Multi-parameter coupling ai machining tool path global optimization method for robot parts

Abstract

The application provides a multi-parameter coupling AI machining tool path global optimization method for robot parts, and relates to the technical field of robot part machining. The method comprises the following steps: switching to a basic tool path before cutting; collecting multiple cutting parameters in the cutting process in real time, and importing the parameters into a preset AI knowledge base; the AI knowledge base compares the collected parameters with a safety baseline, and if it is determined that a parameter is abnormal, the corresponding optimization strategy is automatically matched and executed to correct the tool path. Through real-time and automatic local correction of the tool path by AI, the application effectively suppresses cutting force fluctuations and machining vibration, and through active avoidance, real-time compensation of thermal offset, etc., the surface contour accuracy and mass production qualification rate of robot parts such as harmonic flexspline, RV cycloid wheel, etc. are significantly improved, and the machining efficiency is improved and the tool life is prolonged.

Description

Technical Field

[0001] This application relates to the field of robot component processing technology, specifically to a method for global optimization of toolpaths in multi-parameter coupled AI machining of robot components. Background Technology

[0002] Robot joints are core precision components of industrial robots, and their machining quality directly determines the robot's motion accuracy, lifespan, and overall performance. Key components such as RV cycloidal wheels, harmonic flexible wheels, precision gears, and curved surface parts of robot joints are typically milled using a five-axis CNC machining center. These parts generally have complex structures, thin walls, high material removal rates, and often possess complex geometric features such as non-circular curved surfaces.

[0003] In traditional CNC machining, toolpath planning relies primarily on manual programming based on the experience of process engineers. The basic toolpaths are often fixed trajectories generated from geometric models, making adjustments impossible based on real-time operating conditions during machining. However, in actual cutting processes, factors such as uneven blank allowance, material hardness fluctuations, progressive tool wear, and thermal drift from prolonged machine operation can cause drastic fluctuations in cutting forces, easily leading to resonance or tool breakage. Spindle loads also frequently increase abnormally due to localized high-hardness points. Furthermore, thermal deformation of the machine spindle and workpiece gradually accumulates during continuous machining, resulting in out-of-tolerance profiles of later-machined parts. These problems are particularly pronounced in the thin-walled structures of harmonic flexures, the high-stress areas at the tooth roots of RV cycloidal gears, and the deep groove features of complex articulated surfaces.

[0004] In existing technologies, conservative process parameters are typically used to mitigate the risks associated with the aforementioned problems, but this directly sacrifices processing efficiency. Alternatively, operators rely on their experience to manually stop and adjust the machine in case of anomalies, which not only disrupts the processing cycle but also heavily depends on manual experience, making it difficult to guarantee the stability and consistency of the mass production process. Therefore, how to achieve real-time perception, intelligent decision-making, and adaptive toolpath correction of the processing process to simultaneously achieve high precision, high efficiency, and high mass production stability has become an urgent technical problem to be solved. Summary of the Invention

[0005] In view of this, this application provides a global optimization method for multi-parameter coupled AI machining toolpaths for robot parts, which can achieve real-time perception, intelligent decision-making and adaptive toolpath correction of the machining process, so as to simultaneously achieve high precision, high efficiency and high mass production stability.

[0006] In a first aspect, this application provides a method for global optimization of multi-parameter coupled AI machining toolpaths for robot parts, comprising: switching to a corresponding basic toolpath before the cutting tool performs cutting on the workpiece; real-time acquisition of multiple cutting parameters of the cutting tool when performing cutting with the basic toolpath; importing the multiple cutting parameters into a preset AI knowledge base; if the cutting force is abnormal, executing a cutting force optimization strategy for the basic toolpath based on AI: reducing the toolpath feed rate and decreasing the toolpath depth; if the cutting vibration parameters are abnormal, executing a vibration optimization strategy for the basic toolpath based on AI: reducing the toolpath feed rate and changing the toolpath axis tilt angle; if the cutting zone temperature is abnormal, executing a temperature optimization strategy for the basic toolpath based on AI: obtaining a reverse compensation amount based on the thermal offset and adjusting the toolpath according to the reverse compensation amount; if the cutting spindle load is abnormal, executing a load optimization strategy for the basic toolpath based on AI: reducing the toolpath feed rate and decreasing the toolpath depth.

[0007] In conjunction with the first aspect, in one possible implementation, before switching to the corresponding basic toolpath before the cutting tool performs cutting on the workpiece, the method further includes: preloading the next basic toolpath corresponding to the workpiece when the cutting tool performs any of the basic toolpaths.

[0008] In conjunction with the first aspect, in one possible implementation, before switching to the corresponding basic toolpath before the cutting tool performs cutting on the workpiece, the method further includes: after the cutting tool has completed any of the basic toolpaths, performing a tool lifting action and continuing for a redundant duration.

