Bayesian optimization for learning gaits under uncertainty

被引:207
作者
Calandra, Roberto [1 ]
Seyfarth, Andre [2 ]
Peters, Jan [1 ,3 ]
Deisenroth, Marc Peter [4 ]
机构
[1] Tech Univ Darmstadt, Intelligent Autonomous Syst, Darmstadt, Germany
[2] Tech Univ Darmstadt, Lauflabor Locomot Lab, Darmstadt, Germany
[3] Max Planck Inst Intelligent Syst, Tubingen, Germany
[4] Univ London Imperial Coll Sci Technol & Med, Dept Comp, London, England
关键词
Gait optimization; Bayesian optimization; Robotics; Locomotion; GLOBAL OPTIMIZATION; WALKING; SEARCH;
D O I
10.1007/s10472-015-9463-9
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Designing gaits and corresponding control policies is a key challenge in robot locomotion. Even with a viable controller parametrization, finding near-optimal parameters can be daunting. Typically, this kind of parameter optimization requires specific expert knowledge and extensive robot experiments. Automatic black-box gait optimization methods greatly reduce the need for human expertise and time-consuming design processes. Many different approaches for automatic gait optimization have been suggested to date. However, no extensive comparison among them has yet been performed. In this article, we thoroughly discuss multiple automatic optimization methods in the context of gait optimization. We extensively evaluate Bayesian optimization, a model-based approach to black-box optimization under uncertainty, on both simulated problems and real robots. This evaluation demonstrates that Bayesian optimization is particularly suited for robotic applications, where it is crucial to find a good set of gait parameters in a small number of experiments.
引用
收藏
页码:5 / 23
页数:19
相关论文
共 36 条
[1]  
[Anonymous], 2010, A tutorial on Bayesian optimization of expensive cost functions
[2]  
[Anonymous], 2014, P INT C LEARN INT OP
[3]  
[Anonymous], 2009, 3 INT C LEARN INT OP
[4]  
Bergstra J, 2012, J MACH LEARN RES, V13, P281
[5]  
Bertsekas D. P., 2007, Dynamic Programming and Optimal Control
[6]   A DISCUSSION OF RANDOM METHODS FOR SEEKING MAXIMA [J].
BROOKS, SH .
OPERATIONS RESEARCH, 1958, 6 (02) :244-251
[7]   A LIMITED MEMORY ALGORITHM FOR BOUND CONSTRAINED OPTIMIZATION [J].
BYRD, RH ;
LU, PH ;
NOCEDAL, J ;
ZHU, CY .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1995, 16 (05) :1190-1208
[8]  
Calandra R., 2014, INT C ROB AUT ICRA
[9]  
Chernova S., 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566), P2562
[10]  
Cox DD, 1997, SIAM PROC S, P315