Scholarlink

<p>This research addresses a critical challenge in achieving high precision for non-linear control of robotic systems, where model uncertainty makes it challenging to define accurate mathematical representations. To overcome this issue, this paper proposes an innovative dual neural network structure called the Brain Emotion Nesting Network (BENN), designed to enhance the performance of robot control systems. The BENN structure integrates two complementary sub-systems: the Brain Emotional Learning Network (BEL) and the Cerebellar Model Articulation Control Network (CMAC). Both sub-systems use the signed distance between the actual system tracking errors and the sliding surface as their input, enabling the controller to adapt effectively in uncertain environments. System parameters are designed based on the Lyapunov stability to ensure the overall stability of the BENN controller. The proposed approach is experimentally validated on a five-bar parallel robotic system and compared with conventional CMAC control methods. Experimental results demonstrate that the proposed BENN achieves superior trajectory tracking performance, with errors at least three times lower (0.0004 vs. 0.0012) than existing methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive Position Tracking Controller for Uncertain Parallel Robot System Based on Hybrid Brain Emotion and Cerebellar Model Articulation Control Network

  • 期刊论文

  • 发表日期：

    2026年1月1日

查看全文
  • Thanh Hai Tran,
  • Thanh Quyen Ngo,
  • Bui Thi Cam Quynh,
  • Phan Minh Than,
  • Tong Tan Hoa Le

摘要

This research addresses a critical challenge in achieving high precision for non-linear control of robotic systems, where model uncertainty makes it challenging to define accurate mathematical representations. To overcome this issue, this paper proposes an innovative dual neural network structure called the Brain Emotion Nesting Network (BENN), designed to enhance the performance of robot control systems. The BENN structure integrates two complementary sub-systems: the Brain Emotional Learning Network (BEL) and the Cerebellar Model Articulation Control Network (CMAC). Both sub-systems use the signed distance between the actual system tracking errors and the sliding surface as their input, enabling the controller to adapt effectively in uncertain environments. System parameters are designed based on the Lyapunov stability to ensure the overall stability of the BENN controller. The proposed approach is experimentally validated on a five-bar parallel robotic system and compared with conventional CMAC control methods. Experimental results demonstrate that the proposed BENN achieves superior trajectory tracking performance, with errors at least three times lower (0.0004 vs. 0.0012) than existing methods.