[0009] In conjunction with the first aspect, in one possible implementation, if the cutting force is abnormal, a cutting force optimization strategy for the basic toolpath is executed based on AI: reducing the toolpath feed rate and decreasing the toolpath depth, including: if the fluctuation value of the cutting force is greater than a preset fluctuation threshold, marking the current position of the cutting tool as a key fluctuation point; extracting the first deceleration parameter, the first tool lift amplitude, and the first tool lift length generated by the preset AI knowledge base; the AI, starting from the key fluctuation point, reducing the toolpath feed rate based on the first deceleration parameter, and generating a fluctuation-optimized toolpath with a length equal to the first tool lift length; the AI ​​controlling the cutting tool to execute the fluctuation-optimized toolpath starting from the key fluctuation point.

[0010] In conjunction with the first aspect, one possible implementation further includes: if the fluctuation value of the cutting tool in the fluctuation-optimized toolpath is greater than a preset fluctuation threshold, then controlling the cutting tool to detach from the workpiece; controlling the cutting tool to move to the end position of the fluctuation-optimized toolpath, and then switching to the basic toolpath to perform cutting; and marking the current fluctuation-optimized toolpath.

[0011] In conjunction with the first aspect, in one possible implementation, if the cutting vibration parameters are abnormal, a vibration optimization strategy for the basic toolpath is executed based on AI: reducing the toolpath feed rate and changing the toolpath axis tilt angle, including: if the cutting vibration parameters meet the predicted resonant frequency, marking the current position of the cutting tool as a critical vibration abnormality point; obtaining the tool axis change parameters generated by the preset AI knowledge base; and adjusting the cutting tool axis tilt angle based on the critical vibration abnormality point and the tool axis change parameters until the current basic toolpath is completed.

[0012] In conjunction with the first aspect, one possible implementation further includes: if the cutting vibration parameters again conform to the predicted resonant frequency after the critical vibration location point, then mark the current position as the secondary adjustment critical location point; obtain the secondary tool axis change parameters generated by the preset AI knowledge base; and adjust the tool axis tilt angle of the cutting tool again based on the secondary tool axis change parameters, starting from the secondary adjustment critical location point.

[0013] In conjunction with the first aspect, in one possible implementation, if the cutting zone temperature is abnormal, a temperature optimization strategy for the basic toolpath is executed based on AI: a reverse compensation amount is obtained based on the thermal offset, and the toolpath is adjusted according to the reverse compensation amount, including: if the thermal offset corresponding to the cutting zone temperature is greater than a preset offset, the current position of the cutting tool is marked as a thermal offset key position point; the reverse compensation offset generated by the preset AI knowledge base is extracted; the reverse compensation offset is superimposed on the basic toolpath starting from the thermal offset key position point to obtain a thermal correction toolpath; and the cutting tool is controlled to execute the thermal correction toolpath starting from the thermal offset key position.

[0014] In conjunction with the first aspect, in one possible implementation, if the cutting spindle load is abnormal, an AI-based load optimization strategy for the basic toolpath is executed: reducing the toolpath feed rate and decreasing the toolpath depth for a fourth unit adjustment duration, including: if the cutting spindle load is greater than a preset load, marking the current position of the cutting tool as a critical load abnormality point; extracting the second deceleration parameter, the second tool lift amplitude, and the second tool lift length generated by the preset AI knowledge base; starting from the critical load abnormality point, reducing the toolpath feed rate based on the second deceleration parameter, and generating a load-optimized toolpath with a length equal to the second tool lift length; wherein, the distance between the load-optimized toolpath and the workpiece along the normal direction of the machining surface is the third distance, the distance between the basic toolpath and the workpiece along the normal direction of the machining surface is the fourth distance, and the third distance minus the fourth distance is the second tool lift amplitude; controlling the cutting tool to execute the load-optimized toolpath starting from the critical load abnormality point.

[0015] In conjunction with the first aspect, one possible implementation further includes: if the load on the cutting spindle when the cutting tool is in the load-optimized toolpath is greater than the preset load, then the preset AI knowledge base generates a layered cutting strategy; in the second tool lift-off length, the preset AI knowledge base divides the current cutting depth into a preset number of layers to obtain multiple single-layer cutting toolpaths; the cutting tool is controlled to repeatedly execute cutting with the single-layer cutting toolpaths until the toolpaths within the second tool lift-off length range are cut to the current cutting depth.

[0016] This application achieves a significant improvement in machining quality. During machining, it can detect changes in operating conditions such as cutting force fluctuations, machining vibrations, thermal deformation, and spindle load in real time, and automatically execute targeted toolpath correction strategies, effectively suppressing abnormal states during the cutting process. This ensures the surface contour accuracy and surface quality of high-precision components such as harmonic flexible wheels, RV cycloidal wheels, precision gears, and robot joint curved surfaces, significantly reduces deformation in thin-walled areas, and eliminates machining defects such as vibration marks and tool marks, thereby effectively improving the mass production yield. Simultaneously, through intelligent active avoidance and real-time compensation mechanisms, this method reduces unplanned downtime during machining while protecting the tool and machine spindle from accidental damage, achieving a smoother and more continuous cutting process, improving overall machining efficiency, and effectively extending tool life. This application achieves high-precision machining while simultaneously ensuring high efficiency and high mass production stability. Attached Figure Description

[0017] Figure 1 The diagram shows the steps of a method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts, according to an embodiment of this application.

