Optimized Sustainable Manufacturing Through Fuzzy Control in Image-Based Visual Servoing With Velocity and Field-of-View Constraints

The performance of image-based visual servoing (IBVS) in dynamic, high-speed, and high-precision applications is a major issue in sustainable and smart manufacturing systems. The proposed solution addresses the need for systematic optimization of control laws and constraint treatments in IBVS processes. Limited exploration of this topic is evident in the literature. Central to our approach is a smart fuzzy control-based scheme optimized for the sustainable and intelligent operation of robotic arms in manufacturing environments. The scheme incorporates a Mamdani fuzzy inference method for the adaptive adjustment of servoing gain, improving convergence and aligning with smart manufacturing principles. This method ensures precision and responsiveness, which are essential for high-speed and high-precision tasks. We address field-of-view constraints through an innovative online generation method of virtual features with a variable radius in the image space. The effectiveness of this approach, which synergizes sustainable manufacturing with smart, technology-driven solutions, is demonstrated through various comparative experiments. The experimental results show that the number of convergence iterations and the average initial velocity of the proposed method are reduced to 59% and 12%, respectively, of those of the existing methods on average; the optimization of the convergence efficiency and the continuity of the initial speed are obvious. Additionally, the maximum value of the vertical coordinate of the image is 1011 pixel, and has the best security performance.