[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-142042-105":59,"doc-detail-142042-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","appendix-a-evidence-indicating-transfer-to-real-world","Appendix A Evidence indicating transfer to real-world","","Appendix A and B provide supporting evidence and experimental details for transferring learned policies from simulation to real-world manipulation tasks. The document summarizes convex decomposition and dataset composition across EGAD and YCB, action-space design with relative position control, and student policy requirements for joint positions, object pose, or point clouds. It further explains domain randomization parameters, added observation/action and vision noise, success-rate outcomes under randomization, and torque analysis ensuring feasible peak loads. Appendix B and C extend the setup with environment, state definition, camera configuration, and network architecture references.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/appendix-a-evidence-indicating-transfer-to-real-world/142042/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/appendix-a-evidence-indicating-transfer-to-real-world/142042.png","ImageObject",300,407,{"name":92,"@type":93},"Skyler","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-23","2026-08-25",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"为何实验主要在仿真环境中进行？","Question",{"text":112,"@type":113},"由于问题范围较大且持续受到疫情影响，实验被限制在仿真中进行。随后在论文中给出了仿真策略可在未来转移到真实世界的证据。","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"学生策略需要哪些真实世界可获得的观测信息？",{"text":117,"@type":113},"第一种学生策略需要关节位置和物体姿态，物体姿态可由姿态估计或真实世界的动作捕捉获得。第二种学生策略仅需要场景点云与关节位置，可用 RGBD 相机获取点云。",{"name":119,"@type":110,"acceptedAnswer":120},"域随机化如何帮助缓解仿真到真实的差距？",{"text":121,"@type":113},"文中随机化了物体质量、摩擦、关节与肌腱等阻尼与刚度等参数，并对状态观测与动作指令加入噪声。视觉实验还对点云观测进行数据增强噪声处理，从而在添加随机性后仍保持较高成功率，表明策略具备真实迁移潜力。","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},142042,1787673695,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":41,"preview_page_count":4},2336464648746,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Appendix A Evidence indicating transfer to real-world  \nDue to the scope of the problem and the ongoing pandemic, we limit our experiments to be in simulation. However, we provided evidence indicating that the learned policies can be transferred to the real world in the future in the paper. We summarize this evidence as follows.  \nConvex decomposition The objects after the convex decomposition still have geometrically different and complex geometries as shown in Figure B.3 . The objects in the EGAD dataset are 3D printable. The YCB objects are available in the real world.  \nAction space We control the finger joints via relative position control as explained in Section 2.1. This suffers less sim-to-real gap compared to using torque control on the joints directly.  \nStudent policies We designed two student policies and both of them use the observation data that can be readily acquired from the real world. The first student policy only requires the joint positions and the object pose. Object pose can be obtained using a pose estimation system or a motion capture system in the real world. Our second student policy only require the point cloud of the scene and the joint positions. We can get the point cloud in the real world by using RGBD cameras such as Realsense D415, Azure Kinect, etc.  \nDomain randomization We also trained and tested our policies with domain randomization. We randomized object mass, friction, joint damping, tendon damping, tendon stiffness, etc. Table C.4 lists all the parameters we randomized in our experiments. We also add noise to the state observation and action commands as shown in Table C.4 . For the vision experiments, we also added noise (various ways of data augmentation including point position jittering, color jiterring, dropout, etc.) to the point cloud observation in training and testing as explained in Section D.5 .  \nThe results in Table D.5, Table D.6, and Table 4 show that even after adding randomization/noise, we can still get good success rates with the trained policies. Even though we cannot replicate the true real-world setups in the simulation, our results with domain randomization indicates a high possibility that our policies can be transferred to the real Shadow hand. Prior works [11] have also shown the domain randomization can effectively reduce the sim-to-real gap.  \nTorque analysis We also conducted torque analysis as shown in Section D.4 . We can see that the peak torque values remain in an reasonable and affordable range for the Shadow hand. This indicates that our learned policies are less likely to cause motor overload on the real Shadow hand.  \nAppendix B Environment Setup  \nFigure B.1: We learn policies that can reorient many objects in three scenarios respectively: (1) hand faces upward,(2) hand faces downward with a table below the hand,(3) hand faces downward without any table. The extra object in each figure shows the desired orientation.  \nB.1 State definition  \nThe full state space SE includes joint, fingertip, object and goal information detailed in Table B.1. To compute the angle difference between two quaternion orientations 􀀋 1 and 􀀋 2 , we first compute the difference rotation quaternion: 􀀌 = 􀀋 1 􀀋 1 . Then the angle difference (distance between two rotations) 􀀁􀀒 is computed as the angle of 􀀌 from the axis-angle representation of 􀀌 .  \nTable B. 1: Full state sEt 2 R 134 information. Orientations are in the form of quaternions.  \n\n| Parameter | Description | Parameter | Description | Parameter | Description |\n| --- | --- | --- | --- | --- | --- |\n| qt 2 R24 | joint positions | vft 2 R15 | fingertip linear velocities | 􀀋 g 2 R4 | object goal orientation |\n| t 2 R24 | joint velocities | wft 2 R15 | fingertip angular velocities | vot 2 R3 | object linear velocity |\n| pft 2 R15 | fingertip positions | pot 2 R3 | object position | wot 2 R3 | object angular velocity |\n| 􀀋ft 2 R20 | fingertip orientation | 􀀋ot 2 R4 | object orientation | 􀀌t 2 R4 | 􀀋ot (􀀋g )􀀀1 |\n\nB.2 Dataset  \nWe use ","cbCairhS3HoTBcEw","https://ap.wps.com/l/cbCairhS3HoTBcEw","pdf",10920017,12,"English","# Appendix A\n## Evidence indicating transfer to real-world\n## Convex decomposition\n## Action space\n## Student policies\n## Domain randomization\n## Torque analysis\n# Appendix B Environment Setup\n## State definition\n## Dataset\n## Camera setup\n# Appendix C Experiment Setup\n## Network architecture","[{\"question\":\"为何实验主要在仿真环境中进行？\",\"answer\":\"由于问题范围较大且持续受到疫情影响，实验被限制在仿真中进行。随后在论文中给出了仿真策略可在未来转移到真实世界的证据。\"},{\"question\":\"学生策略需要哪些真实世界可获得的观测信息？\",\"answer\":\"第一种学生策略需要关节位置和物体姿态，物体姿态可由姿态估计或真实世界的动作捕捉获得。第二种学生策略仅需要场景点云与关节位置，可用 RGBD 相机获取点云。\"},{\"question\":\"域随机化如何帮助缓解仿真到真实的差距？\",\"answer\":\"文中随机化了物体质量、摩擦、关节与肌腱等阻尼与刚度等参数，并对状态观测与动作指令加入噪声。视觉实验还对点云观测进行数据增强噪声处理，从而在添加随机性后仍保持较高成功率，表明策略具备真实迁移潜力。\"}]","Appendix A Evidence indicating transfer to real-world | PDF"]