This paper presents an enhanced differential plane mirror interferometer with high resolution for measuring straightness. Two sets of space symmetrical beams are used to travel through the measurement and reference arms of the straightness interferometer, which contains three specific optical devices: a Koster prism, a wedge prism assembly, and a wedge mirror assembly. Changes in the optical path in the interferometer arms caused by straightness are differential and converted into phase shift through a particular interferometer system. The interferometric beams have a completely common path and space symmetrical measurement structure. The crosstalk of the Abbe error caused by pitch, yaw, and roll angle is avoided. The dead path error is minimized, which greatly enhances the stability and accuracy of the measurement. A measurement resolution of 17.5 nm is achieved. The experimental results fit well with the theoretical analysis.

Capacitive sensors are widely used in industrial applications, such as CNC machine tools, where reliable positioning in the micrometer range with nanometer accuracy is required. Hence, these sensors are operated in harsh industrial environments. The accuracy of these sensors is mainly limited by slope errors and nonlinearities. In practice, the required accuracy of these sensors is achieved by a calibration against a metrological high-quality reference such as interferometric displacement measurement systems. This usually involves the use of high-order polynomials as calibration functions based on empirical data. In metrology, this is only the second-best approach and has disadvantages in terms of stability over the measurement range of the instrument. In addition, the validity of these empirical calibrations over time is questionable, and the associated uncertainty can only be roughly estimated. This makes regular recalibration of such sensors at short intervals mandatory to ensure the reliability of the displacement measurement. In this paper, we report on our investigations of the different parameters that affect the accuracy of capacitive sensors. Since the capacitance of these sensors results from the electric fields that build up between the electrodes, these field lines are calculated using FEM simulation models for typical commercial sensors. In the following the influence of various geometric parameters such as edge radius, guard ring size and shape, or thickness of the electrodes are individually analyzed according to their impact on the accuracy of these sensors. Based on these simulations, the deviations of the capacitance as they arise for real detector geometries can then be compared with idealized, de facto unrealizable parallel plate capacitors. This methodology allows overall uncertainty of capacitive sensors to be decomposed into their individual components and sorted in terms of their contribution to the uncertainty budget. The individual FEM-based analysis then enables a systematic analysis of the sources of uncertainty and, thus, reveals possibilities to improve manufacturing processes for capacitive sensors, to put these sensors on a solid metrological basis, and to improve the performance of these displacement measurement systems in the long run, i.e., to provide better sensors for the application.

The charge controller can greatly improve the hysteresis of the piezoceramic without the sensor. It is widely used in open-loop controlled nanopositioning systems. However, conventional charge controllers are not suitable for synchronous high linear operation of multiple piezoelectric actuators. This paper proposes a noval charge control scheme which can be used for synchronous high linear operation of multiple piezoelectric actuators. The control scheme combines similar control and traditional charge control. We verify the feasibility of this control scheme through experiments. The trajectory deviation between the two piezoelectric actuators reaches 0.79% of the travel range, while their respective trajectory errors do not exceed 1.2%. Because the controller in this article has a simpler grounding configuration, it can also replace traditional charge controllers for high linearity operation. The experimental results show that the charge controller achieves 0.65% positioning error of the piezoelectric actuator with grounding configuration. In addition, a T-type resistor network has also been introduced into the charge controller, which enables low-frequency linear operation of piezoelectric actuators with small capacitance.

Surface defect detection is a critical task in the manufacturing industry to ensure product quality and machining efficiency. Image-based precise defect detection faces significant challenges due to defects lacking fixed shapes and the detection being heavily influenced by lighting conditions. Addressing the efficiency demands of defect detection algorithms, often deployed on embedded devices, and the highly imbalanced pixel ratio between foreground and background images, this paper introduces a multi-scale fusion staged U-shaped convolutional neural network (DEU-Net). The network provides segmentation results for defect anomalies while indicating the probability of defect presence. It enables the model to train with fewer parameters, a crucial requirement for practical applications. The proposed model achieves an MIoU of 66.94 and an F1 score of 74.89 with lower Params (36.675) and Flops (19.714). Comparative analysis with FCN, U-Net, Deeplab v3+, U-Net++, Attention U-Net, and Trans U-Net demonstrates the superiority of the proposed approach in surface defect detection.

In order to solve the low coupling displacement and high crosstalk error of multi-dimensional micro-displacement stage, a two-dimensional micro-displacement stage based on flexure hinges and piezoelectric ceramics is designed and analyzed. The whole stage adopts a nested parallel structure. The design parameters of flexure hinges are calculated using Pseudo-rigid body model. Those parameters are confirmed by a simulation before manufacturing the stage. A double frequency laser interferometer is used for measuring the displacement and crosstalk error of the stage when it’s driven. The experimental results show that the maximum crosstalk error of the two-dimensional displacement platform within full travel is 0.98%, the displacement coupling rate approaches 91.4%, and the remaining trajectory deviation caused by hysteresis asymmetry can be reduced to 0.79%.