Adaptive Machine Learning Based Control Method for Visual Robot Manipulator System

The control method proposed in this study is based on adaptive machine learning for a multi-input multi-output visual robotic arm system with uncertain control gain matrices. By relaxing the requirement for known control gain matrices in traditional adaptive learning methods, a composite energy function is designed to prove the convergence of the system. The effectiveness of the adaptive iterative learning method is demonstrated through simulating the motion of the robotic arm system and using an uncalibrated camera for validation.