Adaptive Manipulator Control using Active Inference with Precision Learning

Mohamed Baioumy, Matías Mattamala, Paul Duckworth, Bruno Lacerda, Nick HawesJournal of Robotics & Autonomous Systems 2020 · DOI

Abstract

Active inference provides a framework for decision- making where the optimization is achieved by minimizing free- energy. Previous work has used this framework for control and state-estimation of a robotic manipulator. This required manual definition of precision matrices which serve as controller gains. This paper provides an implementation for control and state-estimation where the precision matrices are tuned during execution-time (precision learning). Learning the precision ma- trices means automatically adjusting the controller’s gains which decreases oscillations and overshoot.
BibTeX
@article{baioumy2020adaptive,
  author = {Mohamed Baioumy and Matías Mattamala and Paul Duckworth and Bruno Lacerda and Nick Hawes},
  title = {{Adaptive Manipulator Control using Active Inference with Precision Learning}},
  journal = {Journal of Robotics \& Autonomous Systems},
  year = {2020},
  doi = {10.31256/gz8eu7c},
}

Loading related content...

Figure