基于深度强化学习的虚拟手自适应抓取研究  

Research on adaptive grasping of virtual hands based on deep reinforcement learning

作　　者：伍一鹤 张振宁 仇栋 李蔚清[2] 苏智勇[1] WU Yi-he;ZHANG Zhen-ning;QIU Dong;LI Wei-qing;SU Zhi-yong(School of Automation,Nanjing University of Science and Technology,Nanjing Jiangsu 210094,China;School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing Jiangsu 210094,China)

机构地区：[1]南京理工大学自动化学院,江苏南京210094 [2]南京理工大学计算机科学与工程学院,江苏南京210094

出　　处：《图学学报》2021年第3期462-469,共8页Journal of Graphics

基　　金：“十三五”装备预研项目(61409230104,1017,315100104);中央高校基本科研业务费专项(30918012203);上海航天科技创新基金(SAST2019009)。

摘　　要：在计算机角色动画的抓取研究中,生成动作序列的自然性、稳定性及自适应能力三者难以同时得到保证,即自然又稳定的抓取控制器往往泛化能力有限,无法适用于其他类型、尺寸物体的抓取任务。通过引入和抓取类型相对应的手部示教数据、设计回报函数,构建了一种基于深度强化学习的虚拟手自适应抓取控制器。实验结果表明,该控制器能够生成兼具自然性和稳定性的抓取运动序列,同时对素材库中不同尺寸、不同类型的基元物体也具备较好的自适应能力。For the grasping of computer character animation,it is difficult to guarantee the naturalness,stability and adaptability of the generated action sequence at the same time.In other words,the natural and stable grasping controller are often limited in generalization and cannot be applied to other types of grabbing tasks.A virtual hand adaptive grasping controller was constructed based on deep reinforcement learning by introducing hand teaching data corresponding to the grasping types and by designing the reward function.Experimental results show that the designed controller can generate a grasping motion sequence with both naturalness and stability,and are also highly adaptive for different sizes and types of primitive objects in the material library.

关 键 词：深度强化学习 示教学习 运动生成 虚拟手 动作捕捉数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]