Straight arm slot machining path planning method and system based on heterogeneous double-head laser cooperation

Abstract

The present disclosure provides a straight arm slot machining path planning method and system based on heterogeneous double-head laser cooperation, relating to the field of precision manufacturing technology, including obtaining a processing workpiece, decomposing the machining task into rough machining and finishing, and assigning them to high-energy galvanometer and high-precision rotary cutting head respectively, constructing a machining path planning and scheduling model based on time consumption, cooling time constraint of heat strategy and machining safety space constraint, determining the feasible domain space, solving the machining path planning and scheduling model by using an improved hybrid evolutionary algorithm, obtaining the access topological sequence of the galvanometer head and the rotary cutting head among all workpieces, and finely arranging the cooperation gap of the two on the unified time axis; constructing an intelligent feedback closed-loop strategy, retaining all parameters in the planned machining path for subsequent adaptive optimization. The present disclosure realizes efficiency optimization, thermal deformation control and safety collision avoidance in large-scale collaborative machining of multiple workpieces, and continuously improves process consistency through data closed loop.

Description

Technical Field

[0001] This disclosure relates to the field of precision manufacturing technology, specifically to a method and system for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] In the field of precision manufacturing, components with high aspect ratio straight-arm slot structures (such as various precision waveguide devices, heat dissipation substrates, microchannel reactors, etc.) have extremely high requirements for processing accuracy and surface quality.

[0004] Currently, existing fabrication methods for this type of structure have the following limitations: (1) The processing of this type of structure mainly adopts precision machining or single laser processing. Precision machining has problems such as tool wear, deformation caused by cutting force, and difficulty in processing hard and brittle materials; single laser processing faces inherent defects such as large heat-affected zone, difficulty in controlling side wall taper, and energy attenuation at the bottom of deep groove.

[0005] (2) If dual laser heads are used for collaborative processing, the existing solutions are mostly simple parallel or sequential operations, lacking a precise task allocation mechanism for different processing characteristics, making it difficult to balance the material removal efficiency of rough processing and the sidewall perpendicularity requirements of fine processing.

[0006] (3) In addition, in the scenario of large-scale continuous production of multiple workpieces, the existing path planning method only considers the shortest movement distance, ignoring the deformation risk caused by heat accumulation and the safety interference problem of parallel operation of two heads, resulting in low actual processing efficiency, high scrap rate and insufficient equipment utilization. Summary of the Invention

[0007] To address the aforementioned issues, this disclosure proposes a method and system for straight-arm groove machining path planning based on heterogeneous dual-head laser collaboration. The machining tasks are decoupled and assigned to a high-energy galvanometer and a high-precision rotary cutting head, respectively. Dynamic path planning is then performed by combining thermal effect strategies and safety constraints. This achieves efficiency optimization, thermal deformation control, and collision avoidance in large-scale collaborative machining of multiple workpieces, and continuously improves process consistency through data closed-loop processing.

[0008] According to some embodiments, the present disclosure adopts the following technical solutions: A method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration includes: The process involves acquiring the workpiece to be processed, extracting features from the workpiece, and classifying the parts of the workpiece accordingly. Based on the classification results, the processing tasks are decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head. The machining parameters are determined and the roughing process time is predicted based on the machining parameters. A thermal field analysis is established, and an adaptive strategy decision on heat utilization and heat avoidance is made according to the characteristics of the parts. Cooling time constraints are introduced to stabilize the finishing temperature field. The finishing rotary cutting dynamic parameters are optimized to determine the finishing process time. A look-ahead sliding time window with a threshold is introduced to perform machining trajectory pre-simulation and to construct machining safety space constraints. Based on the time consumption of roughing and finishing, the cooling time constraint of thermal strategy, and the machining safety space constraint, a machining path planning and scheduling model is constructed, and the feasible domain space is determined. An improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model, obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and finely arrange the matching gap between the two on a unified time axis. A smart feedback closed-loop strategy is constructed to retain all parameters in the planned processing path for subsequent adaptive optimization.

[0009] According to some embodiments, the present disclosure adopts the following technical solutions: A straight-arm groove machining path planning system based on heterogeneous dual-head laser collaboration includes: The heterogeneous task decoupling module is used to acquire the workpiece to be processed, extract features from the workpiece, and classify the parts of the workpiece. Based on the classification results, the processing task is decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head. The constraint construction module is used to determine the machining parameters and predict the roughing process time based on the machining parameters, establish thermal field analysis, make adaptive strategy decisions on heat utilization and heat avoidance according to the characteristics of the parts, introduce cooling time constraints to stabilize the finishing temperature field, optimize the finishing rotary cutting dynamic parameters, determine the finishing process time, introduce a look-ahead sliding time window with a threshold to perform machining trajectory pre-simulation, and construct machining safety space constraints. The path planning module is used to construct a machining path planning and scheduling model based on the time consumption of roughing and finishing, the cooling time constraint of the thermal strategy, and the machining safety space constraint, and to determine the feasible domain space. The improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model to obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and to finely arrange the matching gap between the two on a unified time axis. The feedback optimization module is used to build an intelligent feedback closed-loop strategy, retaining all parameters in the planned processing path for subsequent adaptive optimization.

[0010] According to some embodiments, the present disclosure adopts the following technical solutions: A computer program product includes a computer program that, when executed by a processor, implements the aforementioned method for planning the straight arm groove machining path based on heterogeneous dual-head laser collaboration.

[0011] According to some embodiments, the present disclosure adopts the following technical solutions: A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration.

[0012] According to some embodiments, the present disclosure adopts the following technical solutions: An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the above-described straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration.

[0013] Compared with the prior art, the beneficial effects of this disclosure are as follows: The disclosed method for straight arm groove processing path planning based on heterogeneous dual-head laser collaboration automatically distinguishes between volume removal features and surface trimming features, assigning roughing to a high-energy galvanometer and finishing to a high-precision rotary cutting head. This ensures both material removal efficiency and meets the stringent requirements for sidewall perpendicularity, solving the problem that a single process cannot achieve both.

