High-speed and high-precision control for SMT process under dual-network structure

High-speed and high-precision motion control remains a major challenge in Surface-Mount Technology (SMT), as conventional methods often fail to balance overshoot and settling time under varying task requirements. To address this challenge, a motion control architecture driven by task-specific performance indicators is developed. The proposed framework integrates adaptive position, velocity, and current loop algorithms with a multi-parameter cooperative regulation model, in which control parameters are dynamically updated via a bidirectional radial basis function neural network (BRBFNN) based iterative correction mechanism. Unlike approaches that optimize a single performance metric, the proposed dual-layer architecture enables real-time coordination between precision and response speed, allowing the control system to adapt to the heterogeneous demands of different components. Experimental validation was conducted on the Z-axis, recognized as the most demanding in terms of speed and precision. The results demonstrate that the method can effectively adjust control performance according to varying operational requirements. This study provides a feasible control solution for SMT equipment and offers a general framework for motion systems requiring simultaneous optimization of speed and precision.