Trajectory-Based Dynamic Programming
作者： Christopher G. Atkeson , Chenggang Liu
DOI: 10.1007/978-3-642-36368-9_1
关键词:
摘要: We informally review our approach to using trajectory optimization accelerate dynamic programming. Dynamic programming provides a way design globally optimal control laws for nonlinear systems. However, the curse of dimensionality, exponential dependence memory and computation resources needed on dimensionality state control, limits application in practice. explore trajectory-based programming, which combines many local optimizations global are able solve problems with less than grid-based approaches, we couldn’t before tabular or function approximation approaches.
springer.com
sci-hub.se 参考文章(5)
Stefan Schaal, Christopher G. Atkeson, Robot Learning From Demonstration international conference on machine learning. pp. 12- 20 ,(1997)
Jennie Si, Andrew G Barto, Warren B Powell, Don Wunsch, Handbook of Learning and Approximate Dynamic Programming (2004). ,(2004) , 10.1109/9780470544785
Yuval Tassa, Tom Erez, Emanuel Todorov, Synthesis and stabilization of complex behaviors through online trajectory optimization intelligent robots and systems. pp. 4906- 4913 ,(2012) , 10.1109/IROS.2012.6386025
Dimitri P. Bertsekas, Dynamic Programming and Optimal Control Athena Scientific. ,(1995)
A.G. Barto, R.S. Sutton, Reinforcement Learning: An Introduction ,(1988)