Inverse reinforcement learning for discrete-time nonlinear systems: applications in circuits and systems

Pengxu Ren, Wei Dai, Lei Ma and Chunyu Yang

International Journal of Systems Science, 2026, vol. 57, issue 3, 883-893

Abstract: This article proposes a novel inverse reinforcement learning (RL) algorithm to address the inverse optimal control (IOC) problem in discrete-time nonlinear circuits and systems with unknown performance function. We consider an expert-learner imitation structure, and adopt the learner to mimic the expert’s behaviour trajectories. Initially, a model-based value iteration inverse RL algorithm is developed, enabling the learner to reconstruct the unknown performance function utilising the observed demonstration behaviour trajectories from the expert. The algorithm includes a performance function learning stage derived from IOC and a control policy learning stage derived from optimal control. Subsequently, an actor-critic-based inverse RL algorithm is further implemented to obviate the need for information about the system dynamics for both the learner and expert. Finally, theoretical analysis is provided, and a boost converter circuit is used as an example to demonstrate the effectiveness of the proposed control algorithms.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/00207721.2025.2515227 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:tsysxx:v:57:y:2026:i:3:p:883-893

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/TSYS20

DOI: 10.1080/00207721.2025.2515227

Access Statistics for this article

International Journal of Systems Science is currently edited by Visakan Kadirkamanathan

More articles in International Journal of Systems Science from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().