[0018] Figure 2 The diagram shows the steps for calculating the redundant time of the lifting blade.

[0019] Figure 3 The diagram shows a schematic of the cutting force optimization strategy.

[0020] Figure 4 The diagram shows a strategy for when the cutting force fluctuates again.

[0021] Figure 5 The diagram shows a vibration optimization strategy.

[0022] Figure 6 The diagram shown illustrates the secondary tool axis modification.

[0023] Figure 7 The diagram shows a temperature optimization strategy.

[0024] Figure 8The diagram shows a load optimization strategy.

[0025] Figure 9 The diagram shown illustrates the layered cutting process.

[0026] Figure 10 The diagram shows the basic toolpath of a cutting tool.

[0027] Figure 11 The diagram shown illustrates the change in the cutter shaft tilt angle. Detailed Implementation

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0029] Figure 1 The diagram illustrates the steps of a method for global toolpath optimization in multi-parameter coupled AI machining of robot parts, according to an embodiment of this application. This application provides a method for global toolpath optimization in multi-parameter coupled AI machining of robot parts. In one embodiment, as shown... Figure 1 As shown, the method includes: Step 101: Before the cutting tool performs cutting on the workpiece, switch to the corresponding basic toolpath.

[0030] The basic toolpath in this step is pre-made based on the target cutting pattern of the workpiece. The cutting tool travels along the basic toolpath to machine the cutting pattern. The toolpath depth at each position in the basic toolpath is the set cutting depth at each cutting position.

[0031] Step 102: Real-time acquisition of multiple cutting parameters of the cutting tool when performing cutting with the basic toolpath.

[0032] Step 103: Import multiple cutting parameters into the preset AI knowledge base.

[0033] In this step, the preset AI knowledge base receives multiple cutting parameters collected in real time from step 102, including parameters such as cutting force, vibration, temperature, and spindle load. The AI ​​performs preprocessing such as filtering and noise reduction on the raw sensor data and extracts key features, including: cutting force fluctuation value, cutting vibration parameters, cutting zone temperature, cutting spindle load, and vibration frequency. These parameters are then compared with a preset safety baseline, which defines the normal range of each parameter. This range is derived from historical data and expert experience in the knowledge base. If all parameters are within the normal range, it is determined to be a normal operating condition, and the basic toolpath continues to be executed. If any parameter exceeds the safety baseline, it is determined to be an abnormal cutting condition, and the corresponding strategy branch needs to be entered.

[0034] Step 104: If the cutting force is abnormal, then execute a cutting force optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and reduce the toolpath depth.

[0035] Step 105: If the cutting vibration parameters are abnormal, then execute a vibration optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and change the toolpath axis tilt angle and / or the toolpath milling mode.

[0036] Step 106: If the temperature in the cutting zone is abnormal, then execute a temperature optimization strategy for the basic toolpath based on AI: obtain the reverse compensation amount based on the thermal offset, and adjust the toolpath according to the reverse compensation amount.

[0037] Step 107: If the cutting spindle load is abnormal, execute a load optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and decrease the toolpath depth.

[0038] In this embodiment, a basic toolpath is pre-defined. During the online phase of the machining process, the system collects multiple parameters in real time, such as cutting force, vibration, temperature, and spindle load, using sensors, and compares them with a preset safety baseline. Once an abnormality is detected in a parameter, the AI ​​knowledge base automatically matches and executes the corresponding optimization strategy from the built-in strategy rule base, resulting in various execution schemes in steps 104 to 107. For example, when the cutting force is abnormal, the feed rate and toolpath depth are reduced; when the vibration is abnormal, the tool axis tilt angle is changed; when the temperature is abnormal, the toolpath coordinates are adjusted through reverse compensation; and when the load is abnormal, deceleration and depth of cut are performed, thereby achieving real-time, automated local correction of the basic toolpath.

[0039] This application utilizes AI for continuous fine-tuning of the toolpath, effectively suppressing cutting force fluctuations and machining vibrations. This improves the surface contour accuracy of robot parts, reduces surface roughness, and significantly eliminates defects such as tool marks and vibration patterns. Specifically targeting thin-walled and heat-deformable areas of precision components such as harmonic flexures and RV cycloidal wheels, this method effectively reduces machining deformation through active avoidance and real-time thermal compensation, thereby increasing the mass production yield of key robot components. While improving machining quality, this method also enhances overall machining efficiency and extends tool life by reducing overload cutting and tool wear, achieving high precision, high efficiency, and high mass production stability.

[0040] Figure 2 The diagram illustrates the steps involved in maximizing the lifting tool redundancy time. In one embodiment, as shown... Figure 2 As shown, prior to step 101, the multi-parameter coupled AI machining toolpath global optimization method for robot parts further includes: Step 201: When the cutting tool executes any basic toolpath, preload the basic toolpath corresponding to the next workpiece to be machined.

