Directed Time Series Regression for Control
This work addresses a specific problem in control systems for robotics or automation, but appears incremental as it builds on existing methods like least squares regression and empirical optimization.
The paper tackled the problem of estimating parameters for time-series models in certainty equivalent model predictive control by proposing directed time series regression, which combines least squares regression and empirical optimization, and demonstrated significant improvements in controller performance in a stochastic inverted pendulum balancing problem.
We propose directed time series regression, a new approach to estimating parameters of time-series models for use in certainty equivalent model predictive control. The approach combines merits of least squares regression and empirical optimization. Through a computational study involving a stochastic version of a well known inverted pendulum balancing problem, we demonstrate that directed time series regression can generate significant improvements in controller performance over either of the aforementioned alternatives.