Wireless transmission grating reading head based on flexible circuit

Abstract

The invention relates to the technical field of displacement measurement, and discloses a wireless transmission grating reading head based on a flexible circuit, which comprises a front-end sensing part, a flexible transmission part and a rear-end main control processing part. The front end is integrated with a high-power LED photoelectric sensor and an inertial measurement unit, is connected with the rear end through a flexible wire harness, and can transmit analog photoelectric signals and digital attitude signals at the same time, and the rear end high-performance MCU executes multi-source signal synchronous acquisition, calculates an inclination angle by using inertial data so as to perform Abbe error dynamic compensation and generate a correction displacement value; meanwhile, on the basis of a Kalman filtering model and a dynamic threshold judgment mechanism, an inertial parameter compressed packet or a full-amount correction data packet is adaptively selected to be sent according to an observation residual error. According to the invention, a flexible structure is utilized to break through the limitation of a traditional rigid cable, the problems of narrow space installation, cable mechanical damping interference and dynamic attitude error correction are solved, and the wireless bandwidth occupation is reduced while the measurement precision is ensured.

Description

Technical Field

[0001] This invention relates to the field of displacement measurement technology, specifically to a wireless transmission grating reading head based on flexible lines. Background Technology

[0002] As a high-precision digital metrology and control device, the optical grating ruler displacement sensor is widely used in various intelligent precision geometric measurement instruments. Its working principle typically involves a high-power infrared LED inside the reading head emitting a light beam. This beam illuminates the grating glass, producing diffraction and interference patterns. The photoelectric sensor receives the interference light signal and converts it into an electrical signal, which, after amplification, filtering, and digital processing, reflects the change in the measured length. With the development of measurement technology, applications such as coordinate measuring machine probes place higher demands on multi-degree-of-freedom high-precision measurement and equipment integration. This often requires integrating multiple measurement units in extremely small physical dimensions, and the internal space is insufficient to accommodate complex traditional wiring.

[0003] However, existing grating readheads typically employ discrete metal housings to ensure structural strength and connect to external controllers via thick-diameter rigid shielded cables. This traditional structure results in a large overall size of the readhead, making it difficult to deploy flexibly in confined spaces. Furthermore, the rigid shielded cables themselves have significant bending stiffness and minimum bending radius limitations. When the readhead reciprocates at high frequency with the measured component, the mechanical traction force generated by the cable and the varying damping exert a considerable disturbance force on the moving component. This stress accumulation directly affects the positioning accuracy and repeatability of the precision motion platform.

[0004] Furthermore, the measurement accuracy of traditional readheads heavily relies on rigid mechanical mounting references, with physical clamps strictly constraining the gap and parallelism between the readhead and the grating ruler. If the metal casing is removed to adapt to space constraints or a non-rigid mounting method (such as direct adhesive bonding to a flexible circuit board) is adopted, the readhead is highly susceptible to slight attitude deflections caused by microscopic unevenness of the mounting surface or external vibrations during movement. Existing readhead circuit architectures typically only process single photoelectric displacement signals, lacking the ability to sense and compensate for changes in their own spatial attitude, and cannot eliminate the resulting Abbe error, leading to distorted measurement data under non-rigid connection conditions.

[0005] When attempting to solve cabling issues by introducing wireless transmission technology, existing technologies still face a contradiction between limited wireless bandwidth and the need for high-frequency precision measurement and massive data transmission. To ensure accurate reconstruction of motion trajectories, measurement systems typically require extremely high sampling frequencies. If full data is transmitted in real-time directly via a wireless link, channel congestion and transmission delays are highly likely, making it difficult to meet the stringent requirements of real-time measurement and synchronization in industrial settings. Summary of the Invention

[0006] Existing optical grating readout heads typically employ metal casings and rigid shielded cables, resulting in issues such as large size, limited installation space, and the impact of rigid cable traction on motion accuracy. Furthermore, traditional structures struggle to effectively compensate for dynamic geometric errors generated during installation or movement. To address these shortcomings, this invention provides a wireless transmission optical grating readout head based on flexible wiring, aiming to solve the problems of traditional readout heads, including deployment difficulties in confined spaces, high mechanical resistance, and the lack of dynamic error compensation mechanisms.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A wireless transmission grating readout head based on flexible lines adopts a split architecture design, including a front-end sensing part, a flexible transmission part, and a back-end main control processing part.

[0008] The front-end sensing part is integrated and mounted on the first circuit carrier board, including a high-power LED photoelectric sensor and a sensor; the high-power LED photoelectric sensor and the sensor integrate a high-power light-emitting diode for generating a measurement beam, a photoelectric conversion array for receiving grating moiré fringes and generating analog photoelectric signals, and a microelectromechanical system inertial measurement unit that is rigidly connected to the photoelectric conversion array and is used to generate digital attitude signals.

[0009] The flexible transmission section includes a flexible wiring harness that connects the front-end sensing section and the back-end main control processing section, and is used to simultaneously transmit the analog photoelectric signal and the digital attitude signal from the front-end sensing section.

[0010] The back-end main control processing section is set on the second circuit carrier board and includes a processor, which includes a signal processing module, a wireless transmission module and a high-performance MCU.

[0011] The high-performance MCU receives the analog photoelectric signal conditioned by the signal processing module and the digital attitude signal directly through the digital bus. After fusing the raw measurement data, it transmits the data through the wireless transmission module.

[0012] Preferably, the signal processing module mainly consists of an operational amplifier and a filter circuit, and its processing flow includes: receiving the analog photoelectric signal transmitted via the flexible wire harness; performing hardware-level amplification, shaping, and noise suppression on the signal to generate a conditioned analog signal; sampling the conditioned analog signal using the analog-to-digital converter integrated within the high-performance MCU, and outputting digital optical sampling values ​​for subdivision interpolation calculations.