[0014] This disclosed method for straight-arm groove machining path planning based on heterogeneous dual-head laser collaboration dynamically decides between heat utilization and heat avoidance according to workpiece characteristics: for heat-sensitive workpieces, it forces a cooling window to suppress thermal deformation; for heat-insensitive workpieces, it utilizes residual heat to improve efficiency. For the first time, thermal effects are treated as an active constraint rather than a passive result in path planning, achieving a balance between quality and efficiency.

[0015] The disclosed method for straight arm groove machining path planning based on heterogeneous dual-head laser collaboration establishes a dynamic safety envelope and optical occlusion prediction model, and pre-sets mechanical obstacle avoidance and optical safety as hard constraints for path planning, fundamentally eliminating the risk of dual-head interference and ensuring reliable operation of continuous production.

[0016] This invention discloses a path planning method for straight arm slot processing based on heterogeneous dual-head laser collaboration. It constructs a composite cost matrix that includes motion, process, thermal constraints, and safety costs, and solves the optimal scheduling scheme through the MHGTSP (Multi-Constraint Heterogeneous Generalized Traveling Salesman Problem) model, thereby improving the equipment utilization rate of dual-head asynchronous parallel operation.

[0017] The disclosed method for straight arm groove machining path planning based on heterogeneous dual-head laser collaboration, through full-process parameter retention and sub-model feedback mechanism, continuously optimizes the model at each stage as the number of machining batches increases, forming an evolutionary capability that becomes more accurate with each machining, and significantly improving batch machining consistency. Attached Figure Description

[0018] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0019] Figure 1 This is a flowchart of a straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration, according to an embodiment of this disclosure. Detailed Implementation

[0020] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0022] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0023] Example 1 One embodiment of this disclosure provides a method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration. The method includes the following steps: Step 1: Obtain the workpiece to be processed, extract features from the workpiece, and classify the parts of the workpiece. Step 2: Based on the classification results, the processing task is decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head; Step 3: Determine the machining parameters and predict the roughing process time based on the machining parameters, establish a thermal field analysis, make adaptive strategy decisions on heat utilization and heat avoidance according to the characteristics of the parts, introduce cooling time constraints to stabilize the finishing temperature field; optimize the finishing rotary cutting dynamic parameters, determine the finishing process time, introduce a look-ahead sliding time window of the threshold to perform machining trajectory pre-simulation, and construct machining safety space constraints; Step 4: Based on the time consumption of roughing and finishing, the cooling time constraint of the thermal strategy, and the machining safety space constraint, construct a machining path planning and scheduling model, determine the feasible domain space, and use an improved hybrid evolutionary algorithm to solve the machining path planning and scheduling model to obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and finely arrange the matching gap between the two on a unified time axis. Step 5: Construct an intelligent feedback closed-loop strategy to retain all parameters in the planned processing path for subsequent adaptive optimization.

[0024] As one embodiment, this disclosure presents a path planning method for straight arm groove machining based on heterogeneous dual-head laser collaboration. The machining task is decomposed into roughing and finishing, which are respectively assigned to a high-energy galvanometer and a high-precision rotary cutting head. Dynamic path planning is then performed by combining thermal effect strategies and safety constraints. This method achieves efficiency optimization, thermal deformation control, and collision avoidance in large-scale collaborative machining of multiple workpieces, and continuously improves process consistency through data closed-loop processing. The specific implementation process is as follows: Step 1: Obtain the workpiece to be processed, extract features from the workpiece, and classify the parts of the workpiece; based on the classification results, decompose the processing task into roughing and finishing, and assign different processing tasks to the high-energy galvanometer and the high-precision rotary cutting head; Specifically, the process of distinguishing between the "volume removal features" and "surface finishing features" of a workpiece includes: For each straight arm groove feature to be processed, the geometric parameters such as wall thickness and aspect ratio required for roughing and finishing are extracted simultaneously. The algorithm input is a 3D CAD model of the workpiece. First, curvature analysis or template matching methods are used to locate all straight arm groove structures. For each groove feature, its total volume, layer removal thickness, and other roughing geometric parameters (i.e., volume removal features) are calculated for subsequent galvanometer path scanning, filling, and layer planning. At the same time, finishing geometric parameters such as the normal variation of the sidewalls, taper distribution, and finishing allowance are extracted for the generation of the beam precession trajectory of the rotary cutting head.

[0025] Furthermore, based on the identified classification results and features, precise data inputs are assigned to subsequent heterogeneous laser heads. The galvanometer system, due to its rapid deflection and dynamic characteristics, is assigned high-energy, high-speed scanning tasks, aiming to quickly penetrate the basic contour using pulse superposition effects to achieve rough machining of the straight arm groove. The rotary cutting head receives high-precision commands and utilizes the beam precession characteristics to correct the sidewalls, eliminating the taper and heat-affected zone left by the galvanometer machining, achieving fine machining of the straight arm groove. This decoupling based on feature attributes ensures a high degree of consistency between the machining path and the design intent. The machining targets corresponding to the feature types and the assigned laser heads are shown in Table 1.

[0026] Table 1 Decoupling Classification of Feature Types

[0027] Step 2: Determine the machining parameters and predict the roughing process time based on the machining parameters, establish a thermal field analysis, make adaptive strategy decisions on heat utilization and heat avoidance according to the characteristics of the parts, introduce cooling time constraints to stabilize the finishing temperature field; optimize the finishing rotary cutting dynamic parameters, determine the finishing process time, introduce a look-ahead sliding time window of the threshold to perform machining trajectory pre-simulation, and construct machining safety space constraints; This disclosure divides the implementation of step 2 into four stages, the details of which are as follows: Phase 1: Prediction model for galvanometer roughing parameters and conservative removal strategy; In the roughing stage, parameter settings directly affect the allowance distribution and overall heat accumulation in subsequent finishing. Considering the material properties and the geometry of the straight arm groove, the system adopts a process flow of "offline testing - optimal application - conservative strategy": For new models or workpieces being processed for the first time, several candidate parameter combinations are verified through rapid marking tests before formal machining. The optimal parameters are then selected and applied to roughing, with a finishing allowance reserved. If similar cases exist in the historical database, the existing optimal parameters are directly called, skipping the testing phase.

