Tellez Morales, J Abraham (Instituto Politécnico Nacional & INSA Centre Val de Loire)

Micro-Manipulator Control Using an Optimized PID Based on Neural Networks

Scheduled for presentation during the Regular Session "Robotics for Industry 4.0" (FrBT3), Friday, July 18, 2025, 14:00−14:20, Auditorium

Joint 10th IFAC Symposium on Mechatronic Systems and 14th Symposium on Robotics, July 15-18, 2025, Paris, France

This information is tentative and subject to change. Compiled on August 2, 2025

Abstract

This paper introduces a micro-positioning system using microscopy for visual feed- back, employing advanced methodologies to enhance precision and adaptability. The Denavit- Hartenberg method is utilized for systematic kinematic modeling of the robotic system, pro- viding a solid framework for motion control. An optimized PID controller is implemented, fine- tuning proportional, integral, and derivative gains through iterative optimization to minimize trajectory tracking errors. For smooth motion planning, the system generates trajectories using cubic polynomial equations, ensuring seamless operation. Furthermore, a neural network trained on synthetic datasets predicts inverse kinematics, significantly improving accuracy in nonlinear tasks. Experimental results validate the system’s effectiveness, demonstrating robust performance in micro- and nano-manipulation applications, particularly in biomedical research and nanotechnology processes.