[0013] Preferably, to eliminate the time phase difference of multi-source data, the high-performance MCU performs synchronous acquisition of multi-source signals: it uses an internal hardware timer to generate a hardware trigger signal at a fixed sampling frequency; this hardware trigger signal is simultaneously transmitted to the trigger input of the analog-to-digital converter and the transmission request terminal of the direct memory access controller. In response to the hardware trigger signal, the system controls the analog-to-digital converter to lock onto the conditioned analog signal at the same moment, and controls the direct memory access controller to read the output register of the microelectromechanical system inertial measurement unit, thereby acquiring a digital attitude signal containing triaxial acceleration and triaxial angular velocity data.

[0014] Preferably, to address the latency issues that may be introduced by wireless transmission, the high-performance MCU performs system initialization and clock synchronization operations: controlling the wireless transmission module to establish a communication link with the host computer and starting a precise time protocol; calculating network transmission latency by exchanging messages with hardware timestamps, and calculating the clock deviation value of the local clock relative to the host computer system clock; applying the clock deviation value to correct local timing, establishing a global time reference, and using the global time reference to stamp all collected data with a unified timestamp.

[0015] Preferably, to address the attitude errors that may be introduced by the flexible connection, the high-performance MCU performs a dynamic Abbe error compensation operation: extracting triaxial angular velocity data from the digital attitude signal and calculating the real-time tilt angle of the reading head relative to the measured plane; substituting the real-time tilt angle with the preset Abbe arm length parameter into the Abbe error compensation operation logic to calculate the total geometric error at the current moment; after obtaining the original grating count value based on the analog photoelectric signal, performing a correction operation to subtract the total geometric error from the original grating count value to generate a corrected displacement value.

[0016] Preferably, to optimize data transmission bandwidth, the high-performance MCU performs a data fusion prediction operation: extracting triaxial acceleration data from the digital attitude signal as the control input of the kinematic model; using the control input to run a Kalman filter algorithm, and employing state prediction logic to calculate the prior state estimate for the next moment, which includes the prior displacement prediction value and the prior velocity prediction value.

[0017] Preferably, the high-performance MCU performs the following transmission decision operation: calculating the absolute value of the difference between the prior displacement prediction value and the corrected displacement value to obtain the observation residual; simultaneously calculating a dynamic threshold based on the prior velocity prediction value; and comparing the observation residual with the dynamic threshold to determine the transmission mode. The dynamic threshold is defined as the product of the baseline tolerance constant when the system is stationary and the absolute value of the velocity influence coefficient and the prior velocity prediction value, to accommodate error tolerance at different motion speeds.

[0018] Preferably, the high-performance MCU performs data packaging operations based on the judgment result: when it is determined that the observation residual is less than or equal to the dynamic threshold, a first type of data packet is constructed, the payload of which only includes a global timestamp and the digital attitude signal, for the host computer to reconstruct the trajectory through the kinematic model; when it is determined that the observation residual is greater than the dynamic threshold, a second type of data packet is constructed, the payload of which includes a global timestamp, the corrected displacement value, the real-time velocity estimate, and the system status word, for full correction.

[0019] Preferably, the flexible wire harness is manufactured using flexible printed circuit board technology; the first circuit carrier is fixed to the surface of the moving part under test by adhesive bonding; the microelectromechanical system inertial measurement unit is configured to sense minute attitude changes of the front-end sensing part in real time.

[0020] Preferably, the back-end main control processing part further includes an auxiliary circuit, which includes a power management unit for converting the input power into a stable voltage and providing constant drive current to the high-power LED photoelectric sensor and the high-power light-emitting diode in the sensor, as well as powering the high-performance MCU and each sensor.

[0021] This invention provides a wireless transmission optical grating readout head based on flexible lines. It has the following beneficial effects: 1. This invention physically separates the front-end sensing part from the back-end main control processing part through a split architecture, and uses flexible wire harnesses to replace traditional rigid shielded cables. This structure eliminates the metal shell packaging, reduces the physical size and weight of the reading head, and makes it suitable for installation in confined spaces. The low bending stiffness of the flexible wire harness reduces the mechanical traction and damping interference of the cable on moving parts, and solves the problem of stress accumulation in rigid cables during high-frequency reciprocating motion, which affects the measurement accuracy.

[0022] 2. This invention achieves hardware-level synchronous acquisition of photoelectric signals and inertial attitude signals by directly integrating an inertial measurement unit at the front end. It uses inertial data to capture the attitude changes of the reading head caused by flexible installation or movement in real time, and calculates the geometric error by combining preset arm length parameters and corrects the grating count value. This method solves the measurement instability caused by the lack of rigid reference fixture constraints in split or adhesive installation, and ensures positioning accuracy under non-rigid connection conditions.

[0023] 3. This invention constructs a motion prediction model and transmission decision logic based on Kalman filtering. A dynamic threshold is calculated based on prior velocity prediction values, and the data transmission mode is switched by comparing the observed residuals. When the motion state conforms to the prediction model, only inertial parameters are sent to compress the data volume; when deviations occur, full correction data is sent. This method resolves the contradiction between limited wireless communication bandwidth and the large volume of high-frequency measurement data, reducing the load and latency risks of the wireless link while ensuring data refresh rate and trajectory reconstruction accuracy. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the structure of a novel grating reading head in one embodiment of the present invention; Figure 2 This is a block diagram illustrating the working principle of a grating reading head according to an embodiment of the present invention; Figure 3 This is a flowchart of a flexible split-type wireless grating measurement method based on inertial optical fusion prediction according to an embodiment of the present invention.

