Vibration Measurement, Analysis and Dynamic Parameter Optimization for Delta Robot

Delta robots have been widely used in many manufacturing applications because of their high velocity and acceleration. Vibrations are one of the obstacles blocking them from further releasing all of their performance. Existing solutions include inserting a wait time between cycles, programming with smoother trajectory, introducing damp and limiting the jerk, even integrating compensation mechanisms. However, due to the lack of accurate robot-environment vibration model, typically those dynamic parameters are set empirically. Furthermore, the vibration of robot base/frame is not considered in the current literature. In this paper, an optimization procedure is proposed to solve the dynamic parameter tuning problem. An overlap based pick and place door shape trajectory generation method and jerk limited acceleration profile are used to generate smooth motion. By directly executing the pick and place motion with candidate parameters and measuring the vibration a comprehensive analysis and grid based constraint optimization can be performed to find the optimal dynamic parameters. The whole tuning technique is performed without modifying the existing hardware and software and it can be implemented with different smooth motion generation methods. The experimental results show the effectiveness of the proposed methodology.