https://doi.org/10.1140/epjs/s11734-025-01926-z
Regular Article
Adaptive vibration suppression for a flexible truss-arm system using a hybrid-trained Memristive Spiking Neural Network
School of Mathematics and Statistics, Northwestern Polytechnical University, No. 1 Dongxiang Road, 710129, Xi’an, Shaanxi, China
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
5
July
2025
Accepted:
4
September
2025
Published online:
16
September
2025
In on-orbit servicing and assembly missions, the coupled vibrations between large-scale space manipulators and their flexible space trusses present a core challenge to system stability and mission accuracy. To address the limited adaptability of conventional fixed-gain controllers, this paper proposes a novel Memristive Spiking Neural Adaptive Controller (MSNAC). The core of this strategy is a bio-inspired Spiking Neural Network (SNN) whose synapses exhibit memristor-like characteristics, enabling online, autonomous tuning of Proportional-Derivative (PD) controller gains. We innovatively design a two-stage hybrid training framework to optimize this SNN: first, offline global pre-training of a computationally efficient proxy model is performed using Particle Swarm Optimization (PSO) to obtain high-performance initial weights (a warm start); subsequently, online fine-tuning is conducted on the full spiking model using Reward-modulated Spike-Timing-Dependent Plasticity (R-STDP). The event-driven nature of the SNN architecture, combined with its compact network size and the low-power concept of memristive synapses, provides a theoretical foundation for efficient onboard real-time control. Monte Carlo simulation results demonstrate that the proposed MSNAC strategy not only significantly outperforms the conventional PD controller in vibration suppression performance (a 51.0% reduction in Integrated Absolute Error, IAE), but its warm-start mode also exhibits significant advantages over cold-start (random initialization) in terms of final performance (a further 13.1% IAE reduction), convergence reliability, and policy stability.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
