Adaptive Neural-PID Visual Servoing Tracking Control via Extreme Learning Machine

The vision-guided robot is intensively embedded in modern industry, but it still a challenge to track moving objects real time accurately. In this paper, hybrid adaptive control scheme combined with an Extreme Learning Machine (ELM) and proportional–integral–derivative (PID) proposed for dynamic visual tracking of the manipulator. extracts line features on image plane based laser-camera system determines optimal input guide robot, so that are aligned their desired positions. observation state–space equations first determined by analyzing motion camera object. then represented as autoregressive average extra (ARMAX) valid estimation model. predictor estimates online relevant 3D parameters between object, which subsequently used calculate sensitivity neural network. ELM–PID controller designed adjustment parameters, was validated physical platform. experimental results showed method’s vision-tracking displayed superior performance pure P PID controllers.