Discovering Diverse Athletic Jumping Strategies

被引:33
作者
Yin, Zhiqi [1 ]
Yang, Zeshi [1 ]
Van de Panne, Michiel [2 ]
Yin, Kangkang [1 ]
机构
[1] Simon Fraser Univ, Burnaby, BC, Canada
[2] Univ British Columbia, Vancouver, BC, Canada
来源
ACM TRANSACTIONS ON GRAPHICS | 2021年 / 40卷 / 04期
基金
加拿大自然科学与工程研究理事会;
关键词
physics-based character animation; motion strategy; Bayesian optimization; control policy; OPTIMIZATION; MOTION; ANIMATION;
D O I
10.1145/3450626.3459817
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
We present a framework that enables the discovery of diverse and natural-looking motion strategies for athletic skills such as the high jump. The strategies are realized as control policies for physics-based characters. Given a task objective and an initial character configuration, the combination of physics simulation and deep reinforcement learning (DRL) provides a suitable starting point for automatic control policy training. To facilitate the learning of realistic human motions, we propose a Pose Variational Autoencoder (P-VAE) to constrain the actions to a subspace of natural poses. In contrast to motion imitation methods, a rich variety of novel strategies can naturally emerge by exploring initial character states through a sample-efficient Bayesian diversity search (BDS) algorithm. A second stage of optimization that encourages novel policies can further enrich the unique strategies discovered. Our method allows for the discovery of diverse and novel strategies for athletic jumping motions such as high jumps and obstacle jumps with no motion examples and less reward engineering than prior work.
引用
收藏
页数:17
相关论文
共 117 条
[1]  
Achiam J., 2018, ARXIV180710299
[2]   BIOMECHANICS ASPECTS OF TECHNIQUE OF HIGH JUMP [J].
Adashevskiy, V. M. ;
Iermakov, S. S. ;
Marchenko, A. A. .
PHYSICAL EDUCATION OF STUDENTS, 2013, 17 (02) :11-17
[3]  
Agrawal Shailen, 2013, Symposium on Computer Animation, P37
[4]   Trajectory Optimization for Full-Body Movements with Complex Contacts [J].
Al Borno, Mazen ;
de Lasa, Martin ;
Hertzmann, Aaron .
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, 2013, 19 (08) :1405-1414
[5]   Goal directed multi-finger manipulation: Control policies and analysis [J].
Andrews, S. ;
Kry, P. G. .
COMPUTERS & GRAPHICS-UK, 2013, 37 (07) :830-839
[6]  
[Anonymous], 2010, ACM Transactions on Graphics, DOI DOI 10.1145/1778765.1781157
[7]  
[Anonymous], 2010, THESIS
[8]  
[Anonymous], 2017, ABS170702286 CORR
[9]  
[Anonymous], 2016, Advances in Neural Information Processing Systems, V29
[10]   DReCon: Data-Driven Responsive Control of Physics-Based Characters [J].
Bergamin, Kevin ;
Clavet, Simon ;
Holden, Daniel ;
Forbes, James Richard .
ACM TRANSACTIONS ON GRAPHICS, 2019, 38 (06)