Login / Signup

An Anticipatory Scheme for the Model Predictive Control of Circadian Phase for Expected Environmental Light Changes.

Lindsey S BrownElizabeth B KlermanFrancis J Doyle
Published in: IEEE control systems letters (2021)
The circadian system is critical to timing biological functions in anticipation of daily environmental light changes, but much previous work on the development of molecular control inputs to shift the phase of the circadian system has applied model predictive control (MPC) without considering expected environmental light changes. We augment the MPC algorithm to develop an anticipatory control algorithm, which has advantages over MPC in achieving scheduled phase shifts (as occurs with jet lag and shiftwork). We further extend the algorithm in a model switching control scheme to account for changes in the light environment. Taken together, these two enhancements to the standard MPC framework allow for better control of the circadian oscillator in more realistic environments by anticipating environmental light changes.
Keyphrases
  • machine learning
  • deep learning
  • physical activity
  • risk assessment
  • high frequency
  • life cycle