Embedded Files
Overview
The sampling time of a digital control determines how often the controller closes the loop by updating the control input. Typically, a fixed sampling time is chosen to balance the desired control performance with the cost of input updates. This performance vs. update cost tradeoff depends on the system and its environment. An autonomous car in an urban environment requires frequent updates at the cost of inference time and energy use. In healthcare, updates incur a monetary cost from appointments and tests. With AI agents in the loop, updates come with latency, licensing fees, and usage limits. Furthermore, the "best" choice of sampling time may change depending on the real-time conditions.Â
Adaptive-sampling control adjusts the input update rate of the controller online to manage the tradeoff between performance and update cost in real-time. However, existing methods often fail to guarantee robust constraint satisfaction during rate transitions or require computationally expensive online optimization. We developed a robust adaptive-sampling control framework for linear systems subject to polytopic state and input constraints and bounded additive disturbances (CDC '26). In this framework, an update rate for an MPC controller (IFAC '25) is computed by a parameter selection algorithm to track a time-varying reference update rate while ensuring robust constraint satisfaction at all time steps. The notion of robust M-step hold control invariance (CDC, L-CSS '25) is used offline to precompute invariant sets for a list of update rates and transition sets between them. The parameter selection uses these sets in real time to guarantee recursive feasibility and finite-time transitions to the reference update rate.
Publications:
S. Schutz, C. Vallon, and F. Borrelli, "An Adaptive-Sampling Control Framework for Constrained Linear Systems with Robust Safety Guarantees," IEEE 64th Conference on Decision and Control (CDC) (accepted), 2026
Page updated
Report abuse