[0028] Specifically, the process of generating candidate processing parameter combinations includes: First, the process knowledge base is searched to find historical machining cases similar to the current straight arm groove characteristics (groove depth, groove width, depth-to-width ratio, material batch). If a case with a similarity exceeding a set threshold (e.g., 90%) exists, the optimal parameter combination of that case is directly used as the rough machining benchmark, without the need for repeated testing.

[0029] Secondly, for novel structures lacking historical data, a set of candidate parameter combinations is generated based on machine learning prediction models (such as pre-trained Gaussian process regression or neural network models) or inference engines incorporating expert experience. The machine learning model takes the target groove depth, groove width, and material properties as input, and outputs predicted laser power, scanning speed, filling spacing, and number of scans, along with a prediction confidence interval. The expert experience base uses case-based reasoning (CBR) technology to match several recommended parameter sets from historical processes with similar geometric features. The parameters from these two sources are merged and filtered based on the diversity of parameter distribution, ultimately retaining 3-5 candidate combinations covering typical process windows (e.g., one biased towards high efficiency, one towards low thermal impact, and one in between) for subsequent testing.

[0030] Furthermore, rapid labeling tests and multi-objective optimization are conducted, including: On the edge scrap area of ​​the final workpiece or on the accompanying pieces of the same batch, short-stroke straight-arm groove roughing simulations are performed sequentially according to candidate parameter combinations. Each combination is repeated 2-3 times to evaluate repeatability. After machining, the following key indicators are measured offline using a laser confocal microscope: actual removal depth, groove bottom surface roughness, and sidewall taper angle. Grey relational analysis or weighted scoring method is used to integrate the above multi-objective indicators into a comprehensive score. The candidate parameters are ranked according to the score, and the optimal machining parameter combination is selected. Then, the material removal situation in this process is input into the model of the next stage.

[0031] All data from the testing and optimization process (including candidate parameters, measured removal depth, roughness, heat-affected zone, etc.) are stored in the process knowledge base for subsequent retraining of machine learning models and enrichment of the expert experience base. Simultaneously, this stage outputs the roughing prediction time calculated based on the optimal processing parameters, serving as the time benchmark input for subsequent global path scheduling.

[0032] Phase Two: Processing Strategy Decisions Based on Thermal Effects; First, the thermodynamic state response and bilinear strategy determination are performed: Thermal effects are often considered uncontrollable passive disturbances in traditional machining, easily leading to material warping and loss of precision. The core technological breakthrough of this disclosure lies in the proactive establishment of thermal field analysis, and the adaptive strategy game between "thermal utilization" and "thermal avoidance" based on the characteristics of the parts.

[0033] The established adaptive strategy game is not a simple binary choice, but a comprehensive decision-making process based on multi-physics quantitative indicators. Using the wall thickness, aspect ratio, material thermal diffusivity, structural stiffness, and sidewall verticality requirements obtained in the workpiece feature extraction stage as inputs, a thermal effect risk scoring function is constructed, and a dynamic game and trade-off is carried out between "thermal utilization" and "thermal avoidance" based on this function.

[0034] Specifically, the thermal sensitivity index of each workpiece is first calculated. It is defined as the weighted normalized sum of minimum wall thickness, aspect ratio, material thermal diffusivity, and structural stiffness, with the weights preset based on process experience. The thermal sensitivity index... A larger value indicates that the workpiece is less sensitive to thermal deformation. Two thresholds were determined through offline experiments or historical data analysis. and (and ):like This is determined to be a "heat utilization" strategy, actively retaining the residual heat generated during the rough machining of the galvanometer to improve the absorption rate of the subsequent spin-cutting laser, without the need for cooling; if This is determined to be a "heat avoidance" strategy, forcibly inserting a cooling window between roughing and finishing until the temperature returns to room temperature; if Then, the system enters the game arbitration phase, where it simultaneously evaluates efficiency gains and deformation risks, introducing a dynamic cost function. ,in Cooling time required to avoid heat Weighting the benefits of increased absorption rate from heat utilization The system dynamically adjusts based on current production cycle requirements, selecting to... A smaller strategy is adopted. After the decision is made, the hot strategy label is passed to the global path planning module, and the weights and thresholds are adjusted online based on actual detection data after each batch of processing is completed. This achieves adaptive evolution of the strategy game, thereby maximizing the efficiency potential of high-temperature assisted processing while ensuring accuracy.

[0035] The heat utilization strategy involves using a galvanometer to recover the residual high-temperature thermal field from rough machining of workpieces with thick walls, high structural rigidity, and wide tolerance to microscopic thermal deformation. Based on the Drude free electron model and the dual-temperature model of ultrashort pulse lasers, an increase in the lattice temperature of metallic materials significantly intensifies electron-phonon coupling and scattering frequency, thereby reducing the macroscopic reflectivity of the material to 1064nm and other wavelength lasers. At high temperatures, the optical absorption rate of oxygen-free copper to subsequent rotary cutting laser energy increases dramatically, and the ablation threshold decreases. This can multiply the material removal efficiency of finishing processes, significantly shortening the overall manufacturing time.

