New method models multirate control systems without iterative optimization
Cyclic reformulation creates accurate models from sensors with different sampling rates.
Modern control systems often use heterogeneous sensors operating at different sampling rates, creating intermittently missing outputs that complicate system identification. Hiroshi Okajima and Kakeru Ono introduce a non-iterative, control-oriented identification method for multirate systems based on cyclic reformulation. The technique transforms multirate data into an expanded time-invariant representation, then extracts M distinct parameter sets from a single input-output dataset, where M is the least common multiple of the sensor sampling periods.
These parameter sets serve two complementary purposes: their centroid provides a noise-reduced nominal model, while their convex hull constructs a polytopic uncertainty model compatible with vertex-based LMI robust control design. Numerical simulations show strong performance: an illustrative SISO example achieved higher validation FIT than the best individual vertex and substantially outperformed an interpolation-based baseline. A MIMO multirate sensing example confirmed that the polytope contains models whose validation FIT exceeds 95% on average, even at the highest tested noise level. This framework bridges multirate system identification with robust-control-oriented uncertainty modeling without requiring iterative EM-type optimization.
- Method uses cyclic reformulation to handle sensors with different sampling rates without iterative optimization.
- Generates M parameter sets (M = least common multiple of sampling periods), with centroid for nominal model and convex hull for uncertainty polytope.
- Numerical validation: SISO centroid outperforms interpolation baseline; MIMO polytope yields models with >95% average FIT even at high noise.
Why It Matters
Simplifies robust control design for systems with mixed-rate sensors, reducing computational complexity and improving model accuracy.