Fused quartz substrate integrated laser radar transceiver optical module and adjustment method

Abstract

The application discloses a fused quartz substrate integrated laser radar transceiving optical module and an adjusting method, and belongs to the technical field of optical instruments. The module integrates all optical elements of the transmitting and receiving optical paths on the same fused quartz substrate, and configures a micro sensor and a micro adjusting device for each element. The system comprises a first control loop, a second control loop, a first central signal gating and routing unit, comparison judgment and arbitration logic. Through six pre-defined segment diagnosis modes, the system can accurately locate the misalignment segment in the optical path, and only drive the corresponding micro adjusting device to perform cooperative closed-loop compensation, realizing adaptive alignment from local fine tuning to global search. The system has the beneficial effects of self-maintenance in the whole life cycle, high efficiency and accuracy of adjustment, global optimization of the transceiving link, and significant reduction of the whole life cycle cost, and is particularly suitable for fields such as vehicle-mounted laser radars with high reliability requirements.

Description

Technical Field

[0001] This invention relates to the field of optical instrument technology, specifically to an integrated laser radar transceiver optical module and adjustment method based on a fused silica substrate. Background Technology

[0002] With its high precision and high resolution, lidar has become a core sensor in fields such as autonomous driving and 3D mapping. The performance and stability of its optical system directly determine the overall detection capability of the device.

[0003] Traditional lidar uses discrete optical components (such as discrete fiber lasers, collimating lenses, and receiving lenses) to build the optical path. These components are aligned and fixed one by one through mechanical structures, resulting in inherent drawbacks such as large size, complex assembly and adjustment, and high cost. More importantly, discrete structures are prone to micron-level relative displacement under harsh environmental stresses such as vibration and temperature cycling, leading to optical path misalignment, irreversible performance degradation, and poor long-term stability.

[0004] To address these issues, optical integration technology has become a development trend. For example, solutions using silicon photonic chips or planar optical circuits suffer from problems such as high insertion loss and limited laser power tolerance. Recently, significant progress has been made in integrated solutions based on fused silica substrates: by fabricating precision grooves and light guide channels on a single fused silica substrate, all core optical components of the transmitting and receiving optical paths (such as seed light source fibers, amplifier fibers, collimating lenses, receiving lenses, detectors, etc.) are pre-aligned and permanently glued in place. This solution effectively solves the volume and initial assembly problems of discrete structures and, by utilizing the extremely low coefficient of thermal expansion of fused silica, achieves mechanical and thermal stability far exceeding that of discrete structures.

[0005] However, after in-depth research and practice, the applicant discovered that even such advanced "passive" integrated solutions still have a fundamental technical limitation: all optical components form a rigid "optical black box" after encapsulation and curing. Although the initial alignment accuracy is high, once faced with long-term, significant environmental stress (such as continuous vibration and shock in automotive environments, and temperature cycling from -40°C to 105°C), the residual stress inside the material may slowly release, and the adhesive may also undergo extremely small creep. These factors can lead to slow drift between optical components at the nanometer to micrometer level. Because the module is completely sealed and cured, this minute performance degradation process cannot be detected in real time, let alone compensated or corrected within the system. As a result, the module's performance will slowly and irreversibly degrade throughout its entire lifespan, ultimately affecting the long-term reliability and ranging accuracy of the lidar. This poses a potentially significant risk for automotive-grade applications requiring high-reliability operation for more than 10 years.

[0006] Therefore, existing technologies (including advanced integrated solutions) still face a pressing technical problem: how to enable highly integrated lidar optical modules to not only achieve initial compactness and stability, but also to possess the "adaptive" capability to resist performance degradation throughout their entire lifecycle. That is, to sense the optical path status in real time without human intervention and dynamically maintain the optical alignment in the optimal state, thereby ensuring the ultimate reliability of the end product.

[0007] In view of the above, this application is hereby submitted. Summary of the Invention

[0008] To address the aforementioned problems in existing technologies, an integrated laser radar transceiver optical module and adjustment method based on a fused silica substrate are provided, with the aim of resolving at least one of the aforementioned problems.

[0009] The technical solution to achieve the purpose of this invention is as follows:

[0010] This invention provides an integrated lidar transceiver optical module on a fused silica substrate, comprising a transmitting optical path unit, a receiving optical path unit, an A / D converter, an adder counter, a D / A converter, a control comparator, an adjustment comparator, a clock generator, and an AND gate. The optical elements of the transmitting optical path unit include a seed light source fiber, a fiber amplifier fiber, and a transmitting collimating lens. The optical elements of the receiving optical path unit include a receiving lens, a detector fiber, and a photodetector. The optical elements of the transmitting and receiving optical path units are integrated onto a miniature adjustment device on the fused silica substrate in the order of the optical paths.

[0011] Corresponding miniature sensors are respectively set on the laser emitting side of the optical elements of the transmitting optical path unit and the receiving optical path unit;

[0012] An A / D converter, an adder counter, a D / A converter, and a control comparator are connected in sequence to form the first control loop; an adjustment comparator, a clock generator, and an AND gate are connected in sequence to form the second control loop.

[0013] The first control loop acquires laser information from the optical path interface. The first and second control loops control the micro-adjustment device to perform traversal adjustments within its adjustment range, and to collaboratively search, connect, optimize, and lock the optical path interfaces of the transmitting and receiving optical path units.

[0014] This invention also provides an adjustment method for an integrated lidar transceiver optical module on a fused silica substrate. This method, implemented using an integrated lidar transceiver optical module on a fused silica substrate, includes the following steps:

[0015] Step 1: After the system is powered on or receives a calibration command, it initializes each component; the main controller selects and activates one of the six segmented diagnostic modes according to the preset strategy; the first central signal gating and routing unit selects the corresponding combination of micro sensors according to the selected mode and outputs the first signal and the second signal with a time sequence relationship.

[0016] Step 2: Input the first signal into the adjustment comparator and compare it with the preset offset threshold; if it is determined to be offset, the adjustment comparator outputs an adjustment request signal; at the same time, the first control loop determines whether adjustment is allowed based on the system operating status, and activates the system adjustment enable clock generator when it is allowed.

[0017] Step 3: The control of the adjustment request signal and the system adjustment is executed by performing a logical "AND" operation on the clock pulse output by the clock generator in an AND gate; a valid adjustment execution command is generated only when both are valid simultaneously.

[0018] Step 4: Use the second signal as a feedback voltage input to control the comparator, compare it with the dynamic reference voltage at its non-inverting input terminal, and generate an error voltage signal; effectively adjust the execution command to control the multi-channel adjustment command distributor, and route the error voltage signal to the micro-adjustment device drive circuit corresponding to the current diagnostic mode;

[0019] Step 5: The micro-adjustment device drive circuit that receives the error voltage signal drives the corresponding micro-adjustment device to adjust the pose of the corresponding optical element; at the same time, the first control loop dynamically adjusts the dynamic reference voltage according to the feedback to form a closed-loop control until the error voltage signal meets the locking condition, so that the current diagnostic segment reaches the optimized alignment state.

[0020] Step Six: After optimizing the current diagnostic segment, switch to the next diagnostic mode. The system repeats steps one through five to adjust each optical path segment in turn. If the adjustment fails to converge in any mode, a larger-scale traversal search dominated by the second control loop is triggered.

