開いたかは確認しません（gv を使わず・開始前の y/n も無し。判定は終了後の y/n と動画） [STEP 10] MOVE → APPROACH_PLACE (retreat) segment PLACE → APPROACH_PLACE (retreat) tol=3.0° [rpp.TOL_LIFT_BACK（stage10_lift_back）] from = [ +40.69, +61.78, +1.14, +4.13, -70.22, -31.28] to = [ +41.04, +42.36, +1.66, +2.54, -49.39, -31.90] max |Δ| = 20.83° 1 回送信 speed=20 (~ 1.0s @ 20°/s 目安・到達待ち 最大 8.0s) [STEP 11] MOVE → HOME segment APPROACH_PLACE → HOME tol=2.5° [rpp.TOL_HOME（stage1_to_home / HOME 復帰）] from = [ +41.04, +42.36, +1.66, +2.54, -49.39, -31.90] to = [ +0.00, +0.00, +0.00, +0.00, +0.00, -30.00] max |Δ| = 49.39° 1 回送信 speed=20 (~ 2.5s @ 20°/s 目安・到達待ち 最大 9.4s) ============================================================================== [RTD ] 送信 = 8 回（STEP 1 + move 7 区間）, 想定所要時間 ≈ 29.8s（最悪 73.0s） [RTD ] 内訳: 先頭 settle 3.0s + 移動 12.8s（max|Δ| 合計 256.1° @ 20°/s 目安） + gripper 14.0s [RTD ] 到達確認: 許容は区間ごと（本線 run_pick_place.py の stage 値）/ 最大は clamp(2.0 + max|Δ|/20*3.0, 8.0, 30.0) 秒 [RTD ] 届かないとき: 送信前から 1.0° 未満しか動いていなければ 5.0s 待って**同じ目標へ 1 回だけ**再送（本線と同じ）。それでも届かなければ ABORT（HOME へ自動で戻らない） [RTD ] 実測 j4: [-90.0, +60.0] の外は WARN / [-98.0, +68.0] の外は ABORT ============================================================================== [RTD ] measure: /Volumes/KIOXIA/r2r2r-data/mecharm_calibration/teach_poses_LOG359/measure_20261022_164406.json [RTD ] measured_at = 2026-10-22T16:44:08.337908 block = (+127.6, +216.8) mm box = (-213.6, +178.2) mm [RTD ] 再生直前: 上カメラでマット四隅とブロック画素を確認します [RTD ] 再生直前: block = (+127.2, +216.8) mm（measure の homography で換算） [RTD ] 確認画像: /Volumes/KIOXIA/r2r2r-data/mecharm_calibration/teach_poses_LOG359/prerun_20261022_164432.jpg [RTD ] ブロック位置の照合 OK（画素の差 1.0px <= 許容 10.6px = 5.0mm / mm 換算の差 0.43mm） [RTD ] マット四隅のずれ 最大 4.5px（上限 15.0px・やり直し 1 回） [RTD ] ドライランなので、ここで終了します（アームには何も送っていません） [RTD ] Pi 版と比べる点: 抽出ウェイポイント 4 姿勢と各区間の max|Δ| は同じ。送り方は単発（Pi 版は 0.1s 補間ストリーミング）なので n_steps / ticks は比較対象外。送信 8 回 / 想定所要 ≈ 29.8s (base) shimazueiichi@shimadzueiichinoMacBook-Pro drag_teaching %