[0041] Step 202: After the cutting tool has completed any basic toolpath, perform a tool lifting action and continue for a redundant duration.

[0042] In this embodiment, before the toolpath for machining a workpiece is finished, the system preloads the basic toolpath corresponding to the next workpiece in the background, which can improve machining efficiency and production cycle time. After the current toolpath is completed, the toolpath automatically switches to the preset basic toolpath, and the machine tool performs a normal tool lifting action to avoid collision of the cutting tool when changing parts, and waits for cutting the next workpiece.

[0043] Figure 3 The diagram shows a schematic of a cutting force optimization strategy. In one embodiment, as shown... Figure 3 As shown, step 104 includes: Step 301: If the fluctuation value of the cutting force is greater than the preset fluctuation threshold, then mark the current position of the cutting tool as the critical position point of the fluctuation abnormality.

[0044] In this step, based on the expert experience in the knowledge base, the preset fluctuation threshold is set to 30%.

[0045] Step 302: Extract the first deceleration parameters, first lifting amplitude, and first lifting length generated by the preset AI knowledge base.

[0046] The working condition in step 301 corresponds to the strategy in step 302. The AI ​​knowledge base automatically calls the corresponding first deceleration parameters, first lifting amplitude, and first lifting length.

[0047] Step 303: Starting from the critical location of the fluctuation anomaly, the AI ​​reduces the toolpath feed rate based on the first deceleration parameter and generates a fluctuation-optimized toolpath with a length equal to the first tool lifting length.

[0048] Step 304: The AI ​​controls the cutting tool to execute the fluctuation optimization toolpath starting from the critical position point of the fluctuation anomaly.

[0049] In this step, after the fluctuation optimization toolpath ends, the basic toolpath continues to be executed.

[0050] In this embodiment, when the system detects that the cutting force fluctuation value exceeds the preset fluctuation threshold, half of the preset fluctuation threshold is set to a fluctuation amplitude of 30%. This location is marked as a critical location point of abnormal fluctuation. The AI ​​knowledge base uses the first deceleration parameter, the first tool lift amplitude, and the first tool lift length as the fluctuation-optimized toolpath. The key to this new toolpath is that, while keeping the feed direction unchanged, the tool is lifted a certain distance along the normal direction of the machining surface, i.e., the first tool lift amplitude. This reduces the depth of cut, allowing the tool to pass through the high-risk area with a shallower cutting method. Based on the expert experience in the AI ​​knowledge base, the first tool lift amplitude is set to 30%, meaning the tool lift amplitude is 30% of the currently set depth of cut. Simultaneously, this fluctuation-optimized toolpath also includes a feed rate reduction instruction, i.e., reducing the feed rate by the first deceleration parameter to further reduce the instantaneous cutting load. Based on the expert experience in the knowledge base, the first deceleration parameter is set to 30%, i.e., a 30% reduction in speed. The cutting tool starts from the critical point of abnormal fluctuation and travels along a new path of elevation and deceleration for a toolpath of the first lift-off length. After passing the danger zone, it smoothly cuts back to the basic toolpath. Based on expert experience in the knowledge base, the first lift-off length is set to a value between 10% and 30% of the current basic toolpath length. When generating the fluctuation-optimized toolpath, a toolpath trajectory identical to the basic toolpath trajectory is automatically drawn. At the same time, the cutting depth of the toolpath is adjusted based on the first lift-off amplitude, and the toolpath trajectory length is determined based on the first lift-off length.

[0051] This embodiment effectively avoids the risks of tool breakage or workpiece overload that may result from drastic fluctuations in cutting force by actively reducing the depth of cut. Since the tool does not completely detach from the workpiece, the machining process remains continuous, avoiding cycle time losses caused by sudden stops and starts. Simultaneously, the temporarily reduced depth of cut and feed rate allow for less mechanical impact through high-stress zones, thus maintaining high overall machining efficiency while protecting both the tool and the workpiece. After cutting the current workpiece, the toolpath can be optimized to a set depth of cut for further machining to reach the desired depth. This process can be manually controlled or precisely controlled by AI.

[0052] Figure 4 The diagram illustrates a strategy when the cutting force fluctuates again. In one embodiment, as shown... Figure 4 As shown, the multi-parameter coupled AI machining toolpath global optimization method for robot parts also includes: Step 401: If the fluctuation value of the cutting tool in the fluctuating optimized toolpath is greater than the preset fluctuation threshold, then control the cutting tool to detach from the workpiece.

[0053] Step 402: Control the cutting tool to move to the end position of the fluctuating optimized toolpath, and then switch to the basic toolpath to perform cutting.

[0054] Step 403: Mark the current fluctuation and optimize the toolpath.

