Sign in Subscribe
Periodical publishing

Adaptive Finite-Time Optimal Control for Flexible-Joint Robots

Strategic Insights

— 1 min read

Researchers from Bohai University have developed a new adaptive finite-time neural optimal control technique for flexible-joint robots. The technique, which combines reinforcement learning and radial basis function neural networks, is designed to simplify the control process and solve the singularity problem that occurs in traditional backstepping optimal control techniques. According to the study, the proposed control scheme ensures the boundedness of all system signals within finite time and demonstrates practicability through a 2-link flexible-joint robot simulation example.

Key Takeaways:

  • The researchers propose an adaptive finite-time neural optimal control technique for flexible-joint robots using reinforcement learning and radial basis function neural networks.
  • The technique simplifies the control process and solves the singularity problem that occurs in traditional backstepping optimal control techniques.
  • The control scheme ensures the boundedness of all system signals within finite time.
  • The practicability of the designed control scheme is demonstrated through a 2-link flexible-joint robot simulation example.
  • The researchers suggest that the proposed control technique can be applied to a variety of robots with flexible joints.
  • The study has been peer-reviewed and published in the Nonlinear Dynamics journal.

Statistics:

  • The proposed control technique combines reinforcement learning and radial basis function neural networks to estimate uncertain functions in the considered system.
  • The technique simplifies the control process by releasing two preconditions: persistent excitation and known dynamics.
  • The controller design method proposed in the article not only simplifies the control way but also solves the singularity problem that occurs in the traditional backstepping optimal control techniques.
  • The study uses a 2-link flexible-joint robot simulation example to demonstrate the practicability of the designed control scheme.
  • The researchers receive financial support from the National Natural Science Foundation of China, LiaoNing Revitalization Talents Program, Natural Science Foundation of Liaoning Province, and Major Project of Education Department in Liaoning Province.

Sources:

  • Zhang, et al. (2025). Adaptive Finite-time Optimal Control for Flexible-joint Robots Via an Identifier-critic-actor Reinforcement Learning Algorithm. Nonlinear Dynamics, 2025.
  • Springer (publisher). Van Godewijckstraat 30, 3311 GZ Dordrecht, Netherlands. (www.springer.com)
  • NewsRx (2025). Studies from Bohai University Describe New Findings in Robotics. Journal of Engineering, October 13, 2025; p 4418.

Read more