[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-sim-to-real":3},{"tag":4,"articles":10},{"id":5,"name":6,"slug":6,"article_count":7,"description_zh":8,"description_en":9},"9e87f070-fefc-4f6d-9768-d8b04dc7edfb","sim-to-real",3,"Sim-to-real 指把模擬環境中的學習結果遷移到真實世界，重點在縮小物理落差與資料成本。常見做法包括域隨機化、數位孿生、擴散式軌跡生成與強化學習，特別適用於機器人操作、可變形物體與物理推理。","Sim-to-real focuses on transferring policies or models from simulation to real-world use while reducing the gap in dynamics, sensing, and control. It shows up in domain randomization, digital twins, synthetic trajectory generation, robot manipulation, and physics-based reasoning.",[11],{"id":12,"slug":13,"title":14,"summary":15,"category":16,"image_url":17,"cover_image":17,"language":18,"created_at":19},"81ffcc19-e976-462b-9c51-006a53187ff4","sim1-physics-aligned-deformable-worlds-zh","SIM1：把少量示範變成可訓練資料","SIM1 把真實示範先對齊成物理一致的數位孿生，再用擴散式軌跡生成擴充資料，目標是讓可變形物體操作更省真實資料。","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775801210052-3us2.png","zh","2026-04-10T06:06:34.149021+00:00"]