Fusion and decoupling of Spatial geometric errors in Five-Axis CNC machine tools based on unscented Kalman filter
摘要
Double ball bar (DBB) is a commonly used identification instrument in the measurement process of five axis machine tools, which is easy to use stably, but difficult to break through the measurement range of circles in space. This paper proposes an error fusion method based on double ball bar (DBB) and inertial mass unit (IMU). This method divides the spatial measurement hemisphere and uses DBB and IMU to measure on each formed node. Then, the unscented Kalman filter method is used for data fusion. The error after fusion is not significantly different from the error when using DBB alone, and any NURBS curve can be selected for error identification. After error decoupling verification, the data fused with errors has decouple ability and verifiability.