[0025] Among them, 110 is a high-power LED photoelectric sensor and sensor; 120 is a flexible wire harness; 130 is a processor; 131 is a signal processing module; 132 is a wireless transmission module; 133 is a high-performance MCU; and 134 is an auxiliary circuit. Detailed Implementation

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] See attached document Figure 1 This invention provides a wireless transmission grating readout head based on flexible wiring. This novel grating readout head adopts a split architecture design, primarily comprising a front-end sensing section, a flexible transmission section, and a back-end main control processing section. The front-end sensing section includes a high-power LED photoelectric sensor and sensor 110; the flexible transmission section includes a flexible wiring harness 120; and the back-end main control processing section includes a processor 130.

[0028] Specifically, a high-power LED photoelectric sensor and sensor 110 are integrated and mounted on a first circuit board, the size of which is configured to accommodate installation in confined spaces. The high-power LED photoelectric sensor and sensor 110 not only include a high-power light-emitting diode (LED) for generating a measurement beam and a photoelectric conversion array for receiving grating moiré fringe signals, but also integrate a microelectromechanical system (MEMS) inertial measurement unit. This inertial measurement unit is physically rigidly connected to the photoelectric conversion array and is used to sense microscopic attitude changes of the front-end sensing part in real time, including triaxial acceleration and triaxial angular velocity information. The first circuit board is fixed to the surface of the moving part being measured by adhesive or other non-mechanical fastening methods, eliminating the reliance on precision mechanical fixtures.

[0029] The flexible wiring harness 120 connects the front-end sensing section and the back-end main control processing section, and is manufactured using flexible printed circuit board (FPC) technology. The flexible wiring harness 120 is configured to simultaneously transmit analog photoelectric signals from the photoelectric conversion array and digital attitude signals from the inertial measurement unit. Compared to the circular shielded cables used in conventional grating readout heads in the prior art, the flexible wiring harness 120 has an extremely thin thickness and a very small bending radius, enabling it to adapt to high-frequency reciprocating motion without generating significant mechanical damping.

[0030] The processor 130 is mounted on a second circuit board, which has a larger area than the first circuit board and is used to house complex signal processing circuitry. Specifically, the processor 130 includes a signal processing module 131, a wireless transmission module 132, a high-performance MCU 133, and auxiliary circuitry 134. The signal processing module 131 is located between the input of the flexible wire harness 120 and the high-performance MCU 133. It mainly consists of operational amplifiers and filter circuits, and is used to perform hardware-level amplification, shaping, and noise suppression of the weak analog photoelectric signals transmitted via the flexible wire harness 120 to match the input range of the analog-to-digital converter interface of the high-performance MCU 133.

[0031] The high-performance MCU133 is the core computing unit of the entire system, integrating an analog-to-digital converter (ADC), a floating-point unit, and a direct memory access controller. The high-performance MCU133 is configured to receive optical displacement signals processed by the signal processing module 131, and to directly receive inertial attitude data from the high-power LED photoelectric sensor and sensor 110 via a digital bus. The high-performance MCU133 internally runs an Abbe error compensation algorithm and an inertial-optical fusion prediction model to perform real-time correction and fusion processing on the raw measurement data.

[0032] The wireless transmission module 132 is connected to the high-performance MCU 133 via a communication interface, and is used to send the processed position data and status information to the host computer system. The wireless transmission module 132 supports a timestamp-based synchronous transmission protocol to eliminate the impact of random delays in the wireless communication link on multi-axis synchronous measurement. The auxiliary circuit 134 includes a power management unit and a peripheral debugging interface. The power management unit is used to convert the input power into a stable voltage, provide constant drive current to the high-power LEDs, and power the high-performance MCU 133 and each sensor.

[0033] In contrast, in existing technologies, traditional grating readout heads are typically encased in a large metal shell and connected to a bulky DB interface via a thick, circular shielded cable. Such traditional cables have significant physical limitations, requiring a bending radius typically greater than 20 mm, and their maximum effective transmission length is usually limited to less than 5 meters due to the characteristics of analog signal transmission, in order to avoid signal attenuation. In contrast, the novel grating readout head in this embodiment eliminates the metal shell encapsulation, utilizing the miniaturization of the high-power LED photoelectric sensor and sensor 110, and the high flexibility of the flexible cable harness 120, overcoming the aforementioned physical limitations and enabling flexible deployment and long-distance wireless measurement in confined spaces.

[0034] See attached document Figure 2 This diagram illustrates the complete signal chain from the measured target to the displacement output. In the optical path, a light beam emitted by a high-power LED photoelectric sensor and the light source in sensor 110 passes through a collimating lens and is directed towards a grating ruler, generating diffraction and interference fringes. The beam is then projected onto a four-quadrant detector or photoelectric array via a reflector. In the circuit section, the optical signal is converted into a weak electrical signal and preprocessed in signal processing module 131 (labeled as signal conditioning in the diagram). Unlike existing technologies where the signal is directly converted to a digital signal (A / D) for displacement output, this embodiment introduces a data fusion layer after A / D conversion. MCU 133 not only uses its integrated ADC to read the sampled optical analog signal values ​​conditioned by signal processing module 131 (for subsequent software subdivision interpolation calculations), but also simultaneously reads data from the inertial sensor at the front end. After fusing the optical and inertial measurement values, the final displacement result is output via wireless transmission module 132. Auxiliary circuit 134 ensures the stability of the light source brightness and the continuity of circuit operation during this process. The figure shows in detail the complete signal conversion link from physical displacement input to final displacement output. This link is logically divided into an optical part and an electrical part, and is specifically implemented by the front-end sensing part, the flexible transmission part and the back-end main control processing part of this invention.

