Research on Curvature Adaptive Control Method for ABB Robot Blade Polishing Residuals

Abstract

To address challenges in traditional blade polishing such as low error control precision, irrational path planning, and poor parameter matching of belt pulleys, error control methods were investigated using ABB industrial robots as execution platforms and belt pulleys as polishing tools, combined with curvature adaptive algorithms. By analyzing blade surface curvature characteristics, a correlation model linking polishing errors to parameters and surface curvature was established. A curvature-adaptive error control algorithm was designed to dynamically adjust polishing parameters across different curvature zones. Simulation environments were built using Robot Studio platform to validate error control through both simulation and experimental testing. Results demonstrate that: In high-curvature zones, polishing errors reach 0.012mm at 1500r/min belt speed and 80N polishing force; In medium-curvature zones, errors minimize to 0.009mm at 1700r/min belt speed and 70N force; In low-curvature zones, errors decrease to 0.007mm at 2000r/min belt speed and 60N force. The curvature-adaptive control strategy achieves an average error of 0.011mm and a maximum error of 0.021mm, both remaining within the target error range of 0.1mm. Compared to traditional fixed-parameter methods, this approach improves allowance control precision by 61.1%. These findings provide theoretical support and technical references for automated high-precision blade polishing.