[0036] For precision components with extremely high microscopic dimensional accuracy, thin walls, and "zero tolerance" for thermal deformation, continuous injection of high-temperature energy will induce irreversible thermoelastic warping and structural stress concentration. For such workpieces, an asynchronous parallel "thermal avoidance" machining mode is enforced. In this mode, after roughing, the workpiece does not immediately proceed to finishing; instead, it undergoes forced air cooling or awaits natural cooling. The rotary cutting head intervenes with a lag, ensuring that the workpiece's temperature and stress fields have fully relaxed and returned to a stable ground state close to room temperature when it cuts the sidewalls. This strategy fundamentally avoids the periodic cumulative errors caused by thermal deformation.

[0037] Furthermore, once a heat avoidance strategy is determined, cooling time must be precisely incorporated into the global path planning. This calculation cannot rely on rough empirical estimates but must establish rigorous mathematical equations for transient heat conduction.

[0038] According to the classical Fourier law of heat conduction and the theory of convective heat transfer, the transient temperature field decay of a workpiece is mainly dominated by the thermal diffusivity within the material and the convective heat transfer on its surface. Combining the lumped parameter method and one-dimensional laser thermal relaxation theory, the cooling time is calculated. The complete formula is defined as follows:

[0039] in, The density of the material is (kg / m³). The isobaric specific heat capacity (J / (kg·K)) h The comprehensive convective heat transfer coefficient (W / (m²·K)) of the workpiece surface is used to distinguish between natural cooling and externally assisted forced air cooling. A and V are the effective convective heat transfer surface area and geometric volume of the workpiece, respectively.

[0040] Based on cooldown time Based on Newton's law of cooling and its exponential decay principle, the evolution of the workpiece's transient temperature is predicted. This is to ensure that the overall workpiece temperature is suppressed within a set safe isothermal threshold when the rotary cutting finishing head intervenes. (usually strictly defined as workshop ambient temperature) Within this timeframe, the cooling window time that is forcibly inserted in subsequent path planning. The deduction is as follows:

[0041] in, This refers to the highest peak workpiece temperature obtained at the moment of completion of the galvanometer roughing process, through thermodynamic finite element analysis (or offline sensor calibration). Using this set of rigorous equations built upon a profound foundation in thermophysics, the system quantifies the abstract concept of "thermal avoidance" into concrete millisecond-level time constraints. .

[0042] In the subsequent MHGTSP path planning, this value is transformed into a hard time window constraint, thereby achieving the ultimate optimization of equipment efficiency while ensuring absolute accuracy. After completing the galvanometer roughing and before calculating the finishing parameters, a decision must first be made between the "heat utilization" and "heat avoidance" strategies based on the workpiece characteristics and material thermophysical properties. This is because the temperature field significantly changes the material's absorption rate of laser light, thus affecting the modeling of subsequent rotary cutting parameters.

[0043] Phase 3: Remaining margin analysis and dynamic setting of rotary cutting head precession parameters; After the roughing and thermal strategy decision-making loop is completed, since the "thermal utilization" or "thermal avoidance" strategy in the previous stage has completely changed the optical absorption boundary of oxygen-free copper, it is necessary to assign completely different dynamic parameters to the rotary cutting head.

[0044] This disclosure establishes a unified three-dimensional kinematic overlap rate model and achieves precise control of extreme thermophysical effects by dynamically setting the upper and lower limits of the overlap rate.

[0045] During rotary cutting and finishing, the absolute scanning linear velocity of the beam spot on the sidewall is a vector composed of the high-frequency internal beam precession and the macroscopic platform feed. The global theoretical overlap rate calculation formula is:

[0046] in, Where is the shear radius, The frequency of the rotary shearing is... To improve the platform's feed speed, The laser repetition frequency, To determine the focused spot diameter. Based on the thermodynamic state of the third-stage decision, this unified formula is differentiated into two sets of rigid constraints: (1) Upper limit constraint of anti-overcutting overlap rate for matching hot utilization strategy; In the "thermal utilization" mode, the workpiece maintains a high macroscopic temperature rise, and electron-phonon coupling absorption is intensely excited, leading to a high equivalent laser ablation threshold for the material. (T) drops sharply. To prevent catastrophic sidewall overcutting and thermal collapse caused by single-pulse removal of volume expansion, a low overlap rate / high feed control strategy must be implemented. A strict dynamic upper limit is set for the overlap rate based on the transient temperature field T:

[0047] Among them, the upper limit threshold It is negatively correlated with temperature. Under this constraint, the algorithm is forced to significantly increase the scanning feed rate of the mechanical platform. By actively increasing the pulse spacing, while utilizing the high removal efficiency under high temperature conditions, further local heat accumulation is cut off, thereby achieving a leap in the finishing cycle while ensuring the quality of the base surface.

[0048] (2) Matching the lower limit constraint of mirror-level overlap rate for the "hot avoidance" strategy; When the system decision is set to "thermal avoidance" mode, the workpiece has already undergone forced cooling and returned to its cold room temperature. At this point, oxygen-free copper regains its extremely high reflectivity, and the processing mechanism completely degenerates into "hard cutting" relying on the rotary cutting optical system. To achieve a mirror-level extremely low surface roughness (Ra) under these extremely demanding optical conditions, a hard constraint on the lower limit of the overlap rate must be applied:

[0049] Among them, the lower limit threshold This is typically rigidly locked above a threshold. To achieve fine "micro-grinding" of the sidewalls while satisfying this inequality, the feed rate must be drastically reduced. And matched with extremely high laser repetition frequencies in the megahertz (MHz) range. This ensures that every micron-level area on the sidewall of the straight arm slot receives extremely uniform pulse coverage.

[0050] Within the parameter space that satisfies the upper and lower limits of the overlap rate constraint, the feed rate is maximized. Local optimization is performed to target the final rotary cutting dynamics parameters, and the precise finishing process time is calculated. This is then fed forward as a deterministic quantity into the subsequent path planning and scheduling matrix.