[0035] In the optical section, external displacement input acts on the measurement system. The light source and collimation module correspond to the emitting end of the high-power LED photoelectric sensor and sensor 110 in this embodiment, which includes a high-power LED and a lens module. The light beam emitted by the high-power LED is collimated by the lens module and then projected onto the external steel strip grating ruler. The external steel strip grating ruler is specifically composed of a steel strip grating ruler substrate and a metal coating on its surface. The relative motion between the light beam and the grating ruler generates a moiré fringe effect, thereby modulating and converting the mechanical displacement signal into an optical signal carrying phase information.

[0036] In the electrical section, the aforementioned optical signal is transmitted to the photoelectric conversion module, which corresponds to the high-power LED photoelectric sensor and the receiving end of sensor 110, as well as the signal processing module 131 in this embodiment. Specifically, the photoelectric conversion module includes a four-quadrant detector and a signal processing unit. The four-quadrant detector receives the optical signal from the external steel strip grating ruler and converts it into a weak electrical signal, which is then sent to the signal processing unit (i.e., signal processing module 131) for pre-amplification and filtering conditioning.

[0037] The processed signal enters the data processing module, which in this invention is mainly implemented by the high-performance MCU133 and its internal logic. The data processing module shown in the figure includes a subdivision circuit and an analog-to-digital converter (ADC). In a preferred embodiment, the ADC integrated within the high-performance MCU133 performs the analog-to-digital conversion, achieving a fundamental transformation from optical signals to electrical signals (digital domain). The subdivision circuit's function is primarily achieved through a high-order interpolation algorithm running within the high-performance MCU133. Notably, before generating the displacement output, the data processing module combines attitude data from the high-power LED photoelectric sensor and the inertial measurement unit in sensor 110 to perform Abbe error compensation and data fusion, ultimately outputting a high-precision displacement output via a wireless transmission module.

[0038] See attached document Figure 3 This invention provides a flexible, split-type wireless grating measurement method based on inertial optical fusion prediction. This method operates based on the aforementioned novel grating readhead system and includes the following steps: S10, Perform system initialization and clock synchronization operations: After the system is powered on, the auxiliary circuit 134 supplies power to each sensor and processor 130. The high-performance MCU 133 controls the wireless transmission module 132 to establish a communication link with the host computer and runs a precision time protocol to calibrate the deviation between the local hardware clock and the host computer system clock, and establish a global time reference that is uniformly followed by each node of the system. S20, perform synchronous acquisition of multi-source signals: at each sampling moment triggered by the system clock, the front-end sensing part, based on the global time reference, synchronously acquires the optical displacement signal modulated by the grating and the inertial motion signal characterizing the spatial attitude of the reading head through the high-power LED photoelectric sensor and sensor 110. After signal processing and analog-to-digital conversion, the original measurement dataset containing the original grating count value, triaxial acceleration and triaxial angular velocity is generated. S30, perform Abbe error dynamic compensation operation: The high-performance MCU133 receives the original measurement dataset through the flexible wire harness 120, uses the triaxial angular velocity in it to calculate the real-time tilt angle of the reading head relative to the measured plane, and calculates the geometric error by combining the preset Abbe arm length parameter. The geometric error is then subtracted from the original grating count value to generate a corrected displacement value that eliminates the micro-motion error of the flexible structure. S40, perform data fusion prediction and transmission decision operation: the high-performance MCU133 internal running status observation algorithm uses the triaxial acceleration in the original measurement dataset to make a priori estimation of the current position to obtain the predicted position, calculates the observation residual between the predicted position and the corrected displacement value, and dynamically selects the inertial parameter compression mode or the full data correction mode according to the magnitude of the residual, and constructs a wireless transmission data packet containing the corresponding payload and timestamp. S50 performs the receiver trajectory reconstruction operation: The host computer receives wireless transmission data packets through the wireless network, parses the absolute position coordinates or relative motion increments according to the data packet type identifier, restores the motion trajectory using an integral interpolation algorithm, and then strictly aligns the data from different measurement axes on the time axis according to the global time reference, outputting the final multi-dimensional spatial coordinates.

[0039] The following section will elaborate on the technical implementation details and principles of each of the above steps, using specific mathematical models, control logic, and signal processing algorithms.

[0040] During the system initialization and clock synchronization operation S10, this embodiment employs a Precision Time Protocol (PTP) based on the IEEE 1588 standard or a similar wireless time synchronization mechanism to address the nondeterministic delay problem generated by the wireless transmission module 132 during data transmission. Since the measurement system in this embodiment consists of a spatially separated novel grating readhead and a host computer, each node has an independent local clock source. Therefore, the system needs to establish a unified time reference to ensure that the receiving end can correctly align motion data from different axes.

[0041] Specifically, the process of establishing a global time base mainly includes three stages: establishing a synchronization link, calculating clock skew, and dynamic drift compensation, which can be further subdivided into the following steps: S101, Synchronization Link Establishment: After the new grating reading head is powered on, the high-performance MCU133 controls the wireless transmission module 132 to enter broadcast listening mode and search for the beacon signal of the host computer. After both parties successfully handshake and establish a communication connection, the high-performance MCU133 is configured as the slave clock node, the host computer is configured as the master clock node, and both parties start the synchronization message exchange sequence.

[0042] S102, Round-trip delay measurement and clock skew calculation: The master and slave nodes measure the network transmission delay and the deviation of the local clock relative to the master clock by exchanging messages with hardware timestamps. Specifically, the master clock node records the time t1 when sending the synchronization message, and the slave clock node records the time t2 when receiving the message; subsequently, the slave clock node records the time t3 when sending the delay request message, and the master clock node records the time t4 when receiving the message. All times t1 to t4 are time values ​​captured by the physical layer hardware of the wireless transmission module 132.

