System and method for self-recognition and trajectory planning and polishing of complex-shaped components in different spaces

Abstract

The application discloses a system and method for self-recognition and trajectory planning and polishing of a complex component in a special-shaped space, wherein the system mainly comprises the following steps: a workpiece is fixed through a fixed tooling workpiece; a visual scanning and feature extraction system is used for automatic scanning and feature extraction of all welds on the workpiece, and a polishing path and a polishing posture are planned; according to different process treatment requirements of the welds, a tool library system can automatically replace a tool bit; a task coordination planning algorithm mainly realizes the following points: a task is effectively distributed for online operation facing multiple welds, polishing amount calculation, polishing task homogenization distribution and anti-interference are realized, and a robot cooperates with operation according to the scanned weld position. The application determines the optimal polishing path through laser vision, configures a suitable polishing tool bit according to the weld shape size and the planned path, realizes quick and automatic replacement of the polishing tool through the tool library, and improves work efficiency.

Description

Technical Field

[0001] This invention relates to the field of grinding equipment technology, specifically to a grinding system and method for autonomous identification and trajectory planning of complex, irregularly shaped spatial components. Background Technology

[0002] In the grinding process of square and round molds, traditional methods relied mainly on manual operation. However, this traditional grinding method has many shortcomings. The work site is often filled with a large amount of dust particles, accompanied by high-decibel noise. Such environmental conditions can easily distract workers, making it difficult for them to maintain high-precision work standards for extended periods.

[0003] More importantly, due to the numerous uncontrollable factors inherent in manual operation, the quality of polishing is often inconsistent. The polishing effect can vary from product to product, undoubtedly posing a significant challenge to subsequent production stages. Furthermore, prolonged exposure to dusty environments poses a serious threat to workers' health, potentially leading to various respiratory illnesses.

[0004] Manual sanding requires a significant amount of time and manpower, resulting in low efficiency. Furthermore, inconsistencies in sanding quality may necessitate multiple rework processes, further reducing production efficiency.

[0005] Traditional grinding methods are ill-suited to the demands of high-precision, high-efficiency production. With the continuous development of grinding technology, the requirements for grinding precision and efficiency are becoming increasingly stringent, and traditional grinding methods can no longer meet these needs. Summary of the Invention

[0006] The purpose of this invention is to provide an autonomous identification and trajectory planning system and method for irregularly shaped, complex spatial components to improve the level and accuracy of grinding automation.

[0007] The technical solution to achieve the purpose of this invention is as follows:

[0008] A system for autonomous identification and trajectory planning of complex, irregularly shaped spatial components during grinding is characterized by comprising a visual scanning and feature extraction system, a robotic grinding platform, an intelligent algorithm unit, a grinding system, a dust removal system, and a workpiece fixing fixture. First, the workpiece is placed on the fixing fixture. Then, a laser scanning camera mounted on the end effector of one of the six-axis robots follows the base, moving along the workpiece on a ground track via a slider and helical gears. The laser camera collects information on the position, length, width, and thickness of all weld seams on the workpiece and transmits this information to an industrial control computer via a bus. Finally, a dual-robot task collaboration algorithm calculates the grinding amount based on the collected weld seam information and allocates multiple weld seam tasks based on an anti-interference mechanism. The six-axis robot drives the floating electric spindle at its end effector to complete the grinding operation. A dust cover and dust removal duct provide dust removal during the grinding operation.

[0009] Furthermore, the robotic grinding platform consists of a six-axis robot, cutting tools, a motor, a base and slider, a dust removal bracket, and helical gears. The six-axis robot is mounted on a ground rail via a mounting base and is used to scan the workpiece to be ground and slide axially at different weld positions. A laser scanning camera and a floating electric spindle are installed at its end. The motor is used to rotate the helical gears, and there is a rack on the ground rail that meshes with the helical gears, allowing the robotic grinding platform to slide along the slider on the ground rail. A tool magazine is installed next to the six-axis robot and contains grinding tools. When the tool is worn to a certain extent, it is automatically replaced. The dust removal bracket supports the dustproof and dust removal pipes.

