Scholarlink

<p>The design and control of Lower-Limb Exoskeletons (LLE) is a tedious task to researchers; without a properly designed controller, the control system would show poor step responses and steady-state error. Due to their simplicity, classical control methods are broadly applicable to biomedical systems; however, their effectiveness reduces when highly nonlinear, fixed-gain, time-varying human-robot interaction dynamics exist, and in turn cause serious tracking errors. This manuscript suggests a hybrid optimization scheme which is a common combination of the Grey Wolf Optimization (GWO) and the Artificial Bee Colony (ABC) algorithm, termed GWBC, to optimize a Proportional-Integral-Derivative (PID) controller to a LLE platform. A femoral and tibial joint model based on two second-order transfer-function models which have been validated as being appropriately dynamic is used to model the behavior of both joints. Traditional PID calibration techniques, such as manual tuning, ZieglerNichols procedure, or auto-tuner of the MATLAB toolkit are typically not suitable in the current complex systems and require the use of modern meta-heuristic optimization algorithms. We have carried out a time based analysis to model the performance of the systems. The combined responses attained under the GWBC controller registers an average rise time of 0.130 ± 0.006s and 0.281 ± 0.021s for Femur and Tibia joint Transfer functions respectively. Such findings were possible due to the keen observation that GWO demonstrates fast convergence at the early phase, whereas ABC provides more stability. It is therefore seen that utilizing the complementary capabilities of the constituents, the hybrid GWBC framework provides enhanced transient responsiveness with reduced overshoot and lower steady-state error in both joint models. Thus, the resource of GWBC-based PID tuning approach becomes a more efficient and effective approach to controlling a robotic system. This enhanced performance of control is an indication of the real-life importance of this technology to rehabilitation exoskeletons and other robotic assistive gear requiring clear and stable motion control.</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

A hybrid grey wolf bee colony optimized PID controller for improved transient response and stability of lower limb exoskeleton system

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月9日

查看全文
  • Nirbhay Raut,
  • Ayush Karapagale,
  • Akshit Gangwar,
  • Harsh Goud,
  • Hari Narayan Bhargaw

摘要

The design and control of Lower-Limb Exoskeletons (LLE) is a tedious task to researchers; without a properly designed controller, the control system would show poor step responses and steady-state error. Due to their simplicity, classical control methods are broadly applicable to biomedical systems; however, their effectiveness reduces when highly nonlinear, fixed-gain, time-varying human-robot interaction dynamics exist, and in turn cause serious tracking errors. This manuscript suggests a hybrid optimization scheme which is a common combination of the Grey Wolf Optimization (GWO) and the Artificial Bee Colony (ABC) algorithm, termed GWBC, to optimize a Proportional-Integral-Derivative (PID) controller to a LLE platform. A femoral and tibial joint model based on two second-order transfer-function models which have been validated as being appropriately dynamic is used to model the behavior of both joints. Traditional PID calibration techniques, such as manual tuning, ZieglerNichols procedure, or auto-tuner of the MATLAB toolkit are typically not suitable in the current complex systems and require the use of modern meta-heuristic optimization algorithms. We have carried out a time based analysis to model the performance of the systems. The combined responses attained under the GWBC controller registers an average rise time of 0.130 ± 0.006s and 0.281 ± 0.021s for Femur and Tibia joint Transfer functions respectively. Such findings were possible due to the keen observation that GWO demonstrates fast convergence at the early phase, whereas ABC provides more stability. It is therefore seen that utilizing the complementary capabilities of the constituents, the hybrid GWBC framework provides enhanced transient responsiveness with reduced overshoot and lower steady-state error in both joint models. Thus, the resource of GWBC-based PID tuning approach becomes a more efficient and effective approach to controlling a robotic system. This enhanced performance of control is an indication of the real-life importance of this technology to rehabilitation exoskeletons and other robotic assistive gear requiring clear and stable motion control.