[0021] Compared with the prior art, the beneficial effects of the present invention include:

[0022] (1) It achieves full lifecycle adaptive maintenance, fundamentally solving the long-term performance drift problem. Unlike the "optical black box" formed by the existing "passive fixed" integrated solution, this invention embeds a distributed micro-sensor network and micro-adjustment device in the fused silica substrate and constructs an intelligent closed-loop control system, enabling the module to sense the optical path status in real time and dynamically compensate for misalignment at the nanometer to micrometer level. The module can actively resist the slow performance decay caused by environmental factors such as vibration, temperature cycling, and material stress relaxation, ensuring that the core optical performance of the lidar is actively maintained at a high level throughout its entire service life (such as more than 10 years as required by automotive grade), achieving a fundamental leap from "inevitable performance decay" to "dynamic performance maintenance";

[0023] (2) A segmented diagnostic and precise collaborative adjustment architecture is proposed, which greatly improves adjustment efficiency and system stability. Addressing the problems of adjustment conflicts, oscillations, or getting trapped in local optima that are common in traditional multi-degree-of-freedom adjustment systems, this invention innovatively proposes six segmented diagnostic modes based on an optical path physical model. Through a first central signal gating and routing unit, two-level comparison and decision-making, and central arbitration logic, the system can accurately locate the specific segment of the transmitting or receiving optical path where the misalignment occurs (such as the "seed light source-amplifier" docking segment or the "lens-fiber" coupling segment), much like diagnosing circuit faults. Subsequently, adjustment commands are precisely distributed to 1-2 micro-adjustment devices related to that segment for collaborative compensation, avoiding the inefficiency, interference, and additional losses caused by "one-size-fits-all" global adjustment. This "precise positioning and collaborative adjustment" mode significantly improves calibration speed and accuracy, and ensures the overall stability of the system during the adjustment process.

[0024] (3) Intelligent closed-loop optimization of the entire transmission and reception chain was completed, ensuring the ultimate performance of the system. The adaptive control of this invention covers the entire optical chain from laser emission (seed light source, amplification, collimation) to laser echo reception (convergence, light guiding, detection). It not only ensures the optimal quality of the emitted beam, but also ensures that the sensitivity of the receiving optical path is always at its best. This global optimization of the transmission and reception chain ensures that the key indicators of the lidar, such as detection range, resolution, and signal-to-noise ratio, remain consistent and optimal throughout the entire life cycle and under complex working conditions, meeting the stringent requirements of high-level autonomous driving and other applications for the ultimate reliability of sensors;

[0025] (4) Advanced configurations such as heterogeneous integration were introduced, achieving a leap in performance while maintaining high reliability. In the preferred scheme, a silicon-based optoelectronic chip (integrating SPAD array and processing circuit) was heterogeneously integrated with a fused silica substrate, and a vertical grating coupler was used to achieve efficient optical signal switching. This not only retains the ultra-high thermomechanical stability of the fused silica platform, but also leverages the advantages of silicon materials in high-speed optoelectronic processing and integration, realizing "integrated sensing and computing". This design greatly shortens the transmission path of high-sensitivity analog signals, reduces noise, and improves system bandwidth and detection sensitivity, providing a hardware foundation for the next generation of high-performance lidar;

[0026] (5) Reduced manufacturing and lifecycle maintenance costs, enhancing product competitiveness; On the manufacturing side, the system's powerful self-calibration capability reduces the extreme demands on the absolute precision of initial assembly, improving production yield and efficiency. On the usage and maintenance side, the module's "online self-maintenance" and "self-repair" capabilities greatly reduce the need for downtime, return for calibration, or even replacement of the entire module due to performance degradation. Especially for large-scale commercial deployment scenarios (such as fleets), this significantly reduces long-term operating costs and total cost of ownership;

[0027] (6) Possessing high scalability and intelligent potential, reserving space for continuous technological upgrades; the architecture of this invention is a standardized intelligent optical platform. Its modular design allows performance to be improved by replacing or upgrading silicon photonic chips (such as integrated modulators, multi-wavelength processing units) that integrate more advanced algorithms or functions, without redesigning the core fused silica substrate and adaptive frame. At the same time, the rich sensor data provides a data foundation for advanced intelligent applications such as predictive maintenance and health status monitoring, extending the technology life cycle. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram of the system for adjusting the optical path using modules. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of 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, not all, of the embodiments of the present invention.

[0031] Therefore, the following detailed description of embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0032] It should be noted that, unless otherwise specified, the embodiments and features and technical solutions in the present invention can be combined with each other.

[0033] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0034] The present invention will be further described in detail below with reference to the embodiments.

[0035] like Figure 1 As shown, this invention provides an integrated lidar transceiver optical module on a fused silica substrate, which includes a transmitting optical path unit, a receiving optical path unit, an A / D converter, an adder counter, a D / A converter, a control comparator, an adjustment comparator, a clock generator, an AND gate, a micro-adjustment device, and a micro-sensor.

[0036] Among them, the micro-adjustment device includes a first micro-adjustment device, a second micro-adjustment device, a third micro-adjustment device, a fourth micro-adjustment device, a fifth micro-adjustment device, and a sixth micro-adjustment device; the micro-sensor includes a first micro-sensor, a second micro-sensor, a third micro-sensor, a fourth micro-sensor, a fifth micro-sensor, and a sixth micro-sensor;

[0037] The optical components of the transmitting optical path unit include a seed light source fiber, a fiber amplifier fiber, and a transmitting collimating lens; the optical components of the receiving optical path unit include a receiving lens, a detector fiber, and a photodetector.

[0038] A number of grooves are provided on the quartz substrate, and a first micro-adjustment device, a second micro-adjustment device, a third micro-adjustment device, a fourth micro-adjustment device, a fifth micro-adjustment device and a sixth micro-adjustment device are arranged on the grooves in the order of optical path connection.

[0039] The seed light source fiber, the fiber amplifier fiber, and the transmitting collimator are sequentially connected in optical paths on the corresponding grooves, and respectively pass through the first micro-adjustment device, the second micro-adjustment device, and the third micro-adjustment device to form the transmitting optical path unit of the first light guide channel; the laser output side of the seed light source fiber, the fiber amplifier fiber, and the transmitting collimator are respectively matched with the corresponding first micro-sensor, the second micro-sensor, and the third micro-sensor to collect their escaped laser signals.

[0040] The receiving lens, detector fiber and photodetector are connected in sequence on the corresponding groove, and pass through the matching fourth micro-adjustment device, fifth micro-adjustment device and sixth micro-adjustment device respectively to form the receiving optical path unit of the second light guide channel; the laser output side of the receiving lens, detector fiber and photodetector is matched with the corresponding fourth micro-sensor, fifth micro-sensor and sixth micro-sensor respectively.

[0041] An A / D converter, an adder counter, a D / A converter, and a control comparator are connected in sequence to form the first control loop; an adjustment comparator, a clock generator, and an AND gate are connected in sequence to form the second control loop.

[0042] The first control loop acquires laser information from the optical path interface. The first and second control loops control the micro-adjustment device to perform traversal adjustments within its adjustment range, and to collaboratively search, connect, optimize, and lock the optical path interfaces of the transmitting and receiving optical path units.

[0043] It should be noted that the A / D converter (analog-to-digital converter) converts the analog voltage signal output from the miniature sensor (such as a miniature photodiode) integrated into the key interface of the optical path into a digital signal. The adder counter performs a numerical addition operation between the digital signal from the A / D converter and the input control value for control purposes. The D / A converter (digital-to-analog converter) converts the value of the adder counter back into an analog voltage signal. The fused silica substrate integrated lidar transceiver optical module is a precision opto-mechanical-electrical control system integrating "sensing-decision-execution". By real-time monitoring of the escape laser (i.e., coupling loss) at each key interface in the optical path and driving the micro-adjustment device to perform dynamic compensation, it achieves globally optimal self-alignment and locking of the transmitting and receiving optical paths.