[0010] The tool magazine is an automatic tool changer for adaptive grinding and polishing tools. The tool magazine has two sliding doors on one side. When the tool wears down to a preset threshold, or when different specifications of tool heads need to be replaced according to the specific requirements of the grinding process, the six-axis robot can automatically replace the tool heads placed in the tool magazine. Based on the visual scanning and feature extraction system, the robot scans the workpiece, determines the different types of welds and different process treatment methods, and automatically replaces different tool heads through the tool magazine to complete the process treatment.

[0011] Furthermore, the aforementioned visual scanning and feature extraction system is used to collect all geometric feature information of welds on the workpiece, providing data for grinding operations. It consists of a large-range laser scanning camera and an industrial control computer. The laser scanning camera is mounted on a laser mounting plate at the end of a six-axis robot. The six-axis robot can drive the scanning camera to complete the acquisition of all weld position, length, width, and thickness information, which is transmitted to the industrial control computer via a bus to reconstruct the three-dimensional contour of the weld.

[0012] Furthermore, the intelligent algorithm unit includes a dual-robot task collaborative planning algorithm. The system uses a laser scanning camera to scan and acquire 3D point cloud data of multiple weld seams, extracts geometric parameters, and calculates the grinding amount of each weld seam. According to the anti-interference mechanism, the six-axis robot on the left grinds the weld seams one by one from the initial position on the left end of the workpiece to the center of the workpiece, while the six-axis robot on the right slides from the right end to the center of the workpiece and grinds the weld seams one by one to the right. The motor encoders of the two robot grinding platforms collect the torque and speed of the motors in real time and convert them into pulse signals. These pulse signals are transmitted to the industrial control computer through the bus and converted into the actual position coordinates of the robot grinding platforms on the ground rail. Based on the position coordinates of the two robot grinding platforms, the distance between them is calculated in real time to ensure that the distance between the two machines is ≥3000mm at any time to avoid collisions. When allocating tasks, the total workload of each weld seam is calculated first, and then the weld seam tasks are equally allocated to the two six-axis robots according to the principle of equalization. Finally, the tasks are adjusted according to the real-time situation through dynamic allocation. After the grinding is completed, the system will perform a pre-detection and effect evaluation.

[0013] process:

[0014] 1. Grinding Amount Calculation Model

[0015] Based on the weld geometry (height a, width b, length l, curvature κ) and material removal rate η, the grinding amount for a single weld pass is defined as: Where s is the weld arc length parameter

[0016] 2. Anti-interference mechanism

[0017] Robot Layout:

[0018] The robot (left) is positioned at the initial position on the left end of the workpiece.

[0019] The robot (right) is located at the initial position on the right end of the workpiece.

[0020] Direction of movement:

[0021] The robot (left) grinds the weld seams one by one from left to right, moving in the direction of +X.

[0022] The robot (right) first slides from the right end to the center of the workpiece, and then grinds the weld seams one by one from the center to the right, with the direction of movement being -X.

[0023] Principle: The motor encoder collects the motor's torque and speed in real time and converts them into pulse signals. These pulse signals are transmitted to the control cabinet via a bus and converted into the actual position coordinates of the robotic grinding platforms on the base cabinet. Based on the position coordinates of the two robotic grinding platforms, the distance between them is calculated in real time. The control system compares the calculated actual distance d with a safety distance of 3000mm. If it is less than 3000mm, emergency braking is triggered. Ultimately, this ensures that the distance between the two machines is ≥3000mm at any given time.

[0024] 3. Task allocation methods

[0025] (1) Calculation of total workload: Calculate the total workload of each weld based on the characteristics of the weld.

[0026] The total workload Wi can be expressed as:

[0027] W i =k1·L i +k2·a i +k3·b i

[0028] Where li is the weld length, ai is the weld height, bi is the weld width, and k1, k2, and k3 are weighting coefficients that are adjusted according to the actual situation.