[0055] In this embodiment, a fluctuation-optimized toolpath has already been executed, but the monitored cutting force fluctuation value still exceeds the preset safety threshold. At this point, the system determines that the cutting state in the current area has exceeded the normal controllable range. This could be due to abnormal tool wear or chipping, or abnormal hardness or internal defects in the workpiece in that area. Therefore, further fine-tuning of the toolpath is no longer attempted. The system's strategy is to forcibly interrupt cutting in this area, controlling the cutting tool to immediately detach from the workpiece surface to completely eliminate the risk of damage to the tool and workpiece caused by abnormal cutting. Afterward, the system controls the tool to move to the end of the interrupted fluctuation-optimized toolpath and switches back to the normal basic toolpath from there to continue subsequent machining. Simultaneously, the system marks the segment of the fluctuation-optimized toolpath that was interrupted due to the anomaly and stores it as key process data in a preset AI knowledge base.

[0056] Figure 5 The diagram shows a vibration optimization strategy. In one embodiment, as shown... Figure 5 As shown, step 105 includes: Step 501: If the cutting vibration parameters match the predicted resonant frequency, mark the current position of the cutting tool as the critical location point of vibration abnormality.

[0057] Step 502: Obtain the tool axis change parameters generated by the preset AI knowledge base.

[0058] Step 503: Starting from the critical location of abnormal vibration, AI adjusts the cutting tool axis tilt angle based on the tool axis change parameters until the current basic toolpath is completed.

[0059] In this embodiment, a cutting test is performed in advance to obtain multiple natural frequencies of the tool system during cutting, i.e., the predicted resonant frequency. When the system detects that the main frequency of the current vibration matches the pre-determined resonant frequency, the location is marked as a critical vibration abnormality point. When the abnormal vibration condition in step 501 is triggered, a pre-set tool axis change strategy is invoked, and step 502 invokes the tool axis change parameters corresponding to this strategy. When adjusting the tool axis tilt angle, the tool axis change parameters correspond to a preset angle step. Based on expert experience in the knowledge base, the preset angle step is set to 1°, i.e., the tool axis tilt angle is changed by 1°, which can change the contact geometry between the tool and the workpiece. This can break the stable vibration mode of the current process system and fundamentally avoid resonance. The AI ​​controls the cutting tool to start from the critical vibration abnormality point and execute along the basic toolpath with a new tool axis tilt angle until the end of the current basic toolpath. This can effectively and quickly eliminate resonance phenomena in the machining process and avoid vibration marks, tool chipping, or spindle damage caused by continuous resonance. Step 502 generates a fine-tuning angle command, rather than a large-scale angle change, ensuring a smooth transition when changing working conditions and avoiding secondary impacts or cutting discontinuities caused by drastic adjustments. This embodiment effectively suppresses vibration while maintaining the smoothness of the machining process, protecting the tool and workpiece, and ensuring the machining quality of the part surface.

[0060] Figure 6 The diagram shows a secondary tool axis modification. In one embodiment, as shown... Figure 6 As shown, the multi-parameter coupled AI machining toolpath global optimization method for robot parts also includes: Step 601: If the cutting vibration parameters meet the predicted resonant frequency again after the cutting tool reaches the critical position point of abnormal vibration, then mark the current position as the critical position point for secondary adjustment.

[0061] Step 602: Obtain the secondary tool axis change parameters generated by the preset AI knowledge base.

[0062] Step 603: Starting from the key position point of the secondary adjustment, adjust the cutting tool axis tilt angle again based on the secondary tool axis change parameters.

[0063] In this embodiment, after the toolpath undergoes its first deflection in step 504 and travels a certain distance, the cutting vibration parameters monitored by the system again match the predicted resonant frequency, indicating that the previous single intervention of changing the tool axis tilt angle is insufficient to completely escape the resonance range. At this point, the system marks this new position as a key point for secondary adjustment to trigger a higher-level optimization strategy. Upon response, new secondary tool axis change parameters are generated. These parameters are also based on a preset angle step size, performing a small-angle step-by-step fine-tuning on the basis of the first tool axis tilt angle adjustment, for example, tilting another 1° on top of an already 1° tilt. Starting from this key point for secondary adjustment, the system fine-tunes the tool axis tilt angle again based on the new tool axis change parameters and continues cutting.

[0064] This embodiment provides a progressive and iterative vibration damping solution for continuous or reciprocating resonance by setting up a secondary adjustment mechanism. When the first deflection fails to completely eliminate the resonance, the system does not choose to make a large leap or forcibly interrupt it. Instead, based on the same step-by-step principle, it performs small-angle fine-tuning of the working condition again, thereby finding the optimal working point that completely avoids resonance in a gradual approximation manner. This step-by-step adjustment strategy maintains the continuity and stability of the machining process and avoids the risk of sudden changes in cutting state that may occur due to a one-time large change in toolpath parameters.

[0065] Figure 10 The diagram shows the basic toolpath of a cutting tool. Figure 11 The diagram shows the change in the cutter axis tilt angle. (Refer to...) Figure 10 The dashed line represents the basic toolpath 1001. In the example shown, the basic toolpath 1001 is formed by three reciprocating strokes. The cutting tool 1002 then processes the cutting lines on the workpiece 1003 along the basic toolpath 1001. (Refer to...) Figure 11 The diagram shows the change of the tool axis tilt angle of the cutting tool 1002 twice when resonance is triggered. The change of the tool axis tilt angle effectively suppresses or eliminates the resonance phenomenon during the cutting process.