[0043] The high-performance MCU133 performs clock skew calculation based on the four acquired timestamps. This calculation process does not rely on complex mathematical formulas but is based on the link symmetry assumption, which assumes that the physical transmission times of the uplink and downlink are equal. The high-performance MCU133 first calculates the downlink time difference (i.e., receive time t2 minus transmit time t1) and the uplink time difference (i.e., receive time t4 minus transmit time t3); then it calculates the difference between these two time differences and divides this difference by two to obtain the time skew value of the local clock relative to the master clock. This calculation logic eliminates the impact of physical transmission delay on time synchronization, separating the simple clock phase deviation.

[0044] S103, Local Clock Calibration and Global Time Base Establishment: The high-performance MCU133 uses the calculated time deviation value to calibrate the local time. The specific calibration method can be hardware step adjustment by adjusting the counter register of the high-performance MCU133's internal timer, or maintaining a software-level global time deviation variable, which is automatically added to the timestamp of each sensor data read. The calibrated local time is then defined as the global time base for that measurement node. Under this base, the time axis of all sensor nodes in the system is unified.

[0045] S104, Continuous Monitoring and Dynamic Compensation of Frequency Drift: Considering that the local clock source will experience frequency drift due to temperature changes, a one-time initial synchronization is insufficient to maintain long-term measurement consistency. Therefore, the high-performance MCU133 is configured to periodically repeat the above S102 step, for example, once per second. The high-performance MCU133 records the changing trend of the time deviation values ​​measured multiple times consecutively, estimates the frequency drift rate of the local clock relative to the master clock, and dynamically adjusts the compensation coefficient of the local timing cycle based on this drift rate to offset the cumulative error caused by clock source temperature drift, ensuring the long-term stability of the global time reference.

[0046] For details regarding the specific communication message formats and handshake protocol involved in the above steps, those skilled in the art can refer to the IEEE 1588 standard document and related technical manuals on wireless sensor networks, as these are well-known technologies in the field and will not be elaborated upon here. This embodiment ensures, through the above steps, that each set of data collected by the high-power LED photoelectric sensor and sensor 110 carries a unified time tag, providing a time dimension alignment reference for subsequent multi-axis data fusion.

[0047] In existing technologies, the installation process of traditional grating readheads is typically subject to strict physical constraints. Specifically, traditional installation requires a physical "reference clamp" inserted into the narrow gap between the readhead and the grating ruler to physically force the setting of the readhead's clearance and parallelism relative to the ruler. Furthermore, traditional readheads have stringent tolerance requirements for installation posture; for example, the allowable deviation range for pitch is typically only ±1 degree, and for yaw, only ±0.5 degrees, with lateral offset controlled at the micrometer level. Simultaneously, traditional cable connection methods also have limitations. Their cable bundles typically have large bending radii (e.g., greater than 20mm), making flexible arrangement difficult in tight spaces, and the elastic stress of the cables themselves can easily compromise the positional accuracy after installation.

[0048] Unlike the traditional methods described above, this embodiment employs a digital installation process that no longer relies on physical reference fixtures but instead achieves a "virtual reference" through software and hardware interaction. The specific process includes: the host computer pre-setting "desired installation angle parameters" based on the actual installation environment (such as the tilt of the guide rail), including the desired pitch angle, roll angle, and yaw angle; after the new type of optical grating reading head is powered on, its high-power LED photoelectric sensor and sensor 110 collect its own attitude information in real time and transmit it to the host computer via the wireless transmission module 132; after receiving the data, the host computer compares and calculates the real-time attitude information with the preset desired installation angle parameters and generates visual adjustment instructions (e.g., displaying a 3D attitude model and prompts such as "increase roll angle" on the screen); the installer fine-tunes the angle of the reading head according to the instructions fed back from the host computer screen until the real-time attitude data falls within the preset tolerance range.

[0049] To support the above installation process and subsequent high-precision measurements, this embodiment employs hardware-level multi-source signal synchronous acquisition logic, namely step S201. Specifically, the high-performance MCU133 is internally configured with a high-precision hardware timer, which is set to generate a trigger signal at a fixed sampling frequency (e.g., 5kHz). This trigger signal is configured to be simultaneously connected to the trigger input of the analog-to-digital converter (ADC) of the high-performance MCU133 and the transfer request terminal of the direct memory access controller (DMA).

[0050] At each timer trigger, the ADC immediately initiates conversion, sampling and holding the analog photoelectric signal from the signal processing module 131 to obtain the grating phase information at that moment. Simultaneously, the DMA controller, without CPU intervention, automatically reads the output registers of the high-power LED photoelectric sensor and the inertial measurement unit in sensor 110 via the I2C or SPI digital bus to acquire the triaxial acceleration and triaxial angular velocity data at the same time. The high-performance MCU 133 then combines the optical data converted by the ADC and the inertial data transferred by the DMA into a single data frame in its interrupt service routine, and timestamps this data frame using the global time reference established in step S10. This hardware-triggered parallel acquisition mechanism eliminates time jitter caused by software polling, ensuring strict alignment of optical displacement data and inertial attitude data in the time dimension, providing a zero-phase-difference input source for subsequent data fusion algorithms.

[0051] Through the above steps S201 and the digital installation interaction process, this embodiment utilizes the real-time sensing capability of the high-power LED photoelectric sensor and sensor 110. On the one hand, it replaces the physical reference fixture during the installation stage, allowing installers to achieve high-precision alignment without mechanical reference. On the other hand, it provides accurate and synchronized attitude data during the measurement stage, enabling the system to effectively distinguish between the real displacement generated by the movement of the grating ruler and the false displacement generated by the vibration of the reading head.

