RBF-NN과 I-PID 제어를 적용한 4륜 구동 선박 도장 로봇의 벽면 경로 추종에 관한 연구 (A Study on Wall Climbing of a Four-wheel Drive Ship Painting Robot Using RBF-NN and I-PID Control)

서지정보

· 발행기관 : 제어·로봇·시스템학회
· 수록지 정보 : 제어.로봇.시스템학회 논문지 / 30권 / 7호 / 740 ~ 749페이지
· 저자명 : 김동주, 김성재, 이문행, 성태경, 서진호

초록

Painting operations on ships present significant challenges and risks, highlighting the requirement for developing robotic systems that enhance the safety and precision of these operations. This study introduces an adaptive robust control scheme for wall-climbing robots designed for ship painting applications; this design integrates an intelligent proportional-integral- derivative (PID) controller with a radial basis function neural network. The proposed system overcomes the limitations of traditional PID controllers by effectively estimating complex nonlinear dynamics and uncertainties using time-delay information. Furthermore, the accuracy of path-tracking tasks is improved by compensating for errors in time-delay estimation using a neural network. Additionally, the stability of the control system is proven using a Lyapunov-based analytical method. The simulation is highly reliable because it used models that closely resemble actual ship-painting robots. The proposed controller achieves higher tracking performance than traditional PID controllers. The implications of this study extend beyond ship-painting automation, offering insights into other industrial domains in which the use of high-precision robotic systems is paramount.

영어초록

Painting operations on ships present significant challenges and risks, highlighting the requirement for developing robotic systems that enhance the safety and precision of these operations. This study introduces an adaptive robust control scheme for wall-climbing robots designed for ship painting applications; this design integrates an intelligent proportional-integral- derivative (PID) controller with a radial basis function neural network. The proposed system overcomes the limitations of traditional PID controllers by effectively estimating complex nonlinear dynamics and uncertainties using time-delay information. Furthermore, the accuracy of path-tracking tasks is improved by compensating for errors in time-delay estimation using a neural network. Additionally, the stability of the control system is proven using a Lyapunov-based analytical method. The simulation is highly reliable because it used models that closely resemble actual ship-painting robots. The proposed controller achieves higher tracking performance than traditional PID controllers. The implications of this study extend beyond ship-painting automation, offering insights into other industrial domains in which the use of high-precision robotic systems is paramount.

참고자료

· 없음