[0051] Phase Four: Pre-setting the processing strategy for the protection mechanism; In modern high-density, large-scale, multi-workpiece processing scenarios, two independent laser actuation ends (a bulky, high-precision rotary cutting head and a galvanometer system that performs high-speed dynamic displacement) operate asynchronously and in parallel within a very small gap on the same pallet plane. If only traditional collision-based alarm sensors are relied upon, irreversible equipment damage or optical component contamination can easily occur.

[0052] This disclosure innovatively fully mathematizes and spatializes these protection mechanisms, transforming them into rigid mathematical constraints in the global path planning stage, thereby achieving inherent security through prior knowledge. The specific details are as follows: (1) Geometric modeling of dynamic safety envelope and mechanical obstacle avoidance: To prevent mechanical interference at the algorithmic level, the Convex Set Envelope Theory is used to perform dynamic geometric modeling of the physical entities. The outer contours of the galvanometer laser head and the rotary cutting head are mapped to their respective time-space convex set envelopes. and .

[0053] At any processing time t, the safety space constraint between the two heterogeneous laser heads can be uniquely expressed by calculating the extreme Euclidean distance between the two convex sets:

[0054] To ensure absolute safety in a kinematic and dynamic sense, this shortest distance must not only be greater than the inherent physical static safety clearance threshold of the equipment. Furthermore, it is necessary to dynamically compensate for the following lag error of the servo motor control system and the enormous inertia of the moving components. Therefore, a dynamic redundancy term strongly correlated with the current operating speed is defined. The mechanical safety collision avoidance constraint is ultimately defined by the following inequality:

[0055] Among them, dynamic redundancy items It deeply integrates predictive control theory, and focuses on the time constant of the control system. Integral construction is performed within the timeframe (e.g., communication jitter and servo settling time on the EtherCAT bus).

[0056] in, and These respectively represent the two laser heads in t The three-dimensional spatial velocity vector at time t. This represents the maximum emergency braking deceleration of the CNC system. In actual execution, a look-ahead sliding time window with a threshold is introduced for trajectory pre-simulation. If this safety inequality is detected to be violated during trajectory pre-simulation, the planner will immediately truncate the current sequence, triggering task priority rearrangement or emergency execution of a high-speed lifting and avoidance maneuver on the Z-axis.

[0057] (2) Three-dimensional view cone intersection prediction algorithm for optical occlusion: In a dual-head collaborative system, besides the physical collisions of the mechanical housing, a more insidious and fatal danger lies in the fact that the bulky rotary cutting head can easily intrude into and sever the downward convergence of the high-energy optical path of the galvanometer system during complex movements. This can not only render the current workpiece unusable, but may even burn out the precision lenses of the rotary cutting head due to the reverse diffuse reflection of the powerful laser.

[0058] To address this, a 3D view frustum intersection detection algorithm from advanced computer graphics was introduced to construct the optical occlusion model. Let the origin of the luminous point of the galvanometer optical system be... Its instantaneous three-dimensional spatial velocity vector is The maximum deflection half-angle corresponding to the scanning field is At that moment, the envelope of the beam produced by the galvanometer can be defined as a three-dimensional spatial set of view frustum geometry. :

[0059] Meanwhile, let the physical impact protection shell envelope of the rotary cutting head be a set. The system cleverly transforms the complex optical security check into a set theory problem of finding the intersection for verification. The absolute security constraint of unobstructed optical movement is rigorously defined as the intersection of two sets at any time t being an empty set:

[0060] If the planning simulation finds that this empty set constraint is not satisfied (i.e., the intersection is not empty, indicating that the optical path is blocked), the planner will immediately extract a penalty cost term that tends to infinity and feed it back to the global optimizer. This will force the scheduling algorithm to shift the corresponding galvanometer processing operation backward in the time dimension, or adjust the relative docking position of the large XY mechanical axes in the spatial dimension, thereby completely eliminating the hidden danger of the optical path being blocked at its physical root.

[0061] Step 3: Based on the time consumption of roughing and finishing, the cooling time constraint of the thermal strategy, and the machining safety space constraint, a machining path planning and scheduling model is constructed, and the feasible domain space is determined. The improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model to obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and the matching gap between the two on the unified time axis is finely arranged. Specifically, based on all the physical parameters generated in the above two steps—including basic process time, cooling window constraints of thermal strategies, and mechanical and optical obstacle avoidance constraints—they are ultimately used in the global path planning algorithm.

[0062] When a pallet contains dozens or even hundreds of microchannel / straight-arm slotted workpieces arranged in a distributed manner, simple sequential processing inevitably leads to extremely high equipment idle rates and excessive local heat concentration. This disclosure mathematically abstracts this complex workshop scheduling scenario into a "multi-constraint heterogeneous generalized traveling salesman problem" with extremely high dimensionality and constraint density. The model needs to simultaneously solve for the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and finely arrange the matching gaps between the two on a unified time axis. The specific details are as follows: (1) Construct a processing path planning and scheduling model; The core premise of the MHGTSP (Multi-Constraint Heterogeneous Generalized Traveling Salesman Problem) solution model is to integrate multivariate physical rules and three-dimensional spatial topological dimensionality reduction into a single mathematical cost matrix. Let the cost matrix be... The global total cost required to use a specific laser head k (k=1 represents roughing with a galvanometer, k=2 represents finishing with a rotary cutting head) when the state transitions from workpiece i to workpiece j is expressed by the following mathematical formula:

[0063] The first three terms in the formula constitute the basic time cost of scheduling. The first term is the speed of the mechanical axis. Overcoming Euclidean distance Idle transfer time; The inherent "handshake" and data alignment delay that occurs when switching from the galvanometer process to the rotary cutting process within the same workpiece; This refers to the actual physical cutting time of the corresponding laser head, as output in the previous text. , , The normalized dimensionless weighting coefficients.