[0052] During the Abbe error dynamic compensation operation S30, this embodiment introduces a real-time geometric correction mechanism based on inertial data to address the inherent physical characteristics of the flexible wire harness 120 connection used in the novel grating reading head. Because the flexible wire harness 120 has lower mechanical stiffness than traditional rigid shielded cables, the high-power LED photoelectric sensor and sensor 110 at the front end are easily affected by minor unevenness or straightness errors on the guide rail surface during high-speed reciprocating motion, resulting in slight pitch or yaw oscillations. Although this flexible connection significantly reduces motion resistance, without compensation, these oscillations can cause the measured light spot to deviate from the theoretical measurement baseline, thus generating Abbe error. Therefore, this embodiment achieves high-precision dynamic compensation through the following steps: S301, a dynamic geometric error model under flexible connection is established. The high-performance MCU133 has a pre-built local coordinate system with the sensing center of the photoelectric conversion array as the origin. In this model, the high-power LED photoelectric sensor and sensor 110 are regarded as a rigid body. The system identifies two main types of geometric error components: one is the longitudinal Abbe error caused by the rotation of the reading head around the horizontal axis (i.e., pitch angle change), and the other is the lateral Abbe error caused by the rotation of the reading head around the vertical axis (i.e., yaw angle change).

[0053] To accurately acquire the aforementioned angle changes, the high-performance MCU133 executes an attitude calculation algorithm (such as a complementary filter algorithm or an extended Kalman filter algorithm). This algorithm receives triaxial angular velocity and triaxial acceleration data from the inertial measurement unit. For pitch angle... The algorithm uses the gravity vector sensed by the accelerometer to correct the drift caused by the integral of the angular velocity in real time; for the yaw angle The algorithm calculates the relative angle change using short-time angular velocity integrals. The high-performance MCU133 ultimately outputs the current pitch angle, updated at high frequency. and yaw angle .

[0054] S302, Define the Abbe arm length parameter and apply the error calculation formula. To quantify the specific impact of angle changes on displacement measurement, this embodiment presets the Abbe arm length parameter in the non-volatile memory of the high-performance MCU133. This parameter includes the longitudinal Abbe arm length. And the lateral Abbe arm length .in, Defined as the vertical distance from the sensing center of the photoelectric conversion array to the axis of motion being measured. Defined as the horizontal lateral distance from the sensing center to the measured motion axis. These parameters can be written through factory calibration or configured via a host computer during installation. The high-performance MCU133 uses the Abbe error compensation formula to calculate the total geometric error at the current sampling moment. The Abbe error compensation formula is: ; in, This represents the calculated Abbe error compensation value, with units consistent with the grating displacement measurement units (e.g., micrometers). The length of the longitudinal Abbe arm represents the projected distance from the measurement point to the motion reference axis in the vertical direction; this parameter is a fixed constant. The length of the lateral Abbe arm represents the projected distance from the measurement point to the motion reference axis in the horizontal direction; this parameter is a fixed constant. Represents the pitch angle measured in real time. The sine value, where It is the angle by which the reading head rotates around the Y-axis of the local coordinate system; Represents the yaw angle measured in real time. The sine value, where It is the angle by which the reading head rotates around the Z-axis of the local coordinate system; This overall term represents the longitudinal displacement deviation component caused by pitch motion; This overall term represents the lateral displacement deviation component caused by yaw motion.

[0055] S303 performs real-time correction of the original displacement value. The high-performance MCU133 performs ADC sampling and subdivision calculations on the analog signal output from the signal processing module 131 to obtain the original grating count value. Subsequently, instead of directly using it as the final result, it is immediately combined with the total geometric error calculated above. Perform algebraic correction. The corrected true displacement value. The following modified formula is used to derive: ; The specific correction calculations are performed at the hardware instruction level in the floating-point unit (FPU) of the high-performance MCU133, ensuring that the calculation delay of the correction process is much smaller than the system sampling period. Through this step, this embodiment eliminates geometric errors caused by the micro-motion of the flexible structure in real time, thereby ensuring the accuracy of the measurement data while maintaining the convenience of flexible wiring.

[0056] During the execution of the inertial-optical fusion prediction and adaptive transmission decision operation S40, this embodiment constructs a kinematic prediction model based on Kalman filtering within the high-performance MCU 133. This model uses frequently updated inertial data to predict the position forward and dynamically switches the transmission mode of the wireless transmission module 132 according to the prediction accuracy to balance data refresh rate and wireless bandwidth consumption. The specific implementation steps are as follows: S401, construct a position state prediction model based on Kalman filtering. High-performance MCU133 defines the system's state vector. This vector contains the current time. displacement estimate and speed estimates The system uses a high-power LED photoelectric sensor and sensor 110 to collect the acceleration values ​​of the motion axis. As a control input, the high-performance MCU133 uses a state prediction formula to calculate the prior state estimate for the next time step. The state prediction formula is as follows: ; in, Representative moment The prior displacement prediction value; Representative moment The prior velocity prediction value; Representative moment The posterior displacement estimate (i.e., the best estimate at the previous moment). Representative moment The posterior velocity estimate; The sampling time interval representing inertial data; Representative moment Real-time acceleration measurements along the measurement axis output by a high-power LED photoelectric sensor and sensor 110; matrix This is the state transition matrix; the matrix To control the input matrix.

[0057] S402, calculate the observation residuals and set the transmission mode decision threshold. After obtaining the prior displacement prediction values... Then, the high-performance MCU133 reads the actual optical displacement value at the current moment after correction in step S30. The high-performance MCU133 calculates the observation residuals between the two. ,Right now Subsequently, the high-performance MCU133 processed the observation residual. With preset dynamic threshold Compare them.

[0058] To accommodate different prediction error tolerances at varying motion speeds, the high-performance MCU133 predicts the speed based on the current prior velocity value. Real-time calculation of dynamic threshold The calculation uses a linear mapping relationship: .in, This is the reference tolerance constant when the system is at rest. This is the speed influence coefficient. This mechanism ensures that larger prediction errors are allowed at high speeds, while tightening the decision criteria for precise positioning at low speeds.