[0029] (2) Task pre-allocation

[0030] The principle of equalization allocation: Based on the total workload, the welding task is equally distributed between the two robots. The goal is to make the total workload of the two robots as balanced as possible.

[0031] Pre-allocation process:

[0032] 1. Sort all welds from largest to smallest according to the total amount of work Wi.

[0033] 2. Assign the welds to the robots with the smallest current total workload in sequence.

[0034] 3. Repeat the above steps until all weld seams are assigned.

[0035] The goal of pre-allocation is to make the total workloads WR1 and WR2 of the two robots as close as possible:

[0036] |W R1 -W R2 |≤∈

[0037] Where ∈ is a small tolerance value.

[0038] (3) Dynamic allocation

[0039] Real-time adjustment: During actual execution, inconsistencies in robot speed and deviations in task execution time may occur. The dynamic allocation algorithm adjusts the task based on real-time conditions.

[0040] Dynamic allocation strategy:

[0041] 1. Monitor the real-time workload of the two robots.

[0042] 2. If a robot completes a task quickly, it can take over part of the task from another robot.

[0043] 3. The goal of dynamic adjustment is to maintain a balanced workload between the two robots and avoid one robot being idle while the other is overloaded.

[0044] (4) Residual height detection and effect evaluation

[0045] Online inspection: During the grinding process, the weld reinforcement is detected in real time using a laser scanning camera.

[0046] Excess Height Assessment: The detected excess height is compared with a preset standard value to determine whether it meets the requirements. If the excess height exceeds the standard value, further polishing is required.

[0047] Grinding quality inspection: Use visual or tactile sensors to inspect the surface quality of the weld after grinding and assess whether it meets the preset smoothness and flatness standards.

[0048] Handling of non-conformities: If the grinding effect is not up to standard, the robot will grind again according to the inspection results until the weld quality meets the requirements.

[0049] Furthermore, the grinding system consists of a floating electric spindle and a cutting head. The grinding system is installed at the end of a six-axis robot, and the control system of the floating electric spindle is paired with the main control system via a bus to achieve data interaction and remote control. It has a built-in high-precision displacement sensor to monitor the gap change between the grinding tool and the workpiece surface in real time, transmits the collected gap data to the intelligent control system, calculates the error and generates control commands to optimize the spindle speed and feed rate, and ensures that the cutting head always maintains the best contact state with the workpiece surface.

[0050] The grinding operation begins. Following the path planned by the intelligent algorithm unit, the floating electric spindle moves with the six-axis robot to the part of the workpiece to be ground. When the grinding tool approaches the workpiece surface by about 1mm, the displacement sensor activates a high-precision detection mode, transmitting the gap data between the tool and the workpiece to the intelligent control system in real time. Based on the preset process requirements, the control system quickly calculates the radial and axial displacement adjustments required for the electric spindle and drives the floating module to adjust its movement, ensuring that the grinding tool precisely conforms to the workpiece surface. During the grinding process, the operator can monitor the operating status of the floating electric spindle in real time via a host computer, including parameters such as rotational speed, vibration amplitude, and displacement compensation.

[0051] Furthermore, the dust removal system consists of a dust removal duct, a dust cover, and a dust purification box; the grinding operation is carried out in the dust cover, which is supported by a dust removal bracket, and the dust generated during the operation is purified through the dust removal pipe, the dust removal duct, and the dust purification box (4).

[0052] Furthermore, the workpiece fixing fixture consists of a V-shaped support roller frame and a cylinder clamping block; the workpiece has straight ribs and semi-circular ribs on its sides; when the workpiece is placed, the semi-circular ribs are supported by the rollers in the V-shaped bracket, while the cylinder clamping block next to the bracket fixes the straight ribs located below the workpiece.

[0053] A method for autonomous identification and trajectory planning of complex irregular spatial components includes the following steps:

[0054] Step 1: Manually select the workpiece model, and hoist the workpiece to be ground onto the V-shaped support roller frame using a crane. Place the semi-circular ribs on the rollers.