[0044] In a preferred embodiment of the present invention, the fused silica substrate is precision-processed using semiconductor-compatible processes such as photolithography, reactive ion etching, and laser direct writing to form complex three-dimensional structures such as grooves, vias, grating couplers, and microchannels. In the emitting optical path unit of the first light guide channel, high-precision fiber positioning V-grooves and lens mounting cavities can be fabricated to ensure the initial assembly accuracy of optical components. In the receiving optical path unit of the second light guide channel, embedded vertical grating couplers can be directly fabricated within the fused silica substrate, achieving efficient and low-loss vertical coupling of optical signals from the fused silica light guide layer to the underlying silicon-based optoelectronic chip. This integrates high-performance silicon-based arrays and processing circuits (e.g., amplifiers) on the same substrate platform in chip form. Simultaneously, microchannels for thermal management or hermetically sealed structures for packaging can also be integrated within the substrate, further enhancing the module's environmental robustness. The surface of the fused silica substrate is coated with an anti-reflection film. In the prior art, the substrate materials for lidar optical modules are mostly metals, ordinary glass, or silicon-based materials. While metal substrates offer high mechanical strength, their coefficients of thermal expansion differ significantly from those of optical components (such as optical fibers and glass lenses), making them prone to thermal stress mismatch during temperature changes, leading to optical path alignment failure. Silicon-based materials, though advantageous for optoelectronic integration, suffer from high thermal conductivity and a coefficient of thermal expansion that doesn't match common optical glass, easily causing thermally induced optical distortion and mechanical misalignment under high-power laser operation or drastic temperature fluctuations. Fused silica substrates offer the possibility of physically integrating the transmitting and receiving optical paths onto a single substrate. By arranging the transmitting and receiving optical paths in parallel on the same substrate, not only is the module size significantly reduced, but more importantly, both are placed in the same temperature and stress fields. This integrated layout, combined with thermal expansion matching characteristics, allows the transmitting and receiving optical paths to produce coordinated, predictable, minute deformations during environmental changes, rather than random, disordered drift.

[0045] In a preferred embodiment of the present invention, the module includes a fused silica substrate as a support structure, on which an emitting optical path unit and a receiving optical path unit are integrated. Miniature photoelectric sensors (such as miniature photodiodes) are integrated near several key optical interfaces, including the junction of the seed light source fiber and the fiber amplifier fiber, the coupling point of the fiber amplifier fiber and the emitting collimating lens, and the coupling point of the receiving lens and the detector fiber. These sensors are used to acquire, in situ and in real time, escape laser signals or backscattered light signals that are separated from the main optical path due to coupling misalignment. The intensity of these signals directly characterizes the alignment loss at the interface. Meanwhile, the mounting structure supporting each optical element is not a fixed groove, but a miniature precision adjustment device, such as a multi-degree-of-freedom micro-stage driven by piezoelectric ceramics or microelectromechanical systems. Each miniature sensor and miniature adjustment device is connected to an integrated control circuit, which includes a first control loop and a second control loop. The first control loop forms a high-precision locking loop, which, after acquiring, comparing, and integrating the sensor signals, outputs a signal to control the second control loop, achieving micron-level maintenance and optimization of the alignment state. The second control loop forms a global search loop. When the module is powered on or when it becomes severely inaccurate due to severe environmental shocks, the control and adjustment device performs a coordinated traversal scan within a certain range to quickly reconstruct the optical path.

[0046] It should be further explained that while existing advanced integrated solutions achieve high initial alignment accuracy and excellent passive stability through precision machining and adhesive bonding, all optical interfaces form a permanent rigid connection after encapsulation. This structure cannot cope with the slow release of residual stress within the material under long-term temperature cycling, nor can it compensate for the minute creep that may occur in the colloid. This means that once a hidden offset at the nanometer to micrometer level occurs, the module's performance will irreversibly degrade and cannot be detected or corrected, thus becoming a "black box."

[0047] The fundamental difference between this invention and the aforementioned prior art lies in its innovative shift from a "passive fixing" integration concept to an "active maintenance" approach. By embedding a distributed optical path state sensing network and micro-actuators within a physically integrated framework, and introducing intelligent control algorithms, the module transforms from a static component whose performance only exhibits a degradation curve after completion into an intelligent optical instrument that dynamically seeks and locks in peak performance throughout its entire lifecycle. The module can monitor and automatically compensate for any slow optical path drift caused by temperature shocks, mechanical vibrations, material aging, etc., ensuring that the core optical performance of the lidar remains high over a service life of more than ten years, meeting automotive-grade reliability requirements. The module possesses "self-healing" capabilities; after experiencing extreme environments, it can automatically recover to its optimal alignment state through an internal closed-loop system, significantly improving the overall uptime and reliability of the device under complex operating conditions. On the manufacturing side, this reduces the stringent requirements for initial assembly and adjustment precision, improving production yield; on the usage and maintenance side, it avoids complete replacement due to slow performance degradation, saving on total lifecycle costs.

[0048] In a preferred embodiment of the present invention, the photodetector portion of the receiving optical path unit employs a heterogeneous integrated optical interface design. Specifically, in the area designated for photodetection on the fused silica substrate, a silicon photonic chip integration trench with precise positioning characteristics is formed by deep etching using semiconductor technology. The key optical structure of this trench is an embedded vertical grating coupler fabricated at the interface between the fused silica and the silicon chip. The silicon-based optoelectronic integrated chip is precisely packaged within this integration trench. This chip not only differs from the fused silica substrate in material but also achieves a high degree of integration in function. Its core is a single-photon avalanche diode array serving as the detection unit. This array is directly interconnected with a low-noise transimpedance amplifier, a high-speed time-to-digital converter, and preprocessing logic circuitry integrated in situ on the chip, forming a complete "sensing-computing integrated" microsystem. The signal light from the detector fiber is efficiently vertically redirected from the fused silica light guide layer and coupled to the optical waveguide of the underlying silicon chip via the vertical grating coupler, and is ultimately received by the single-photon avalanche diode SPAD array. The entire photoelectric conversion and primary signal processing process is rapidly completed on a millimeter-scale silicon chip.

[0049] It should be further explained that existing photodetector integration solutions, whether using discrete component bonding or conventional integration, physically separate the detector from subsequent amplification and processing circuitry. Signals must be transmitted to external processing units via long wire bonding or PCB traces, introducing significant parasitic capacitance, inductance, and transmission delay, becoming a bottleneck restricting system bandwidth improvement and noise reduction. This separation of "sensing" and "processing" architecture is particularly inadequate for next-generation high-resolution, high-frame-rate LiDAR applications.

[0050] The heterogeneous integration scheme of this embodiment of the invention is not a replacement for the adaptive function, but rather a performance leap based on its robust reliability. It utilizes the optical path stability of fused silica and the powerful microelectronic integration capabilities of silicon materials. Through the key heterogeneous integration interface of the vertical grating coupler, it achieves low-loss, high-efficiency mode conversion and function transfer of optical signals from an "ultra-stable optical transmission platform" to a "high-speed optoelectronic processing engine." Signal processing near the detector on the silicon chip significantly shortens the transmission path of high-sensitivity analog signals, effectively suppressing noise introduction and signal distortion, resulting in a substantial improvement in key indicators such as the module's detection sensitivity, system bandwidth, and ranging accuracy. This design achieves extreme compactness of the optoelectronic functional unit physically and "sensing-computing synergy" at the system level. The logic circuits on the chip can execute pixel-level signal processing algorithms (such as time-of-flight extraction and background light suppression) in real time, uploading the processed digital signal instead of the fragile analog signal, greatly enhancing the system's anti-interference capability and intelligence level. This heterogeneous integration architecture provides a standardized and modular platform for the functional expansion of the module. System performance can be continuously improved by replacing or upgrading silicon photonic chips that integrate more advanced algorithms or more functions (such as integrated on-chip laser modulators, multi-wavelength processing units, etc.) without redesigning the entire fused silica substrate, thus protecting investment and extending the technology life cycle.

[0051] In a preferred embodiment of the present invention, the micro-adjustment device is a dedicated micro-motion platform that is integrated with a fused silica substrate in three-dimensional space through micromachining or precision assembly processes. The micro-adjustment device includes a substrate and mounting interface, a composite flexible hinge guide mechanism, a multi-mode piezoelectric drive unit, a motion stage, and an in-situ sensing system.

[0052] It should be further explained that the substrate and mounting interface are the fixed parts of the device, which are rigid substrates made of microcrystalline ceramic or silicon-based materials. This substrate is permanently fixed to the mounting cavity pre-processed on the fused silica substrate using a low-stress adhesive or anodic bonding process with a matching coefficient of thermal expansion. The substrate has precise mechanical reference surfaces to define the initial coordinate system for motion.

