Sign in Subscribe
Machine learning

Adaptive Finite-Time Optimal Control of Stochastic Nonlinear Systems

Strategic Insights

— 1 min read

Researchers at Bohai University have developed an adaptive finite-time neural optimal control technique for stochastic nonlinear systems (SNSs) using reinforcement learning (RL). According to the study, the proposed method simplifies the control scheme and addresses the singularity problem that occurs in traditional backstepping optimal control techniques. The technique uses radial basis function neural networks (RBFNNs) to estimate uncertain functions in the system, ensuring stability and boundedness of all system signals within a finite time.

Key Takeaways:

  • The researchers propose an adaptive finite-time neural optimal control technique for SNSs using RL, which simplifies the control scheme and addresses the singularity problem.
  • The technique uses RBFNNs to estimate uncertain functions in the system, allowing for stability and boundedness of all system signals within a finite time.
  • The proposed method is applied to an intelligent ship autopilot simulation example, demonstrating the effectiveness of the theoretical results.
  • The controller design method proposed in this article can be used for other SNS control applications, including mechanical systems, electrical systems, and power systems.
  • The research was supported by a Major Project of Education Department in Liaoning Province, China.
  • The study concludes that the proposed algorithm is efficient and effective for SNS control, and can be applied to other control applications.

Statistics:

  • The proposed control technique is capable of ensuring stability and boundedness of all system signals within a finite time.
  • The RBFNNs used in the technique can estimate uncertain functions in the system with high accuracy.
  • The intelligent ship autopilot simulation example demonstrated the effectiveness of the proposed algorithm in real-world scenarios.
  • The research was conducted by a team of researchers from Bohai University, including Huanqing Wang, Sijia Jia, and Siwen Liu.
  • The study was published in the International Journal of Robust and Nonlinear Control in 2025.

Sources:

  • VerticalNews - 2025 OCT 13
  • International Journal of Robust and Nonlinear Control - 2025
  • Bohai University - 2025
  • NewsRx - 2025 OCT 13

Read more