[0064] Thermal constraint cost penalty function : Dynamically generated based on the thermal strategy determination result of step 2. For workpiece i marked with the "thermal avoidance" strategy, if its absolute timestamp of rough machining is completed... Absolute timestamp of the start of finishing process with rotary cutting Time difference between Smaller than the calculated forced cooling window The function then uses the Big M method to return an infinite penalty (i.e., This measure completely blocks the generation of such dangerous paths from the algorithm's underlying layer, making the thermal constraint an insurmountable red line.

[0065] Security Conflict Cost Penalty Function The prediction model is established based on the fourth stage of step 2. If the simulation shows that the mechanical interference threshold is broken at any time during the transition to node j ( Or optical obstruction occurred ( ),but .

[0066] Further, the feasible region is determined, and the process of determining the feasible region includes: Due to the vast search space, this disclosure employs a hierarchical filtering strategy to gradually narrow down the feasible region: 1. Safety constraint pre-filtering: Based on the envelope model in step 2, pre-filtering is performed on all workpiece pairs. The parallel processing possibilities between different heads are evaluated offline. A safety compatibility matrix is ​​generated, marking which workpiece pairs can be processed by both heads simultaneously (i.e., the spatial layout allows for parallel operations without interference) and which must be processed serially. This matrix is ​​used to avoid dual-head conflicts in subsequent scheduling.

[0067] 2. Thermal Strategy Constraint Transformation: Based on the thermal strategy, an earliest finishing start time is generated for each workpiece (for thermal avoidance workpieces, this is equal to the roughing completion time plus the forced cooling window time; for thermal utilization workpieces, this is equal to the roughing completion time). This time is added as a hard constraint to the scheduling model.

[0068] 3. Dynamic Cost Matrix Generation: Combining motion overhead, task handover overhead, process time, and the constraints mentioned above, a complete cost matrix is ​​constructed. For paths that violate the constraints, the cost is set to infinity.

[0069] (2) Hybrid evolutionary solution and dynamic local rearrangement; After constructing the aforementioned high-dimensional, nonlinear feasible domain space, this disclosure employs a hybrid evolutionary algorithm for efficient optimization.

[0070] In the initial stage of the algorithm, a large population encompassing both galvanometer and rotary cutting access sequences is randomly generated. During the fitness evaluation phase, the time-series simulator performs millisecond-level extrapolation calculations for each individual chromosome. Because an infinite penalty term is embedded in the cost matrix, any sequence that violates the hot strategy red line or the spatiotemporal safety red line will be assigned an extremely low fitness and quickly eliminated from the gene pool.

[0071] Meanwhile, the algorithm incorporates heuristic local search operators (such as the improved 2-opt algorithm) and dynamically adjusts the weight coefficients during the evolutionary generations. To dynamically balance the load rates of the galvanometer and the rotary cutting head, preventing one of the devices from becoming the bottleneck of the overall production cycle.

[0072] Even more advanced is the ability to address environmental disturbances commonly found in actual workshop environments (such as laser output power drift and actual cooling time caused by workshop ambient temperature fluctuations). To mitigate biases, a highly agile local time window rescheduling mechanism is retained on the foundation of offline global optimization. If the online infrared sensor array detects that the actual cooling rate of a workpiece lags behind the theoretical model, it will not trigger a global replay that would cause computing power collapse. Instead, it will only freeze and extract the scheduling subgraph within a future time window (such as within the next 5000ms, or the five adjacent workpiece queues thereafter), and perform local minor topology corrections and heuristic stitching. This mechanism ensures sub-second industrial response latency while maximizing the optimality of global scheduling.

[0073] Furthermore, since laser power fluctuations, material differences, or sudden interruptions may occur during actual processing, the system retains real-time replanning capabilities on top of offline optimization: To address the computational latency issue of global replay under high-dimensional constraints, this disclosure introduces a local time window rearrangement mechanism: The system monitors the actual temperature of each workpiece online (using thermal sensors). If the cooling rate of a workpiece deviates from expectations, resulting in an extended cooling window, a global recalculation is not performed. Instead, the scheduling sequence within a future time window (e.g., the next 5000ms or the next 5 workpieces) is extracted for local heuristic correction. If the fourth stage safety monitoring system in step 2 predicts potential interference, it will trigger an emergency collision avoidance action (e.g., Z-axis lifting) and reallocate priorities only for the currently conflicting "dual-head-workpiece" subset. The adjusted local state is then integrated into the original global planning path, maintaining overall efficiency while ensuring sub-second response.

[0074] As one example, for large slow-wave structures or large-size RF substrates whose length exceeds a threshold and is beyond the coverage of a single field lens, a single optical scan is insufficient to achieve global coverage. The traditional "stepping + scanning" mode leaves fatal micro-steps and seams at each field-to-field stitching point.

[0075] To this end, this disclosure deeply integrates Infinite Field of View (IFOV) synchronization technology at the execution layer. Customized spatial kinematic decoupling strategies are designed for heterogeneous dual-head systems: the originally continuous long trajectory G-code is decomposed into two independent vector components using a low-level interpolation algorithm. 1. The XY mechanical platform, driven by an ultra-high precision linear motor, performs long-distance, low-frequency, and smooth macroscopic reference feed.

[0076] 2. The high-frequency, extremely short-stroke microscopic complex contour deflection scanning is performed by the lightweight optical lenses inside the galvanometer or rotary cutter head.

[0077] These two distinct mechanical and electrical systems are synchronized at high frequency and compensated for feedforward errors within the same sub-millisecond clock domain via a distributed clock on an industrial-grade EtherCAT bus. This ensures that the absolute positioning accuracy of the high-energy laser spot on the workpiece surface remains below the micrometer level. Even during the high-speed continuous operation of the chassis mechanical shaft, the cross-field machining of the long straight arm groove can still achieve a seamless and smooth connection of shape and taper, completely eliminating splicing errors.

