A Minimum-Footprint Implementation of Discrete-Time ADRC

TL;DR

This paper presents the most efficient discrete-time ADRC implementation, significantly reducing computational and storage demands to enable deployment on low-cost, high-speed, and resource-constrained embedded systems.

Contribution

It introduces a minimal-footprint linear discrete-time ADRC implementation that preserves performance while minimizing computational effort and storage needs.

Findings

Achieves the lowest computational effort among existing implementations.

Maintains exact performance characteristics of the state-space form.

Enables ADRC deployment on low-cost embedded systems.

Abstract

To foster the adoption of active disturbance rejection control (ADRC) and support its deployment even on low-cost embedded systems, this article introduces the most efficient implementation of linear discrete-time ADRC to date. While maintaining all features and the exact performance characteristics of the state-space form, computational efforts and storage requirements are reduced to a minimum compared to all existing implementations. This opens up new possibilities to use ADRC in applications with high sample rates, tight timing constraints, or low computational power.