[0059] S403, executes data packing in inertial parameter compression mode (Mode A). When the calculated observation residuals... Less than or equal to the dynamic threshold This indicates that the current motion state conforms to the inertial prediction model. The high-performance MCU133 controls the wireless transmission module 132 to enter the inertial parameter compression mode. In this mode, the high-performance MCU133 constructs the first type of data packet, whose payload only includes: packet header identifier A, global timestamp, raw triaxial acceleration data, and raw triaxial angular velocity data. At this time, the wireless transmission module 132 suspends the transmission of absolute displacement values. The host computer at the receiving end uses the prediction model with parameters consistent with those in S401 to deduce the displacement trajectory based on the received acceleration data.

[0060] S404, performs data packing in full data correction mode (Mode B). This is done when the calculated observation residuals... Greater than the dynamic threshold This indicates a significant deviation in the inertial prediction model. The high-performance MCU133 immediately controls the wireless transmission module 132 to switch to full data correction mode. In this mode, the high-performance MCU133 constructs a second type of data packet, whose payload includes: packet header identifier B, global timestamp, and corrected absolute optical displacement value. The data packet contains real-time velocity estimates and system status words. Upon receiving this type of data packet, the receiver stops prediction calculations and directly resets the trajectory tracking filter's state using the absolute optical displacement value within the packet.

[0061] In the process of performing receiver trajectory reconstruction and multi-axis synchronization operation S50, this embodiment not only solves the problem of data discontinuity at the algorithm level, but also changes the limitations of traditional physical connections at the system architecture level.

[0062] In existing technologies, traditional cable-type readout heads typically employ a point-to-point physical connection architecture. This means that each readout head must be connected to a separate physical interface of the client electronic device via a dedicated physical cable (usually containing shielding, power lines, and differential signal lines). This one-head-one-line-one-interface architecture necessitates stacking multiple corresponding signal processing boards or occupying multiple hardware channels to build a multi-axis measurement system (e.g., a coordinate measuring machine), resulting in high hardware costs and complex wiring. In contrast, this embodiment achieves wireless multi-axis integration based on a single-node host computer through the following steps S501 to S503.

[0063] S501 performs trajectory interpolation and reconstruction based on hybrid data packets. The host computer (receiver) maintains an independent Kalman filter state observer in memory for each registered new grating readhead. The mathematical model of this observer is consistent with the model running in the high-performance MCU133 at the transmitter. When the host computer receives a data packet through the wireless transmission module 132, it first parses the packet header identifier.

[0064] If the received data packet is in inertial parameter compression mode (Mode A), the host computer extracts the raw acceleration data from it. The system uses the reconstructed state from the previous moment to perform forward extrapolation. To ensure the continuity of the extrapolation, the host computer needs to calculate not only the displacement but also update the velocity state simultaneously. The host computer uses reconstruction formulas for calculation, which include displacement reconstruction formulas and velocity update formulas: ; ; in, Represents the current moment The calculated reconstructed displacement values; Represents the current moment Updated speed estimate; Represents the previous moment Determined displacement reconstruction values; Represents the previous moment The determined speed estimate; This represents the time interval between two consecutive data packets, and the value is determined by the difference in global timestamps between the current data packet and the previous data packet. This represents the acceleration data along the measurement direction carried in the current data packet; Represents the velocity increment caused by acceleration; This term represents the displacement increment component caused by acceleration.

[0065] If the received data packet is in full data correction mode (Mode B), the host computer directly extracts the corrected absolute optical shift value from it. And real-time velocity estimates, forcibly overriding the current observer's state variables. and This eliminates the cumulative drift that may be introduced by long-term integration.

[0066] S502 performs time-domain alignment and spatial coordinate synthesis of multi-axis data. In multi-axis simultaneous measurement scenarios (e.g., simultaneous movement of the X, Y, and Z axes), due to the communication mechanism of the wireless channel, there is a slight timing deviation in the arrival time of data packets from different axis reading heads to the host computer. To obtain accurate spatial coordinates, the host computer establishes a circular buffer with a time index.

[0067] The host computer sets a unified virtual rendering clock. For each measurement axis, the host computer searches the buffer for timestamps located at... Two adjacent reconstructed data points and (in Subsequently, the host computer uses a linear interpolation algorithm to calculate... The synchronous displacement value at time 1. This algorithm is based on... Compared to and The time ratio is used to calculate the displacement components at corresponding moments. This logic ensures that the final synthesized data is accurate. In spatial coordinates, all components correspond to the same physical moment, eliminating trajectory distortion caused by asynchronous sampling.

[0068] S503 analyzes the advantages of a multi-node wireless networking architecture. This embodiment adopts a star wireless network topology, unlike the traditional one-to-one physical connection. The host computer acts as the central node, managing multiple new optical grating readout heads simultaneously through the wireless protocol stack. Each readout head is assigned a unique logical ID upon joining the network.

[0069] Since data processing, error compensation (step S30), and transmission optimization (step S40) are all completed within the high-performance MCU133 at the reading head, the host computer only needs to receive the processed standardized data stream, significantly reducing the computational load. This allows a regular personal computer to connect to and parse data from multiple reading heads via a single wireless transceiver port using a time-division processing mechanism. The system only needs to add new node instances at the host computer software level to expand the number of measurement axes, without the need for additional signal acquisition cards or dedicated processors, thereby reducing the overall hardware cost and integration complexity of the multi-axis measurement system.