[0078] Step 4: Construct an intelligent feedback closed-loop strategy to retain all parameters in the planned processing path for subsequent adaptive optimization; The final stage of the path optimization process disclosed herein is to build an intelligent feedback loop, which is the key to achieving continuous process optimization.

[0079] From the feature decoupling results, candidate parameters and measured optimal data in the first step, to the thermal strategy decision and cooling time constant in the second step, to the trial cutting optimal results of rotary cutting precession parameters and dynamic safety envelope data, to the path planning cost matrix and execution records in the third step—all the input parameters, intermediate variables, final results and corresponding workpiece batch information of all stages are completely retained and structured and stored in the process knowledge base.

[0080] The retained data is input into the corresponding prediction models step by step: Step 1 Galvanometer parameter prediction model: Receive the actual removal depth, bottom roughness of the trench and measured values ​​of the heat-affected zone, analyze the prediction deviation through machine learning, automatically correct physical model parameters such as material absorption rate and ablation threshold, or update the training set of the Gaussian process regression model, so that the generation of candidate parameters for subsequent similar features is more accurate.

[0081] Step 2 Thermal strategy decision model: Receive actual cooling curves and post-processing dimensional inspection data, verify the accuracy of the cooling time constant calculation, and optimize the judgment threshold for heat utilization / heat avoidance.

[0082] Rotary cutting parameter setting model: Receives measured sidewall perpendicularity, surface roughness and precession parameters, corrects the precession radius, incident angle and pulse overlap rate dynamic model, and improves the scoring weight of trial cutting selection.

[0083] Safety envelope model: Receives actual motion trajectory and interference records, optimizes the value of dynamic redundancy terms, and improves the sensitivity and accuracy of obstacle avoidance prediction.

[0084] Step 3 path planning model: Receive the actual total processing time, equipment idle time, and thermal constraint satisfaction rate, correct the weight coefficients in the cost matrix and the cooling window forced waiting logic, and improve the efficiency and balance of subsequent batch scheduling.

[0085] This mechanism of full parameter retention and sub-model feedback enables self-evolution after each batch of workpieces is processed. As the number of processing cycles increases, the prediction accuracy of the model at each stage continuously improves, the process knowledge base becomes increasingly rich, and subsequent similar workpieces can gradually reduce offline testing and directly call mature parameters, forming a self-learning closed loop from design to finished product, continuously improving batch processing consistency and equipment utilization.

[0086] Example 2 One embodiment of this disclosure provides a straight arm groove machining path planning system based on heterogeneous dual-head laser collaboration, including: The heterogeneous task decoupling module is used to acquire the workpiece to be processed, extract features from the workpiece, and classify the parts of the workpiece. Based on the classification results, the processing task is decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head. The constraint construction module is used to determine the machining parameters and predict the roughing process time based on the machining parameters, establish thermal field analysis, make adaptive strategy decisions on heat utilization and heat avoidance according to the characteristics of the parts, introduce cooling time constraints to stabilize the finishing temperature field, optimize the finishing rotary cutting dynamic parameters, determine the finishing process time, introduce a look-ahead sliding time window with a threshold to perform machining trajectory pre-simulation, and construct machining safety space constraints. The path planning module is used to construct a machining path planning and scheduling model based on the time consumption of roughing and finishing, the cooling time constraint of the thermal strategy, and the machining safety space constraint, and to determine the feasible domain space. The improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model to obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and to finely arrange the matching gap between the two on a unified time axis. The feedback optimization module is used to build an intelligent feedback closed-loop strategy, retaining all parameters in the planned processing path for subsequent adaptive optimization.

[0087] Example 3 One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration.

[0088] Example 4 One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions. When these computer instructions are executed by a processor, they implement the straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration.

[0089] Example 5 One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the above-described straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration.

[0090] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration, characterized in that, include: The process involves acquiring the workpiece to be processed, extracting features from the workpiece, and classifying the parts of the workpiece accordingly. Based on the classification results, the processing tasks are decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head. The machining parameters are determined and the roughing process time is predicted based on the machining parameters. A thermal field analysis is established, and an adaptive strategy decision on heat utilization and heat avoidance is made according to the characteristics of the parts. Cooling time constraints are introduced to stabilize the finishing temperature field. The finishing rotary cutting dynamic parameters are optimized to determine the finishing process time. A look-ahead sliding time window with a threshold is introduced to perform machining trajectory pre-simulation and to construct machining safety space constraints. Based on the time consumption of roughing and finishing, the cooling time constraint of thermal strategy, and the machining safety space constraint, a machining path planning and scheduling model is constructed, and the feasible domain space is determined. An improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model, obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and finely arrange the matching gap between the two on a unified time axis. A smart feedback closed-loop strategy is constructed to retain all parameters in the planned processing path for subsequent adaptive optimization.

2. The method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration as described in claim 1, characterized in that, The process involves acquiring the workpiece to be processed, extracting features from the workpiece to classify its parts, and based on the classification results, decomposing the processing task into roughing and finishing, and assigning different processing tasks to the high-energy galvanometer and the high-precision rotary cutting head, including: For each straight arm groove feature to be processed, the wall thickness and aspect ratio geometric parameters required for roughing and finishing are extracted simultaneously. The 3D CAD model of the workpiece is input, and all straight arm groove structures are first located using curvature analysis or template matching methods. For each groove feature, its total volume and layer removal thickness are calculated as roughing geometric parameters for subsequent scanning and filling of the galvanometer path and layer planning. At the same time, the normal variation, taper distribution, and finishing allowance of the side wall are extracted as finishing geometric parameters for generating the beam precession trajectory of the rotary cutting head. Based on the identified features, precise data input is assigned to the subsequent heterogeneous laser head. Due to its rapid deflection and dynamic characteristics, the galvanometer system is assigned a high-energy, high-speed scanning task to quickly open up the basic contour using the pulse superposition effect. The rotary cutting head receives high-precision instructions and uses the beam rotation precession characteristics to correct the sidewalls and eliminate the taper and heat-affected zone left by the galvanometer processing.