[0053] It should be further explained that the composite flexible hinge guiding mechanism (the core structure for achieving "traversal" and "fine-tuning") is the core connecting the base and the motion stage. Its unique design lies in integrating two elastic deformation units with different stiffnesses and motion ranges into one unit. The composite flexible hinge guiding mechanism includes a large-stroke scanning unit and a high-precision micro-motion unit. The large-stroke scanning unit consists of one or more pairs of low-stiffness, long-arm curved flexible hinges. This unit allows the motion stage connected to it to generate a continuous displacement range of tens to hundreds of micrometers in the XY direction parallel to the optical interface plane under the action of driving force. This range is the "adjustment range," providing the necessary physical motion space for the traversal search driven by the second control loop. The geometric parameters of the hinge (such as thin wall thickness and arm length) have been optimized through finite element analysis to ensure good linearity of motion within the large deformation range and to prevent plastic deformation or fatigue. The high-precision micro-motion unit is nested inside the motion chain of the large-stroke scanning unit and is a set of high-stiffness, short-arm parallel plate-type flexible hinges. Once the motion stage is driven to the vicinity of the target area via the large-stroke unit, the control logic switches to a micro-motion mode dominated by this high-stiffness unit. This unit provides only a few micrometers of working stroke, but due to its high stiffness and excellent motion guidance accuracy, it can achieve displacement resolution better than 10 nanometers and extremely high attitude repeatability, and is specifically used for the final optimization and locking of the first control loop.

[0054] It should be further explained that the multi-mode piezoelectric drive unit uses a customized composite stacked piezoelectric ceramic actuator as its drive source. This actuator integrates two piezoelectric ceramic groups with different performance characteristics in its physical structure. The multi-mode piezoelectric drive unit includes a stacked piezoelectric ceramic for large-range scanning and a shear piezoelectric ceramic for precision locking. The stacked piezoelectric ceramic for large-range scanning has a large displacement output capability (e.g., rated displacement > 20 μm) and a fast response speed, used to drive the large-stroke scanning unit to perform rapid grid or helical traversal motion. The shear piezoelectric ceramic for precision locking has extremely high displacement resolution (down to the sub-nanometer level) and stiffness, nested inside the stacked ceramic or arranged in parallel, used to drive the high-precision micro-motion unit during the precision locking stage to achieve final alignment fine-tuning. The electrodes of the two piezoelectric ceramics are independently led out, receiving independent or coordinated control signals from the first and second control loops.

[0055] It should be further explained that the motion stage and in-situ sensing system are as follows: the motion stage is made of a low thermal expansion alloy and has precision interfaces for mounting optical components (such as fiber optic V-grooves or lens holders). High-reflectivity miniature optical mirrors or capacitive sensing electrodes are integrated into the motion stage. At corresponding positions on the substrate, miniature laser displacement sensor probes or capacitive sensor electrodes are fixed. Together, they form an in-situ, high-bandwidth position measurement closed loop, which measures the absolute displacement of the motion stage relative to the substrate in real time. The measurement data is directly fed back to the control loop, forming a "position loop-performance loop" dual closed-loop control with the optical path loss signal (performance target), ensuring search efficiency and locking accuracy.

[0056] Furthermore, in the optical transmission unit, the first, second, and third microsensors do not simply process their signals in parallel. Instead, they are connected to a first central signal selection and routing unit implemented by logic circuitry (e.g., a dedicated switch array). This first central signal selection and routing unit, configured by upper-level control logic, can sequentially select the signals from the three sensors into three specific pairwise combinations according to a preset timing sequence and send them to the back-end processing circuit. The first central signal selection and routing unit has two output terminals: a first output terminal and a second output terminal. During the operating cycle of any selected sensor combination, the two output terminals output signals in a strict timing sequence: the signal at the first output terminal is valid earlier than that at the second output terminal. This design allows the back-end circuitry to distinguish the "sequential" or "master-slave" relationship between the signals of two sensors within the same group. The specific combinations and functional definitions are as follows:

[0057] First combined mode (selecting the first and second micro-sensors): In this mode, the system focuses on the interface between the seed light source fiber and the fiber amplifier fiber. The first output terminal outputs the signal from the first micro-sensor, and the second output terminal outputs the signal from the second micro-sensor. By comparing the difference in escape laser intensity and the timing relationship detected by the two sensors, the control system can accurately determine the lateral offset, angular misalignment, or end-face gap abnormality at the interface, and is dedicated to driving the first and second micro-adjustment devices to perform coordinated compensation.

[0058] Second combined mode (selecting the second and third micro-sensors): In this mode, the system focuses on the coupling interface between the fiber optic amplifier's fiber output and the transmitting collimator. The first output terminal outputs the signal from the second micro-sensor, and the second output terminal outputs the signal from the third micro-sensor. The signal difference in this combination directly reflects the coupling efficiency of the amplified laser beam injected into the collimator, and is used to diagnose and drive the second and third micro-adjustment devices to optimize the mode field matching and position alignment of the beam in front of the collimator.

[0059] The third combination mode (gating the first and third micro-sensors): This mode is a global fine-tuning and diagnostic mode. By skipping the intermediate amplifier fiber interface, it directly compares the escape light signal at the seed light source output end and the final collimated output end. This signal difference reflects the cumulative loss change of the entire transmission optical path link. This mode is used to perform closed-loop fine-tuning of the overall performance of the entire transmission optical path after completing the first two stages of local calibration, or to quickly diagnose long-term trend drift in optical path performance.

[0060] The signal path of the first output terminal is as follows: digitized sequentially by an A / D converter and input to an adder counter. The core function of the adder counter here is as an addressable digital potentiometer or an integral accumulator. Its count value (corresponding to digital voltage) can be increased or decreased according to the decision result of the control algorithm. After being converted into an analog voltage by a D / A converter, this value is sent to the positive input terminal (+) of the control comparator.

[0061] The signal at the second output terminal is directly (or after necessary buffering and scaling) connected to the inverting input terminal (-) of the control comparator.

[0062] The core function of the control comparator is thus redefined; it is no longer a simple fixed-threshold comparator, but a dynamic reference comparator. The voltage at its positive input (from the output of the adder counter / DAC link) is a dynamic reference voltage V_ref, intelligently set by the system, representing the "desired minimum loss" or "optimization target." Its inverting input receives a real-time feedback voltage V_fb from the second output of the currently selected sensor combination, representing the actual loss at the current monitoring point.

[0063] The control comparator continuously compares V_ref with V_fb and outputs an error signal. This error signal is the direct basis for subsequent adjustments. More importantly, this error signal is also fed into the second control loop for the operation of the gated clock generator. For example, when the error exceeds a certain threshold, it indicates that the current local optimization can no longer meet the requirements, and a mode switch or a larger-scale search may need to be initiated. At this time, the signal output by the control comparator can trigger or modulate the clock generator, thereby coordinating the switching of the entire system's operating state.

[0064] Compared with existing technologies, common processing methods for optical alignment systems with multi-sensor feedback include: simple summation or averaging, parallel independent processing, or fixed-order polling.

[0065] Parallel independent processing involves each sensor being connected to an independent amplification and comparison channel. The system needs to process multiple error signals that may conflict with each other, resulting in complex control logic. It is difficult to determine the primary and secondary contradictions of the imbalance, which can easily lead to adjustment oscillation or getting trapped in local optima.

[0066] Simple summation or averaging combines all sensor signals into a single comprehensive error index. This method completely loses its fault location capability, failing to distinguish between seed light source coupling issues and collimating lens coupling issues, leading to blind adjustment actions. All adjustment devices may operate simultaneously, resulting in low efficiency and potential mutual interference.

[0067] While fixed-sequence polling can read data from each sensor in a time-sharing manner, it lacks intelligent combination and correlation analysis based on optical path physical logic. For example, it cannot establish a global correlation diagnostic mode that transcends intermediate links, such as the combination of "first sensor and third sensor".