[0066] Figure 7 The diagram illustrates a temperature optimization strategy. In one embodiment, as shown... Figure 7 As shown, step 106 includes: Step 701: If the thermal offset corresponding to the temperature of the cutting zone is greater than the preset offset, then mark the current position of the cutting tool as the thermal offset key position point.

[0067] Step 702: Extract the reverse compensation offset generated by the preset AI knowledge base.

[0068] Step 703: Starting from the key position of thermal offset, superimpose the reverse compensation offset on the basic toolpath to obtain the thermal correction toolpath.

[0069] Step 704: Control the cutting tool to execute the thermal correction toolpath starting from the thermal offset critical position.

[0070] In this embodiment, the preset offset is the system's allowable error range, which can be derived from long-term machining data. Exceeding this preset offset will cause the cutting texture error of the workpiece to fail to meet standards and the scrap rate to fail to meet requirements. By monitoring the temperature of the cutting zone, when the offset exceeds the preset offset, the system marks this location as a critical thermal offset point. After responding, the system extracts a specific reverse compensation offset from historical data or an online lookup table. This value is opposite in direction and equal in magnitude to the thermal offset. Subsequently, the system uses this critical thermal offset point as the starting point and superimposes this reverse compensation offset on the original basic toolpath coordinates to generate a new thermally corrected toolpath. The cutting tool starts from this abnormal point and continues cutting along this new toolpath with the superimposed compensation, thereby actively offsetting the dimensional deviation caused by thermal deformation.

[0071] This embodiment, through the aforementioned mechanism, can proactively compensate for toolpath position deviations caused by thermal deformation of the machine tool or workpiece in real time, thereby effectively eliminating contour deviation problems caused by heat accumulation in the later stages of continuous machining. This application uses a lookup table to quickly calculate the compensation amount, rather than complex online finite element solutions, thus achieving millisecond-level rapid response and ensuring the real-time nature of the compensation action. Simultaneously, this embodiment avoids global replanning of the entire toolpath by employing a strategy of only locally superimposing compensation after exceeding limits, maintaining the efficiency and continuity of machining. Ultimately, this solution can significantly improve the dimensional accuracy stability during long-term machining, ensuring that precision parts such as RV cycloidal wheels maintain consistent machining quality even after continuously machining multiple surfaces.

[0072] Figure 8 The diagram shows a load optimization strategy. In one embodiment, as shown... Figure 8 As shown, step 107 includes: Step 801: If the load on the cutting spindle is greater than the preset load, mark the current position of the cutting tool as a critical position point of abnormal load.

[0073] Step 802: Extract the second deceleration parameters, second lifting amplitude, and second lifting length generated by the preset AI knowledge base.

[0074] Step 803: Starting from the critical load abnormality point, reduce the toolpath feed rate based on the second deceleration parameter, and generate a load-optimized toolpath with a length equal to the second tool lifting length.

[0075] Step 804: Control the cutting tool to execute the load-optimized toolpath starting from the critical position point of abnormal load.

[0076] In this embodiment, when the spindle load exceeds a preset load, the location is marked as a critical load anomaly point. The preset load is determined by the machine tool's factory-set basic parameters. The preset AI knowledge base responds and invokes pre-defined optimization strategies: a second deceleration parameter, a second tool lift amplitude, and a second tool lift length. Based on expert experience in the AI ​​knowledge base, the second tool lift amplitude is set to 30%, meaning the tool lift amplitude is 30% of the currently set cutting depth. The second deceleration parameter is also set to 30%, meaning a 30% speed reduction. The second tool lift length is set to 30% of the basic toolpath length based on expert experience. Subsequently, the AI ​​generates a load-optimized toolpath starting from this anomaly point, identical to the basic toolpath trajectory but with adjusted cutting depth and feed rate. Specifically, this optimized toolpath, while maintaining the feed direction, lifts the tool a distance (the second tool lift amplitude) along the normal to the machined surface, effectively reducing the depth of cut per pass and allowing the tool to pass through high-load areas with a shallower depth of cut. Simultaneously, the AI ​​reduces the feed rate of this section of the toolpath based on the second deceleration parameter, further reducing the instantaneous load on the spindle. Starting from the critical point of abnormal load, the cutting tool travels a preset second lifting length along this new lifting and deceleration path. After crossing the high-load area, it smoothly cuts back to the basic toolpath to continue execution.

[0077] This embodiment effectively avoids the risk of motor overload alarms or spindle damage that may result from abnormally high spindle load by actively reducing the depth of cut and lowering the feed rate. Since the tool does not completely detach from the workpiece, the machining process remains continuous, avoiding cycle time losses caused by sudden stops and starts. This maximizes overall machining efficiency while protecting the core safety of the spindle and tool. After the entire workpiece is cut, the system automatically performs a final finishing cut on the undercut areas where the overload-optimized toolpath was previously applied, according to the original set depth of cut. This ensures that the final dimensional accuracy and machining quality of the part fully meet the drawing requirements.