Claims

1. A wireless transmission optical grating reading head based on flexible lines, characterized in that, include: The front-end sensing part is integrated and installed on the first circuit carrier board. The front-end sensing part includes a high-power LED photoelectric sensor and a sensor (110). The high-power LED photoelectric sensor and the sensor (110) integrate a high-power light-emitting diode for generating a measurement beam, a photoelectric conversion array for receiving grating moiré fringes and generating analog photoelectric signals, and a microelectromechanical system inertial measurement unit that is rigidly connected to the photoelectric conversion array and is used to generate digital attitude signals. The flexible transmission section includes a flexible wire harness (120) connected between the front-end sensing section and the back-end main control processing section, and is configured to simultaneously transmit the analog photoelectric signal and the digital attitude signal from the front-end sensing section. The back-end main control processing section is set on the second circuit carrier board and includes a processor (130). The processor (130) includes a signal processing module (131), a wireless transmission module (132), and a high-performance MCU (133). The high-performance MCU (133) is configured to receive the analog photoelectric signal conditioned by the signal processing module (131), and to receive the digital attitude signal directly through the digital bus, and to transmit the original measurement data through the wireless transmission module (132) after performing fusion processing.

2. The wireless transmission grating reading head based on flexible lines according to claim 1, characterized in that, The signal processing module (131) mainly consists of an operational amplifier and a filter circuit. The processing flow of the signal processing module (131) is configured as follows: Receive the analog photoelectric signal transmitted via the flexible wire harness (120); The analog photoelectric signal is amplified, shaped, and noise suppressed at the hardware level to generate a conditioned analog signal; The conditioned analog signal is sampled using the analog-to-digital converter integrated within the high-performance MCU (133), and digital optical sample values ​​are output. These digital optical sample values ​​are used for subdivision interpolation operations.

3. The wireless transmission grating reading head based on flexible lines according to claim 2, characterized in that, The high-performance MCU (133) is configured to perform synchronous acquisition of multi-source signals as follows: The internal hardware timer generates a hardware trigger signal at a fixed sampling frequency. The hardware trigger signal is simultaneously transmitted to the trigger input terminal of the analog-to-digital converter and the transmission request terminal of the direct memory access controller of the high-performance MCU (133). In response to the hardware trigger signal, the analog-to-digital converter is controlled to lock the conditioned analog signal at the same time, and the direct memory access controller is controlled to read the output register of the microelectromechanical system inertial measurement unit, thereby obtaining the digital attitude signal containing triaxial acceleration and triaxial angular velocity data.

4. The wireless transmission grating reading head based on flexible lines according to claim 1, characterized in that, The high-performance MCU (133) is configured to perform system initialization and clock synchronization operations as follows: The wireless transmission module (132) is controlled to establish a communication link with the host computer and start the precision time protocol; Network transmission delay is measured by exchanging messages with hardware timestamps, and the clock deviation value of the local clock relative to the host computer system clock is calculated. The local timing is corrected using the clock deviation value to establish a global time base, and the collected data is then stamped with a unified timestamp using the global time base.

5. A wireless transmission grating reading head based on flexible lines according to claim 1, characterized in that, The high-performance MCU (133) is configured to perform Abbe error dynamic compensation operation as follows: The triaxial angular velocity data is extracted from the digital attitude signal, and the real-time tilt angle of the reading head relative to the measured plane is calculated. The real-time tilt angle is combined with the preset Abbe arm length parameter and substituted into the Abbe error compensation formula to calculate the total geometric error at the current moment. After obtaining the original grating count value based on the analog photoelectric signal, the total geometric error is subtracted from the original grating count value using a correction formula to generate a corrected displacement value.

6. A wireless transmission grating reading head based on flexible lines according to claim 5, characterized in that, The high-performance MCU (133) is configured to perform data fusion prediction operations as follows: Three-axis acceleration data are extracted from the digital attitude signal and used as control inputs for the kinematic model. The algorithm based on Kalman filtering is run using the control input, and the prior state estimate for the next time step is calculated using the state prediction formula. Output the prior state estimate, which includes the prior displacement prediction and the prior velocity prediction.

7. A wireless transmission grating reading head based on flexible lines according to claim 6, characterized in that, The high-performance MCU (133) is configured to perform the transmission decision operation as follows: The absolute value of the difference between the predicted prior displacement value and the corrected displacement value is calculated to obtain the observation residual; Calculate the dynamic threshold based on the prior velocity prediction value; The transmission mode is determined by comparing the observed residual with the dynamic threshold. The dynamic threshold is defined as the product of the baseline tolerance constant when the system is stationary, the velocity influence coefficient, and the absolute value of the prior velocity prediction value.

8. A wireless transmission grating reading head based on flexible lines according to claim 7, characterized in that, The high-performance MCU (133) is configured to perform a data packaging operation based on the judgment result as follows: If the observation residual is determined to be less than or equal to the dynamic threshold, a first type of data packet is constructed, and the payload of the first type of data packet is configured to contain only the global timestamp and the digital attitude signal. If the observed residual is determined to be greater than the dynamic threshold, a second type of data packet is constructed. The payload of the second type of data packet is configured to include a global timestamp, the corrected displacement value, the real-time velocity estimate, and the system status word.

9. A wireless transmission grating reading head based on flexible lines according to claim 1, characterized in that, The flexible wire harness (120) is manufactured using flexible printed circuit board technology; the first circuit carrier is fixed to the surface of the moving part under test by adhesive bonding; the microelectromechanical system inertial measurement unit is configured to sense the micro-attitude changes of the front-end sensing part in real time.

10. A wireless transmission grating reading head based on flexible lines according to claim 1, characterized in that, The back-end main control processing part also includes an auxiliary circuit (134), which includes a power management unit. The power management unit is used to convert the input power into a stable voltage and provide constant driving current to the high-power LED photoelectric sensor and the high-power light-emitting diode in the sensor (110), and to supply power to the high-performance MCU (133) and each sensor.

Images