Institute Homepage
Institute Homepage EN Sign In

Back

Autonomous Motion Article 2010

Stochastic Differential Dynamic Programming

We present a generalization of the classic Differential Dynamic Programming algorithm. We assume the existence of state- and control-dependent process noise, and proceed to derive the second-order expansion of the cost-to-go. Despite having quartic and cubic terms in the initial expression, we show that these vanish, leaving us with the same quadratic structure as standard DDP.

Author(s): Theodorou, E. and Tassa, Y. and Todorov, E.
Book Title: In the proceedings of American Control Conference (ACC 2010)
Year: 2010
Bibtex Type: Article (article)
Cross Ref: p10307
Electronic Archiving: grant_archive
Note: clmc
Links:

BibTex 

@article{EvangelosACC2010,
  title = {Stochastic Differential Dynamic Programming},
  booktitle = {In the proceedings of American Control Conference (ACC 2010) },
  abstract = {We present a generalization of the classic Differential Dynamic Programming algorithm. We assume the existence of state- and control-dependent process noise, and proceed to derive the second-order expansion of the cost-to-go. Despite having quartic and cubic terms in the initial expression, we show that these vanish, leaving us with the same quadratic structure as standard DDP. 
  },
  year = {2010},
  note = {clmc},
  slug = {evangelosacc2010},
  author = {Theodorou, E. and Tassa, Y. and Todorov, E.},
  crossref = {p10307}
}