3. The method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration as described in claim 1, characterized in that, The process of determining machining parameters and predicting roughing time based on these parameters, establishing thermal field analysis, making adaptive strategy decisions on heat utilization and heat avoidance according to part characteristics, and introducing cooling time constraints to stabilize the finishing temperature field includes: In the roughing stage, considering the material properties and the geometric features of the straight arm groove, an offline testing-optimal application-conservative strategy is adopted. For new models or workpieces being processed for the first time, several candidate parameter combinations are verified through rapid marking tests before formal processing. The optimal parameters are then selected and applied to roughing, with a allowance reserved for finishing. If similar cases exist in the historical database, the existing optimal parameters are directly called, skipping the testing stage. After determining the implementation of the heat avoidance strategy, the cooling time is precisely inserted into the global path planning, the transient heat conduction mathematical equation is established, and the evolution of the transient temperature of the workpiece is predicted based on the exponential decay law of Newton's cooling law. The cooling window time is calculated, and when entering the subsequent path planning, the cooling window time is transformed into a time window hard constraint in the time sequence.

4. The method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration as described in claim 1, characterized in that, The optimization of the dynamic parameters for finishing rotary cutting includes: After the roughing and thermal strategy decision-making closed loop is completed, distinctly different dynamic parameters are assigned to the rotary cutting head. By establishing a unified three-dimensional kinematic overlap rate model and dynamically setting the upper and lower limits of the overlap rate, precise control of extreme thermophysical effects can be achieved. During rotary cutting and finishing, the absolute scanning linear velocity of the spot on the sidewall is a vector composed of the precession of the high-frequency internal beam and the feed of the macroscopic platform. Based on the thermodynamic state of the third-stage decision, the global theoretical overlap rate is divided into two sets of rigid constraints. Within the parameter space that satisfies the two sets of rigid constraints, local optimization is performed with the goal of maximizing the feed rate to determine the final rotary cutting dynamic parameters and calculate the precise finishing process time.

5. The method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration as described in claim 1, characterized in that, The proposed look-ahead sliding time window with introduced thresholds is used for processing trajectory pre-simulation to construct processing safety space constraints, including: The outer contours of the galvanometer laser head and the rotary cutting head are mapped to the time-space convex set envelopes by using the convex set envelope theory. At any processing time, the safety space constraint between the two heterogeneous laser heads is represented by calculating the extreme Euclidean distance between the two convex sets. In actual execution, a forward sliding time window with a threshold is introduced to perform trajectory pre-simulation. If the safety inequality is detected to be violated during trajectory pre-simulation, the planner will immediately truncate the current sequence, trigger task priority rearrangement or emergency execution of high-speed lifting and avoidance action on the Z-axis. A 3D view cone intersection detection algorithm from advanced computer graphics is introduced to construct an optical occlusion model. The absolute safety constraint of optical occlusion is defined as the intersection of two sets at any time being an empty set.

6. The method for planning the machining path of a straight arm groove based on heterogeneous dual-head laser collaboration as described in claim 1, characterized in that, A processing path planning and scheduling model is constructed and mathematically abstracted into a multi-constraint heterogeneous generalized traveling salesman problem with extremely high dimensionality and constraint density. The core premise for solving the problem is to integrate multivariate physical rules and three-dimensional spatial topology dimensionality reduction into a single mathematical cost matrix. A hierarchical screening strategy is adopted to gradually narrow down the feasible region. After constructing the feasible region space, an improved hybrid evolutionary algorithm is used for efficient optimization. In the initial stage of the algorithm, a large population covering galvanometer and rotary cutting access sequences is randomly generated. In the fitness evaluation stage, the time-series simulator performs millisecond-level extrapolation calculations for each individual chromosome. Since an infinite penalty term is embedded in the cost matrix, any sequence that violates the hot strategy red line or the spatiotemporal safety red line will be assigned an extremely low fitness and quickly eliminated from the gene pool. A heuristic local search operator is introduced, and the weight coefficients are dynamically adjusted in the evolutionary generations to dynamically balance the load rate of the galvanometer and rotary cutting head, preventing one of the devices from becoming the bottleneck of the global production cycle.

7. A straight arm groove machining path planning system based on heterogeneous dual-head laser collaboration, characterized in that, include: The heterogeneous task decoupling module is used to acquire the workpiece to be processed, extract features from the workpiece, and classify the parts of the workpiece. Based on the classification results, the processing tasks are decomposed into roughing and finishing, and different processing tasks are assigned to the high-energy galvanometer and the high-precision rotary cutting head. The constraint construction module is used to determine the machining parameters and predict the roughing process time based on the machining parameters, establish thermal field analysis, make adaptive strategy decisions on heat utilization and heat avoidance according to the characteristics of the parts, introduce cooling time constraints to stabilize the finishing temperature field, optimize the finishing rotary cutting dynamic parameters, determine the finishing process time, introduce a look-ahead sliding time window with a threshold to perform machining trajectory pre-simulation, and construct machining safety space constraints. The path planning module is used to construct a machining path planning and scheduling model based on the time consumption of roughing and finishing, the cooling time constraint of the thermal strategy, and the machining safety space constraint, and to determine the feasible domain space. The improved hybrid evolutionary algorithm is used to solve the machining path planning and scheduling model to obtain the access topology sequence of the oscillating head and the rotary cutting head among all workpieces, and to finely arrange the matching gap between the two on a unified time axis. The feedback optimization module is used to build an intelligent feedback closed-loop strategy, retaining all parameters in the planned processing path for subsequent adaptive optimization.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration as described in any one of claims 1-6.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the straight arm groove machining path planning method based on heterogeneous dual-head laser collaboration as described in any one of claims 1-6.

Images