[0068] The hierarchical gating control architecture of this invention differs fundamentally from the aforementioned existing technical solutions in both logic and performance. The leap in diagnostic accuracy and intelligence is achieved through three predefined combination modes implemented by the first central signal gating and routing unit. These modes are not random or simple polling, but are meticulously designed based on the physical model of the transmitted optical path (serial link) and fault tree analysis. This enables the system to "locate the segment where the misalignment occurs," knowing whether it's a problem in the "AB segment," "BC segment," or the "AC as a whole," which is a prerequisite for achieving precise and efficient adjustment. The dynamic benchmark and closed-loop tuning mechanism places the adder-D / A converter link at the comparator benchmark, creating an optimization target point that can be dynamically set by the algorithm. The system can adjust the value of the adder counter through algorithms (such as gradient descent) to actively and tentatively find the V_ref that minimizes V_fb (i.e., minimizes loss), thereby driving the adjustment device to reach that state. This is an active optimization process, rather than a passive error elimination. System-level coordination and state machine transitions utilize the output of the control comparator to simultaneously drive execution and regulate the second loop (search loop), achieving organic linkage and smooth switching between the first loop (optimization loop) and the second loop (search loop). For example, in a certain combination mode, if V_ref cannot be adjusted to a satisfactory level, the continuous large error signal output by the comparator can directly trigger the second loop to intervene, switching sensor combinations or initiating an traversal search, forming a highly coordinated adaptive state mechanism. First, it determines which part of the optical path the misalignment occurs in, and then adjusts the corresponding actuator accordingly, avoiding the inefficiency and potential interference caused by "one-size-fits-all" adjustments, greatly improving the speed, accuracy, and reliability of the adjustment. Through precise positioning, each adjustment only requires the use of 1-2 related micro-adjustment devices, rather than all devices, reducing unnecessary mechanical movement, lowering power consumption, and increasing the service life of the actuators. The third combination mode (first and last sensor combination) provides a global monitoring mechanism. Even if an intermediate sensor temporarily fails, the system can still maintain overall performance through this mode, enhancing the system's fault tolerance. This solution organically combines hardware switching, timing control, dynamic reference setting, and multi-loop coordination to form a complete set of dedicated control logic for adaptive alignment of multi-node serial optical paths.

[0069] Preferably, the adaptive control of the receiving optical path unit uses integrated fourth, fifth, and sixth micro-sensors as core sensing nodes, and constructs its signal processing and decision-making logic according to the principle of "segmented diagnosis and coordinated adjustment." Specifically:

[0070] The outputs of the fourth, fifth, and sixth microsensors are all connected to a dedicated first central signal gating and routing unit. Similar to the transmitting optical path unit, the first central signal gating and routing unit is configured by the main control logic and can intelligently organize the signals from the three receiving sensors into three diagnostic combination modes according to the physical connection relationship of the optical path, and gating them according to a preset timing sequence.

[0071] The first central signal gating and routing unit is provided with a first output terminal and a second output terminal. In any combination mode, the signal of the first output terminal is valid before that of the second output terminal to establish the reference and feedback relationship for signal comparison.

[0072] The three diagnostic combination modes of the receiving optical path are defined as follows: The first combination mode (gating the fourth and fifth micro-sensors) focuses on the coupling interface between the receiving lens and the detector fiber. The first output terminal outputs the signal from the fourth micro-sensor (monitoring stray light at the receiving lens), and the second output terminal outputs the signal from the fifth micro-sensor (monitoring the input end of the detector fiber). The difference between the two signals directly reflects the coupling efficiency of the echo beam injected into the detector fiber after being focused by the lens. This mode is specifically used to drive the fourth and fifth micro-adjustment devices to optimize the alignment of the lens and the focusing spot at the fiber end face. The second combination mode (gating the fifth and sixth micro-sensors) focuses on the opto-electric coupling interface between the detector fiber output end and the photodetector (such as the SPAD array chip in Example 2). The first output terminal outputs the signal from the fifth micro-sensor, and the second output terminal outputs the signal from the sixth micro-sensor (monitoring escape light near the photosensitive surface of the detector chip). This combination is used to diagnose the matching status between the fiber-emitter light and the detector's sensitive area, and to drive the fifth and sixth micro-adjustment devices to compensate, ensuring efficient conversion of optical power into electrical signals. The third combination mode (gating the fourth and sixth miniature sensors) is a global performance monitoring mode for the receiving optical path. It directly compares the signals at the receiving lens inlet and the final photodetector position by traversing the intermediate fiber optic transmission links. This combination reflects the total insertion loss change of the entire receiving optical path and can be used for closed-loop fine-tuning of overall performance after segmented calibration, or for long-term monitoring of the performance degradation trend of the receiving link, providing data for preventative maintenance.

[0073] In the signal processing path, the first output signal of the first central signal gating and routing unit is sequentially digitized by the A / D converter and then sent to the adder counter. In the receiving optical path control context, the adder counter acts as a digital reference setting unit. Its accumulated or decremented count value is converted by the D / A converter to generate a dynamic reference voltage V_ref_rx representing the current expected optimal value, and is applied to the positive input (+) of the control comparator.

[0074] At the same time, the second output signal of the first central signal gating and routing unit is directly (or after conditioning) sent to the inverting input (-) of the control comparator as the current actual feedback voltage V_fb_rx;

[0075] Therefore, during the optical path control period, the control comparator is configured as a dynamic reference comparator, which continuously compares V_ref_rx with V_fb_rx. The output error signal is used, on the one hand, to drive the corresponding micro-adjustment device to perform precise positioning compensation; on the other hand, the error signal is fed to the second control loop to monitor or trigger the clock generator. For example, when the error continues to exceed the threshold, it indicates that the current segmentation optimization is ineffective, and it may be necessary to switch the diagnostic mode or initiate a global re-search of the optical path. At this time, the output of the control comparator will prompt the second loop to intervene.

[0076] Compared to existing technologies, in the field of lidar receiving optical paths, current technologies handle optical path stability in a passive manner. Among these, the most common approach is complete static curing, where the receiving lens, fiber optic cable, and detector are aligned and glued together in one go, treating it as an unadjustable whole. Its performance relies entirely on the initial assembly accuracy and the long-term stability of the materials, falling under the "black box" model described in the background, and is unable to handle internal drift. Single signal monitoring, in a few cases, may monitor the final electrical signal output (such as the detector's total current), but this only detects whether overall performance has degraded, and cannot pinpoint the fault location—whether it's in the receiving lens, transmission fiber, or detector coupling. Once performance deteriorates, only the entire system can be replaced; targeted repair or adjustment is impossible. The lack of dedicated control logic means that existing technologies lack a smart gating and dynamic comparison control architecture specifically designed for the multiple physical interfaces of the receiving optical path, with segmented diagnostic capabilities. The receiving end is typically treated as a single component, and its internal state adjustment is not included in the system's adaptive framework.

[0077] This invention presents a control architecture for the receiving optical path that represents a generational leap forward compared to existing technologies. The receiving optical path actively senses and adjusts its internal state by deploying fourth to sixth micro-sensors and corresponding adjustment devices, transforming it from a "static component" into an "adjustable system" and enabling it to adapt to internal interface drift. Precise segmentation and localization of receiving link faults are achieved through three combined modes implemented by the first central signal gating and routing unit. The system can accurately determine, like a diagnostic circuit, whether performance degradation stems from "lens-fiber" coupling (first mode), "fiber-detector" coupling (second mode), or overall link aging (third mode). This provides a fundamental basis for efficient and precise compensation adjustments. Achieving a fully adaptive closed-loop transmission and reception system, this scheme, together with the control architecture of the transmitting optical path, constitutes a complete intelligent maintenance system covering the entire optical link of the lidar "transmit-receive." It ensures that not only is the emitted light optimal, but the path of the received light also remains optimal, thus achieving unprecedented long-term performance stability at the system level. The system can actively compensate for micron-level misalignment at each interface in the receiving optical path caused by vibration and temperature drift, ensuring that the system's receiving sensitivity (i.e., detection capability) does not deteriorate throughout its entire lifecycle. This is crucial for maintaining the long-range, weak-signal detection capability of the lidar. When receiving performance degrades, the system can automatically diagnose the problem and attempt self-repair, preventing the entire receiving module from being scrapped due to a tiny offset at a single interface, significantly reducing the overall lifecycle maintenance cost. In harsh vibration and high / low temperature cycling environments, such as in automotive applications, the receiving and transmitting optical paths can work together to adaptively adjust, jointly ensuring the performance consistency of the radar optical front end under complex operating conditions and meeting automotive-grade reliability requirements.