[0078] Figure 9 The diagram shows a layered cutting process. In one embodiment, as shown... Figure 9 As shown, the multi-parameter coupled AI machining toolpath global optimization method for robot parts also includes: Step 901: If the spindle load of the cutting tool under load-optimized toolpath is greater than the preset load, the preset AI knowledge base generates a layered cutting strategy.

[0079] Step 902: In the second lifting length, the preset AI knowledge base divides the current cutting depth into preset layers to obtain multiple single-layer cutting toolpaths.

[0080] Step 903: Control the cutting tool to repeatedly execute cutting with a single-layer cutting toolpath until the toolpath within the second lifting length range is cut to the current cutting depth.

[0081] In this embodiment, even after load-optimized toolpaths have been executed, the spindle load monitored by the system still exceeds the preset safety threshold, indicating that a single intervention to reduce the depth of cut is insufficient to control the load within a safe range. At this point, the system determines that the cutting state in the current area is at a high-risk level and no longer attempts to resolve the issue by fine-tuning the toolpath parameters. The system executes a layered cutting strategy, dividing the current depth of cut into a preset number of layers (e.g., 3 layers) within the entire toolpath range corresponding to the second lift-off length, thereby generating multiple independent single-layer cutting toolpaths with smaller depths of cut in each layer. The AI ​​then controls the cutting tool to repeatedly execute cutting within this range using these single-layer toolpaths, removing only a very small amount of material in each layer, until the toolpath within the entire second lift-off length range is cut to the set depth of cut for the current workpiece, thereby reducing the spindle load of a single cut to a safe level. This embodiment improves the safety and continuity of the machining process under extremely high load conditions. It ensures that even in extreme situations such as abnormal material hardness or tool degradation, the system can complete cutting stably, avoiding serious accidents such as tool breakage, spindle overload shutdown, or even workpiece scrap. This significantly enhances the robustness and adaptability of the entire machining system. The cutting depth after the second tool lift in the current load-optimized toolpath is then used to cut along the cutting texture again until the set cutting depth is reached, after executing the load-optimized toolpath and the layered cutting strategy.

[0082] This application also provides a multi-parameter coupled AI machining toolpath global optimization system for robot parts, including a setting module, a data acquisition module, and a strategy execution module. The setting module is configured to switch to the corresponding basic toolpath before the cutting tool performs cutting on the workpiece. The data acquisition module is communicatively connected to the setting module and is configured to: acquire multiple cutting parameters in real time when the cutting tool performs cutting with the basic toolpath; and import these multiple cutting parameters into a preset AI knowledge base. The strategy execution module is communicatively connected to the acquisition module. The strategy execution module is configured to: if the cutting force is abnormal, execute a cutting force optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and decrease the toolpath depth; if the cutting vibration parameters are abnormal, execute a vibration optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and change the toolpath axis tilt angle and / or toolpath milling mode; if the cutting zone temperature is abnormal, execute a temperature optimization strategy for the basic toolpath based on AI: obtain a reverse compensation amount based on the thermal offset and adjust the toolpath according to the reverse compensation amount; if the spindle load is abnormal, execute a load optimization strategy for the basic toolpath based on AI: reduce the toolpath feed rate and decrease the toolpath depth.

[0083] In its application, this application effectively suppresses cutting force fluctuations and machining vibrations by continuously fine-tuning the toolpath using AI, improving the surface contour accuracy of robot parts by 30% to 50%, reducing surface roughness by more than 40%, and significantly eliminating defects such as tool marks and vibration patterns. Specifically targeting thin-walled and heat-deformable areas of precision parts such as harmonic flexures and RV cycloidal wheels, this method effectively reduces machining deformation, improves the contour accuracy of key components, and increases mass production yield through active avoidance and real-time thermal compensation. Regarding machining efficiency and cost, while achieving a leap in machining quality, this method also effectively improves overall machining efficiency and extends tool life by reducing overload cutting and tool wear, truly achieving a balance between high precision, high efficiency, and high mass production stability.

[0084] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0085] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0086] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0087] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features of the invention herein.

[0088] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts, characterized in that, include: Before the cutting tool performs cutting on the workpiece, switch to the corresponding basic toolpath; Real-time acquisition of multiple cutting parameters of the cutting tool when performing cutting with the basic toolpath; Import the aforementioned multiple cutting parameters into a preset AI knowledge base; If the cutting force is abnormal, an AI-based cutting force optimization strategy is executed on the basic toolpath: reduce the toolpath feed rate and decrease the toolpath depth. If the cutting vibration parameters are abnormal, a vibration optimization strategy for the basic toolpath is executed based on AI: reduce the toolpath feed rate and change the toolpath axis tilt angle. If the temperature in the cutting zone is abnormal, a temperature optimization strategy for the basic toolpath is executed based on AI: the reverse compensation amount is obtained based on the thermal offset, and the toolpath is adjusted according to the reverse compensation amount. If the cutting spindle load is abnormal, an AI-based load optimization strategy for the basic toolpath is executed: reducing the toolpath feed rate and decreasing the toolpath depth.

2. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, Before switching to the corresponding basic toolpath before the cutting tool performs cutting on the workpiece, the method further includes: When the cutting tool executes any of the basic toolpaths, the next basic toolpath corresponding to the workpiece is preloaded.

3. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, Before switching to the corresponding basic toolpath before the cutting tool performs cutting on the workpiece, the method further includes: After the cutting tool completes any of the basic toolpaths, a tool lifting action is performed and continues for a redundant duration.

4. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, If the cutting force is abnormal, an AI-based cutting force optimization strategy is executed on the basic toolpath: reducing the toolpath feed rate and decreasing the toolpath depth, including: If the fluctuation value of the cutting force is greater than the preset fluctuation threshold, the current position of the cutting tool is marked as a critical position point of abnormal fluctuation. Extract the first deceleration parameters, first blade lifting amplitude, and first blade lifting length generated by the preset AI knowledge base; Starting from the key location of the fluctuation anomaly, the AI ​​reduces the toolpath feed rate based on the first deceleration parameter and generates a fluctuation-optimized toolpath with a length equal to the first tool lifting length. AI controls the cutting tool to execute the fluctuation-optimized toolpath starting from the critical location point of the fluctuation anomaly.

5. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 4, characterized in that, Also includes: If the fluctuation value of the cutting tool in the fluctuation-optimized toolpath is greater than a preset fluctuation threshold, then the cutting tool is controlled to detach from the workpiece. Control the cutting tool to move to the end position of the fluctuation-optimized toolpath, and then switch to the basic toolpath to perform cutting; Mark the currently described fluctuation-optimized toolpath.

6. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, If the cutting vibration parameters are abnormal, a vibration optimization strategy for the basic toolpath is executed based on AI: reducing the toolpath feed rate and changing the toolpath axis tilt angle, including: If the cutting vibration parameters match the predicted resonant frequency, then the current position of the cutting tool is marked as a critical location point of vibration abnormality. Obtain the tool axis change parameters generated by the preset AI knowledge base; Starting from the critical location of the vibration anomaly, the AI ​​adjusts the cutting tool axis tilt angle based on the tool axis change parameters until the current basic toolpath is completed.

7. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 6, characterized in that, Also includes: If, after the critical point of abnormal vibration, the cutting vibration parameters again meet the predicted resonant frequency, then the current position is marked as the critical point of secondary adjustment. Obtain the secondary tool axis change parameters generated by the preset AI knowledge base; Starting from the key position point of the secondary adjustment, the tool axis tilt angle of the cutting tool is adjusted again based on the secondary tool axis change parameters.

8. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, If the cutting zone temperature is abnormal, a temperature optimization strategy for the basic toolpath is executed based on AI: a reverse compensation amount is obtained based on the thermal offset, and the toolpath is adjusted according to the reverse compensation amount, including: If the thermal offset corresponding to the temperature of the cutting zone is greater than the preset offset, then the current position of the cutting tool is marked as a key thermal offset position point. Extract the reverse compensation offset generated by the preset AI knowledge base; Starting from the aforementioned key thermal offset location, the reverse compensation offset is superimposed on the basic toolpath to obtain the thermally corrected toolpath; The cutting tool is controlled to execute the thermal correction toolpath starting from the thermal offset critical position.

9. The method for global optimization of toolpaths in multi-parameter coupled AI machining of robot parts according to claim 1, characterized in that, If the cutting spindle load is abnormal, an AI-based load optimization strategy for the basic toolpath is executed: reducing the toolpath feed rate and decreasing the toolpath depth, for a continuous adjustment period of four units, including: If the load on the cutting spindle is greater than the preset load, the current position of the cutting tool is marked as a critical position point of abnormal load. Extract the second deceleration parameters, second blade lifting amplitude, and second blade lifting length generated by the preset AI knowledge base; Starting from the critical load abnormality point, the toolpath feed rate is reduced based on the second deceleration parameter, and a load-optimized toolpath with a length equal to the second tool lifting length is generated; wherein, the distance between the load-optimized toolpath and the workpiece along the normal of the machining surface is the third distance, the distance between the basic toolpath and the workpiece along the normal of the machining surface is the fourth distance, and the third distance minus the fourth distance is the second tool lifting amplitude; The cutting tool is controlled to execute the load-optimized toolpath starting from the critical load anomaly point.

10. The method for global optimization of multi-parameter coupled AI machining toolpaths for robot parts according to claim 9, characterized in that, Also includes: If the load on the cutting spindle is greater than the preset load when the cutting tool is in the load-optimized toolpath, then the preset AI knowledge base generates a layered cutting strategy; In the second lifting length, the preset AI knowledge base divides the current cutting depth into preset layers to obtain multiple single-layer cutting toolpaths; The cutting tool is controlled to repeatedly perform cutting with the single-layer cutting toolpath until the toolpath within the second lifting length range is cut to the current cutting depth.

Images