Key points are not available for this paper at this time.
Introduction Robotic grinding suffers from problems such as unstable contact force, limited control in large-area grinding, and difficulty in improving the grinding quality of complex curved surfaces. Methods To address these issues and achieve precise force-position control, adaptive adjustment, and multi-objective balance during the grinding process, this study designs a rigid-flexible coupling grinding mechanism and a human-machine interactive architecture for an integrated grinding and measurement system. A force-position trajectory optimization strategy grounded in an improved non-dominated sorting genetic algorithm II is proposed, and an adaptive strain stiffness force-position control algorithm based on impedance control is designed for the integrated grinding and measurement system. Results Results show that the improved non-dominated sorting genetic algorithm II achieves an accuracy of approximately 0.95 after 100 iterations, with the loss function stabilizing below the order of 10-4 after 40 iterations, and the delay time stabilizing at approximately 5 ms. The integrated grinding and measurement adaptive control algorithm maintains a surface roughness below 1.2 μm under different working conditions, achieving a material removal rate of 0.62 mm 3 /s in working condition 2, with a single-cycle total energy consumption as low as 1.78 kW·h. Discussion The integrated grinding and measurement adaptive variable stiffness force-position scheme, which combines a human-machine interactive grinding system with a multi-objective optimization algorithm, is effective and significantly improves overall grinding performance. This research provides a more efficient and precise control solution for the field of robotic grinding, which can be applied to the grinding of materials such as stainless steel, meeting the industrial demand for high-precision and low-energy grinding, and promoting the development of grinding automation technology.
Yu et al. (Tue,) studied this question.