[0078] Furthermore, in the transmitting optical path unit, the first, second, and third micro-sensors respectively monitor the key optical interfaces at the seed light source fiber output end, the fiber amplifier fiber input / output end, and the transmitting collimator input end. In the receiving optical path unit, the fourth, fifth, and sixth micro-sensors respectively monitor the key optical interfaces at the receiving lens focal spot position, the detector fiber input end, and the photosensitive surface position of the photodetector (such as a SPAD array). The output signals of the above six micro-sensors (usually voltage signals proportional to the escape light intensity or backscattered light intensity) are all connected to the first central signal gating and routing unit (the first central signal gating and routing unit). This unit is essentially a high-performance, multi-channel analog switch array, controlled by mode selection commands issued by the main controller (such as an FPGA or microprocessor).

[0079] The first central signal gating and routing unit does not simply output all sensor signals in parallel. Instead, it organizes them into six predefined "diagnostic modes" based on the physical topology of the optical path and fault diagnosis logic. Each mode gating a pair of logically related sensors to characterize the coupling state of a specific optical segment:

[0080] Mode 1 is a front-end diagnostic of the transmitting optical path, selecting the first and second micro-sensors. This combination is specifically used to diagnose the fiber optic connection segment between the seed source fiber and the fiber amplifier fiber. The alignment status of this segment is mainly determined by the first micro-adjustment device (carrying the seed source fiber) and the second micro-adjustment device (carrying the fiber amplifier fiber input segment).

[0081] Mode two is a post-transmit optical path diagnostic, selecting the second and third micro-sensors. This combination is specifically used to diagnose the spatial optical coupling segment between the fiber amplifier's fiber output and the transmitting collimator. The alignment of this segment is primarily determined by the second micro-adjustment device (carrying the fiber amplifier's fiber output segment) and the third micro-adjustment device (carrying the transmitting collimator).

[0082] Mode 3 is a global evaluation of the emitted optical path, selecting the first and third micro-sensors. This combination bypasses intermediate stages, directly evaluating the overall efficiency of the entire emission link from the seed source to the emitted beam. It is used for closed-loop fine-tuning of overall performance after fine adjustments in the first two modes, or for long-term monitoring of performance drift trends in the emitted optical path. In this mode, the first, second, and third micro-adjustment devices can be invoked for collaborative optimization.

[0083] Mode four is a front-end diagnostic for the receiving optical path, selecting the fourth and fifth microsensors. This combination is specifically used to diagnose the spatial optical coupling segment between the converging spot of the receiving lens and the fiber optic input of the detector. The alignment of this segment is primarily determined by the fourth micro-adjustment device (carrying the receiving lens) and the fifth micro-adjustment device (carrying the fiber optic input segment of the detector).

[0084] Mode 5 is for post-optical path diagnostics, selecting the fifth and sixth microsensors. This combination is specifically used to diagnose the optical-electric coupling section between the detector fiber output and the photosensitive surface of the photodetector. The alignment of this section is primarily determined by the fifth micro-adjustment device (carrying the detector fiber output section) and the sixth micro-adjustment device (carrying the photodetector).

[0085] Mode six is ​​a global evaluation of the receiving optical path, selecting the fourth and sixth micro-sensors. This combination is used to evaluate the overall performance of the entire receiving link from the receiving lens entrance to the electrical signal output. It can be used for overall fine-tuning after segmented calibration or for long-term trend monitoring, and can call upon the fourth, fifth, and sixth micro-adjustment devices for coordinated adjustment.

[0086] When any diagnostic mode is activated, the first central signal gating and routing unit outputs two analog signals with strict timing relationships: a first output signal and a second output signal. The setup and hold times of the first signal are designed to lead those of the second signal. This timing relationship provides a stable reference for the back-end comparator circuit.

[0087] Furthermore, the two signals from the first central signal gating and routing unit are fed into a two-stage comparison and decision circuit. The first output signal is sent to the non-inverting input (+) of the adjustment comparator. The adjustment comparator is configured as a fast-response window comparator, with its inverting input (-) connected to a programmable offset threshold voltage. Its function is to perform a fast "normal / offset" binary judgment. When the first signal (representing the front-end state of the diagnosed segment) is lower than the preset offset threshold, the adjustment comparator outputs a high level (logic "1"), generating a primary flag signal indicating "this segment may be inaccurate, adjustment requested." The second output signal is sent to the inverting input (-) of the control comparator. The non-inverting input (+) of the control comparator receives a dynamic reference voltage generated from the adder counter and digital-to-analog converter link, representing the current desired optimal performance target. The control comparator essentially constitutes a precision error amplifier, whose output voltage is proportional to (dynamic reference voltage - second feedback signal). This error voltage accurately reflects the polarity and magnitude of the misalignment and is used for subsequent precision driving. The "adjustment request" signal output by the adjustment comparator, along with a clock pulse output by a system adjustment enable clock generator managed by the core algorithm of the first control loop, are input together into a global adjustment arbitration AND gate. The system adjustment enable clock generator is not continuously operational; it is only activated within a "time window" when the system deems it safe and suitable for maintenance operations (such as after power-on initialization, during preset periodic maintenance intervals, or during calibration request periods triggered by the performance monitoring module). The global adjustment arbitration AND gate performs a logical "AND" operation. This means that the AND gate will only output a high-level valid adjustment execution command when both conditions are simultaneously met: "a segment is detected as potentially out of balance" and "the system's global timing allows for adjustment." This mechanism is crucial for ensuring the system's main detection function is not interfered with, achieving intelligent time-division multiplexing of business operations and background maintenance.

[0088] It should be noted that the central signal selection and routing unit includes a first central signal selection and routing unit and a second central signal selection and routing unit. The first and second central signal selection and routing units have the same function and are controlled by the same control system for data coordination, such as a shared chip control system. The control output of the second central signal selection and routing unit is connected to the first, second, third, fourth, fifth, and sixth micro-adjustment devices, respectively, while the control input of the second central signal selection and routing unit is connected to the input of the adjustment comparator.

[0089] Furthermore, the valid adjustment command output from the AND gate is sent as a gating control signal to a multiplexer adjustment command distributor. The channel switching logic of this distributor is strictly synchronized with the currently active diagnostic mode of the first central signal gating and routing unit. When the adjustment command is valid, the multiplexer adjustment command distributor, based on the current mode, precisely routes the precise error voltage signal from the control comparator to the drive circuit of the micro-adjustment device corresponding to the current diagnostic mode. For example, if the system is operating in mode two and arbitration passes, the error signal is distributed to the drive amplifiers of the second and third micro-adjustment devices. These drive circuits drive piezoelectric ceramic actuators based on the error signal to perform coordinated nanometer-level pose adjustments on the fiber optic amplifier's fiber output and the transmitting collimator to minimize coupling loss in this segment until the error voltage output from the control comparator approaches zero, at which point the system enters a locked state.

[0090] Existing technologies for achieving optical module stability or limited adjustment typically follow several methods, all of which have inherent limitations. The most common is monitoring only the final output optical power or received electrical signal strength. This approach is akin to diagnosing a disease solely by body temperature; it can only sense the system's "illness" but cannot pinpoint the specific link in the optical path where the "lesion" is located. Once performance degrades, adjustments are often blind or global, resulting in low efficiency and potentially introducing new misalignments. Another method involves setting independent sensors and controllers for each adjustment point. While this method can achieve localized adjustment, it lacks communication and coordination between closed loops. When multiple interfaces experience slight drift simultaneously, the independent closed loops may generate conflicting adjustment commands, leading to overall system oscillation, getting trapped in local optima, or a lengthy adjustment process, ultimately reducing system stability. Yet another method involves adjusting according to a fixed schedule without considering the actual state of the optical path and the system's operational phase. This can lead to unnecessary adjustments during critical detection tasks for the lidar, interfering with the main function; or adjustments that are truly needed may not be executed due to the timing, missing the optimal compensation window.

