Reservoir Predictive Path Integral Control for Unknown Nonlinear Dynamics
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911426557247488 |
|---|---|
| author | Inoue, Daisuke Matsumori, Tadayoshi Tanaka, Gouhei Ito, Yuji |
| author_facet | Inoue, Daisuke Matsumori, Tadayoshi Tanaka, Gouhei Ito, Yuji |
| contents | Neural networks have found extensive application in data-driven control of nonlinear dynamical systems, yet fast online identification and control of unknown dynamics remain central challenges. To meet these challenges, this paper integrates echo-state networks (ESNs)--reservoir computing models implemented with recurrent neural networks--and model predictive path integral (MPPI) control--sampling-based variants of model predictive control. The proposed reservoir predictive path integral (RPPI) enables fast learning of nonlinear dynamics with ESNs and exploits the learned nonlinearities directly in MPPI control computation without linearization approximations. This framework is further extended to uncertainty-aware RPPI (URPPI), which achieves robust stochastic control by treating ESN output weights as random variables and minimizing an expected cost over their distribution to account for identification errors. Experiments on controlling a Duffing oscillator and a four-tank system demonstrate that URPPI improves control performance, reducing control costs by up to 60% compared to traditional quadratic programming-based model predictive control methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_03839 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Reservoir Predictive Path Integral Control for Unknown Nonlinear Dynamics Inoue, Daisuke Matsumori, Tadayoshi Tanaka, Gouhei Ito, Yuji Systems and Control Machine Learning Optimization and Control Chaotic Dynamics Neural networks have found extensive application in data-driven control of nonlinear dynamical systems, yet fast online identification and control of unknown dynamics remain central challenges. To meet these challenges, this paper integrates echo-state networks (ESNs)--reservoir computing models implemented with recurrent neural networks--and model predictive path integral (MPPI) control--sampling-based variants of model predictive control. The proposed reservoir predictive path integral (RPPI) enables fast learning of nonlinear dynamics with ESNs and exploits the learned nonlinearities directly in MPPI control computation without linearization approximations. This framework is further extended to uncertainty-aware RPPI (URPPI), which achieves robust stochastic control by treating ESN output weights as random variables and minimizing an expected cost over their distribution to account for identification errors. Experiments on controlling a Duffing oscillator and a four-tank system demonstrate that URPPI improves control performance, reducing control costs by up to 60% compared to traditional quadratic programming-based model predictive control methods. |
| title | Reservoir Predictive Path Integral Control for Unknown Nonlinear Dynamics |
| topic | Systems and Control Machine Learning Optimization and Control Chaotic Dynamics |
| url | https://arxiv.org/abs/2509.03839 |