[0091] It should be further explained that this invention, based on a "segmented" diagnostic architecture of optical path physical topology, intelligently divides continuous optical paths into logical segments and establishes a precise mapping relationship of "specific sensor combination ↔ specific segment ↔ specific regulator combination". This achieves precise fault location and precise allocation of adjustment resources at the system architecture level. A three-level decision-making mechanism is established: "rapid initial judgment (hardware comparison) → central arbitration (timing and logic gating) → precise distribution (mode synchronization routing)". In particular, the central arbitration stage, by introducing a system adjustment enable clock, transforms adaptive maintenance behavior into a controlled and schedulable system task, ensuring that all adjustment actions absolutely obey the system's highest-level work rhythm and task priority, achieving seamless coexistence of business functions and maintenance functions. This system is not a simple error reactor, but a proactive maintenance system with predictive and scheduling capabilities. It can automatically and orderly inspect and optimize each segment within system idle or reserved maintenance windows, keeping the optical module near its peak performance, achieving a leap from "failure prevention" to "continuous optimization". The system can automatically complete the entire process from "full-link scanning and locating inaccurate segments" to "arbitration decision-making and driving the corresponding regulator for precise compensation." This transforms the complex assembly, adjustment, and maintenance work that traditionally relies on precision instruments and experienced technicians into a built-in, automatically running daily process, physically realizing "full lifecycle self-sensing and self-optimization." The central arbitration mechanism fundamentally eliminates the "regulator conflict" and oscillation problems common in multi-degree-of-freedom control systems. Based on global timing trigger conditions, it ensures the absolute priority and continuity of the core detection function of the lidar. The system can "silently" repair performance damage caused by vibration, temperature drift, etc., in the background, exhibiting extremely high robustness and uptime.

[0092] This invention also provides an adjustment method for an integrated lidar transceiver optical module on a fused silica substrate. This method, implemented using an integrated lidar transceiver optical module on a fused silica substrate, includes the following steps:

[0093] Step 1: The module automatically initiates the adjustment process upon initial power-on, upon receiving an external calibration command (such as a signal triggered by a temperature, humidity, or vibration sensor), or when the internal timer reaches the preset maintenance cycle. First, the system performs initialization: powering on and resetting the first to sixth micro-adjustment devices, the first to sixth micro-sensors, and each control circuit; the main controller (such as an FPGA) loads the fixed control algorithm and parameters, including configuration tables for six diagnostic modes, the offset threshold V_th for each mode, the initial value of the dynamic reference voltage V_ref, and the displacement range of the adjustment device, etc.

[0094] After initialization, the main controller selects and activates the first diagnostic mode according to a preset strategy. This strategy can be sequential execution (from mode one to mode six) or priority execution based on historical performance data (e.g., prioritizing segments with significant recent drift). The first central signal gating and routing unit receives the mode command and physically connects the corresponding micro-sensors. For example, if mode one (front-end diagnostic of the transmit optical path) is activated, the first micro-sensor (monitoring the seed light source fiber output) and the second micro-sensor (monitoring the fiber amplifier fiber input) are selected. This unit outputs two signals: a first signal (from the "upstream" sensor of the segment, such as the first micro-sensor) and a second signal (from the "downstream" sensor of the segment, such as the second micro-sensor), and internal timing circuitry ensures that the establishment time of the first signal is earlier than that of the second signal, providing a stable reference for subsequent comparisons. The selected mode also determines the combination of micro-adjustment devices to be adjusted in this step (e.g., mode one corresponds to the first and second micro-adjustment devices).

[0095] Step Two: The first signal is fed into the non-inverting input of the adjustment comparator and compared with a programmable offset threshold voltage V_th. V_th is preset according to the characteristics of each mode and represents the maximum tolerable loss boundary of that segment. If the voltage value of the first signal is lower than V_th, it indicates that the front-end coupling state of that segment has been significantly degraded, and there is a clear risk of offset. The adjustment comparator then outputs a high-level adjustment request signal. This is a fast hardware-level "anomaly detection".

[0096] Meanwhile, the high-level management algorithm in the first control loop evaluates the overall system state to determine whether the current time window for adjustment is open. The criteria for this judgment include: whether the lidar is in a laser emission gap or standby state, whether the system is not executing a high-priority task, and whether the power supply is stable. This design ensures that adjustment behavior will never interfere with normal laser detection and ranging functions. If the conditions are met, the first control loop activates the system adjustment enable clock generator, causing it to output a series of periodic clock pulses. Conversely, even if an imbalance is detected, adjustment will be delayed until a safe window appears.

[0097] Step 3: The adjustment request signal and system adjustment control are executed by performing a logical AND operation on the clock pulse output from the clock generator within an AND gate. A valid adjustment execution command is generated only when both input signals are simultaneously valid. This AND gate performs a logical AND operation: its output is high only when both input signals are high simultaneously. This arbitration mechanism is key to achieving "uninterrupted service" self-maintenance in this invention. It generates a valid adjustment execution command, which is a clock-synchronized pulse signal, indicating that the system has been authorized to perform specific adjustment actions on the current diagnostic segment.

[0098] Step 4: The second signal, serving as the feedback voltage V_fb characterizing the actual coupling loss of the current segment, is sent to the inverting input (-) of the control comparator. The non-inverting input (+) of the control comparator receives the dynamic reference voltage V_ref from the adder counter and D / A converter link; V_ref is not a fixed value, but a performance target value set by the control algorithm. The control comparator calculates V_ref - V_fb in real time and outputs an error voltage signal proportional to it. This signal not only contains the direction (positive or negative) of the offset but also quantifies the degree of offset.

[0099] A valid adjustment execution command acts as a strobe signal, acting on a multiplexer adjustment command distributor. The distributor's channel mapping is strictly synchronized with the current diagnostic mode of the first central signal strobe and routing unit. When the command is valid, the distributor controls the error voltage signal output by the comparator, precisely routing it to the micro-adjustment drive circuit corresponding to the current mode. For example, in mode one, the error voltage is simultaneously sent to the high-precision drive amplifiers of both the first and second micro-adjustments.

[0100] Step 5: The drive circuit that receives the error voltage signal converts it into a drive current or voltage, which is then applied to the corresponding micro-adjustment device (such as a piezoelectric ceramic actuator). In the example of Mode 1, the first and second micro-adjustment devices work together under the guidance of the error signal to fine-tune the lateral position, axial gap, or pitch / yaw angle of the seed light source fiber and the fiber amplifier fiber. During the adjustment process, the feedback signal from the micro-sensor changes in real time, and V_fb changes accordingly.

[0101] Meanwhile, the intelligent algorithm (such as gradient descent) in the first control loop continuously monitors the changing trend of the error voltage. The algorithm dynamically changes V_ref by adjusting the value of the adder counter, aiming to guide the adjustment action so that V_fb continuously approaches V_ref, i.e., the error voltage tends towards zero. This forms a closed-loop optimization process of "setting the target - executing the adjustment - collecting feedback - correcting the target". When the absolute value of the error voltage is less than a preset, extremely small locking threshold, the algorithm determines that the current segment has reached the optimal or near-optimal alignment state, and then controls the drive circuit to enter the holding mode, locking the position of the adjustment device and completing the optimization of that segment.

[0102] Step Six: After optimizing and locking one diagnostic segment, the main controller automatically switches to the next diagnostic mode (e.g., from mode one to mode two) according to a predetermined strategy. The system repeats steps one through five to diagnose and adjust the coupling segment between the fiber optic amplifier fiber and the transmitting collimator. This cycle continues, finely adjusting each preset segment of the transmitting and receiving optical paths in turn.

[0103] If, in a certain mode (e.g., mode two), closed-loop adjustment continues but the error voltage consistently fails to converge to within the lockout threshold, it indicates that the system may be trapped in a local optimum or that environmental shocks have caused severe inaccuracies, exceeding the range of local fine-tuning. In this case, the persistently large error signal output by the control comparator will trigger the intervention of the second control loop. The second control loop takes over control, ordering the relevant micro-adjustment devices (the second and third micro-adjustment devices in mode two) to temporarily exit the closed-loop optimization mode and instead perform a traversal search motion of a preset pattern (such as a grid or spiral) within their entire mechanical travel range. During this process, the system continuously monitors sensor signals to find performance peaks. Once a region with significantly better performance is recaptured, control is returned to the first control loop, and the closed-loop optimization process restarts from the vicinity of that region (i.e., jumps back to step four).

[0104] Through the cyclical execution and intelligent switching of the above six steps, this method can systematically establish, restore and maintain the optimal alignment state of all key interfaces of the integrated transmission and reception optical path on the entire fused silica substrate. It realizes the all-round autonomous adjustment capability from local fine-tuning to global search, from periodic maintenance to real-time adaptation, thereby ensuring the long-term reliability and performance consistency of the lidar optical front end under extreme environments.

[0105] It should be noted that technical features that are not fully explained will be addressed using conventional technical methods.

[0106] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described herein. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or equivalent substitutions to the present invention, as well as all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.

Claims

1. An integrated laser radar transceiver optical module on a fused silica substrate, comprising a transmitting optical path unit, a receiving optical path unit, an A / D converter, an adder counter, a D / A converter, a control comparator, an adjustment comparator, a clock generator, and an AND gate; the optical elements of the transmitting optical path unit include a seed light source fiber, a fiber amplifier fiber, a transmitting collimating lens, a first central signal gating and routing unit, and a multiplexer adjustment command distributor; the optical elements of the receiving optical path unit include a receiving lens, a detector fiber, and a photodetector, characterized in that... The optical elements of the transmitting optical path unit and the receiving optical path unit are integrated onto the micro-adjustment device on the fused silica substrate in the order of the optical path. Corresponding miniature sensors are respectively set on the laser emitting side of the optical elements of the transmitting optical path unit and the receiving optical path unit; An A / D converter, an adder counter, a D / A converter, and a control comparator are connected in sequence to form the first control loop; an adjustment comparator, a clock generator, and an AND gate are connected in sequence to form the second control loop. The first central signal selection and routing unit is used to select the corresponding combination of micro-sensors according to six predefined segmented diagnostic modes, and output a first signal and a second signal with a timing relationship. The first signal is input to the adjustment comparator and compared with a preset offset threshold to generate an adjustment request signal. The adjustment request signal and the clock pulse output by the clock generator are logically ANDed in an AND gate to generate an adjustment execution command. The second signal is used as a feedback voltage input to the control comparator and compared with the dynamic reference voltage at its non-inverting input terminal to generate an error voltage signal. The adjustment execution command is sent as a strobe control signal to the multiplexer adjustment command distributor. The multiplexer adjustment command distributor, based on the current mode, accurately routes the error voltage signal from the control comparator to the drive circuit of the micro-adjustment device corresponding to the current diagnostic mode. The first control loop acquires laser information from the optical path interface. The first and second control loops control the micro-adjustment device to perform traversal adjustments within its adjustment range, and to collaboratively search, connect, optimize, and lock the optical path interfaces of the transmitting and receiving optical path units.

2. The fused silica substrate integrated lidar transceiver optical module of claim 1, wherein, The micro-sensors include a first micro-sensor, a second micro-sensor, a third micro-sensor, a fourth micro-sensor, a fifth micro-sensor, and a sixth micro-sensor; the micro-adjustment devices include a first micro-adjustment device, a second micro-adjustment device, a third micro-adjustment device, a fourth micro-adjustment device, a fifth micro-adjustment device, and a sixth micro-adjustment device; wherein, the first to third micro-sensors and the first to third micro-adjustment devices are respectively disposed on the laser output side of the seed light source fiber, the fiber amplifier fiber, and the transmitting collimator of the transmitting optical path unit; the fourth to sixth micro-sensors and the fourth to sixth micro-adjustment devices are respectively disposed on the laser output side of the receiving lens, the detector fiber, and the photodetector of the receiving optical path unit.

3. The integrated lidar transceiver optical module based on a fused silica substrate according to claim 1, characterized in that, The six diagnostic modes include: Mode 1: Select the first micro-sensor and the second micro-sensor to diagnose the connection segment between the seed light source fiber and the fiber amplifier fiber; Mode 2: Select the second and third micro-sensors, corresponding to the coupling segment between the fiber optic amplifier fiber and the transmitting collimator; Mode 3: Select the first and third micro-sensors to diagnose the overall performance of the emission optical path; Mode 4: Select the fourth and fifth microsensors, corresponding to the coupling segment between the diagnostic receiving lens and the detector fiber; Mode 5: Select the fifth and sixth microsensors, corresponding to the coupling segment between the diagnostic detector fiber and the photodetector; Mode 6: Select the fourth and sixth micro-sensors to diagnose the overall performance of the receiving optical path.

4. The fused silica substrate integrated lidar transceiver optical module according to claim 1 or 2, characterized in that, The photodetector adopts a heterogeneous integrated structure, including a silicon-based optoelectronic chip packaged in an integrated trench of a fused silica substrate. The chip integrates a single-photon avalanche diode array, a low-noise transimpedance amplifier, a high-speed time-to-digital converter, and preprocessing logic circuitry. The detector fiber and the silicon-based optoelectronic chip are coupled by an embedded vertical grating coupler.

5. The fused silica substrate integrated lidar transceiver optical module of claim 2, wherein, The micro-adjustment device is a composite flexible hinge guide mechanism, including a large-stroke scanning unit and a high-precision micro-motion unit; the large-stroke scanning unit is used to achieve traversal search on the order of tens to hundreds of micrometers under the control of the second control loop; the high-precision micro-motion unit is used to achieve displacement resolution and locking adjustment better than 10 nanometers under the control of the first control loop.

6. The fused silica substrate integrated lidar transceiver optical module of claim 5, wherein, The micro-adjustment device also includes a multi-mode piezoelectric drive unit, which includes stacked piezoelectric ceramics for driving the large-stroke scanning unit and shear piezoelectric ceramics for driving the high-precision micro-motion unit.

7. The fused silica substrate integrated lidar transceiver optical module of claim 3, wherein, The first control loop also includes a dynamic reference voltage generation link formed by an adder counter and a D / A converter; the dynamic reference voltage is input to the non-inverting input of the control comparator and is used to compare with the second signal from the first central signal gating and routing unit to generate an error voltage signal.

8. The fused silica substrate integrated lidar transceiver optical module of claim 6, wherein, The micro-adjustment device also includes a motion stage and an in-situ sensing system. The motion stage is equipped with a precision interface for mounting optical components and integrates position sensing markers; a corresponding position sensor is installed on the substrate, forming a closed loop for position measurement.

9. A method for adjusting an integrated lidar transceiver optical module on a fused silica substrate, implemented using the integrated lidar transceiver optical module on a fused silica substrate as described in any one of claims 1 to 8, characterized in that... It includes the following steps: Step 1: After the system is powered on or receives a calibration command, initialize all components; Step 2: Input the first signal into the adjustment comparator and compare it with the preset offset threshold; if it is determined to be offset, the adjustment comparator outputs an adjustment request signal; at the same time, the first control loop determines whether adjustment is allowed based on the system operating status, and activates the system adjustment enable clock generator when it is allowed. Step 3: The control of the adjustment request signal and the system adjustment is executed by performing a logical "AND" operation on the clock pulse output by the clock generator in an AND gate; a valid adjustment execution command is generated only when both are valid simultaneously. Step 4: Use the second signal as the feedback voltage input to control the comparator, compare it with the dynamic reference voltage at its non-inverting input, and generate an error voltage signal; An effective adjustment execution command control multi-channel adjustment command distributor routes the error voltage signal to the micro-adjustment device drive circuit corresponding to the current diagnostic mode; Step 5: The micro-adjustment device drive circuit that receives the error voltage signal drives the corresponding micro-adjustment device to adjust the pose of the corresponding optical element; at the same time, the first control loop dynamically adjusts the dynamic reference voltage according to the feedback to form a closed-loop control until the error voltage signal meets the locking condition, so that the current diagnostic segment reaches the optimized alignment state. Step six: after completing the optimization of the current diagnosis paragraph, switch to the next diagnosis mode, and the system repeats steps one to five to adjust each light path paragraph in turn; if the adjustment cannot converge in any mode, trigger a larger range of search dominated by the second control loop.

Images