[0055] Step 2: The cylinder clamping block on the fixed fixture clamps and fixes the straight rib plate below the workpiece.

[0056] Step 3: After the workpiece is placed and fixed, the robot grinding platform on the left moves along the workpiece direction on the ground rail by the helical gear driven by the motor. At the same time, the laser scanning camera at the end of the six-axis robot also starts to work, scanning the position, length, width and thickness of all welds on the workpiece, and transmitting the information to the industrial control computer.

[0057] Step 4: After the industrial control computer obtains the weld information, the dual-robot task collaboration algorithm first calculates the amount of weld grinding to be done, and then allocates the grinding tasks according to the anti-interference mechanism to ensure that all welds are reasonably allocated.

[0058] Step 5: According to the task assignment in Step 4, both robot grinding platforms start to move, and the six-axis robot drives the floating electric spindle at the end to perform grinding operations according to the preset path.

[0059] Step 6: During the grinding operation, the dust removal system also starts working simultaneously to ensure the safety and cleanliness of the work area;

[0060] Step 7: After the polishing is completed, the system will perform a pre-test and effect evaluation. If there are no problems, all equipment in the system will return to their initial positions, the six-axis robot will return to its initial position, and the display terminal will show that the polishing operation is complete and the parts are ready for manual removal.

[0061] The beneficial effects of this invention are: determining the optimal grinding path through laser vision; configuring appropriate grinding heads according to the shape and size of the weld and the planned path; and achieving rapid and automatic replacement of grinding tools through the tool magazine, thereby improving work efficiency. Attached Figure Description

[0062] Figure 1 This is a three-dimensional structural diagram of the autonomous identification and trajectory planning of irregularly shaped complex spatial components according to the present invention.

[0063] Figure 2 This is a schematic diagram of the planar structure for autonomous identification and trajectory planning of complex irregular spatial components according to the present invention.

[0064] Figure 3 This is a three-dimensional structural diagram of the robot grinding platform of the present invention.

[0065] Figure 4 This is a three-dimensional structural diagram of the workpiece fixing fixture of the present invention.

[0066] Figure 5 This is a three-dimensional structural diagram of the V-shaped roller frame of the present invention.

[0067] Figure 6 A general flowchart is provided for the polishing scheme of this invention.

[0068] Figure 7 This is a flowchart of the dual-robot task collaborative planning process of the present invention. Detailed Implementation

[0069] The present invention will now be described in detail with reference to the accompanying drawings.

[0070] The present invention and its embodiments are described below. This description is not restrictive, and actual embodiments are not limited thereto. In short, if those skilled in the art are inspired by this description and, without departing from the spirit of the invention, design similar structures and embodiments to this technical solution, all such designs should fall within the protection scope of the present invention. The present invention will now be described in further detail with reference to the accompanying drawings.

[0071] like Figure 1-5 As shown, the autonomous identification and trajectory planning system for complex irregular spatial components includes a visual scanning and feature extraction system, a robot grinding platform, an intelligent algorithm unit, a grinding system, a dust removal system, and workpiece fixing fixtures.

[0072] The aforementioned visual scanning and feature extraction system is used to collect geometric feature information of all welds on the workpiece, providing data for grinding operations;

[0073] The workpiece fixing fixture is used to support and fix the workpiece (18) during the grinding process by means of the V-shaped roller frame (20) and the cylinder clamping block (21);

[0074] In this embodiment, the workpiece fixing fixture consists of 8 V-shaped roller frames (20) and 4 cylinder clamping blocks (21). The semi-circular rib plate (19) of the workpiece is placed on the V-shaped roller frames (20), and the straight rib plate (22) is located below the workpiece (18).

[0075] The cylinder clamping block (21) clamps the cylinder;

[0076] In this embodiment, the robot grinding platform consists of a six-axis robot (10), a dust removal bracket (9), a tool magazine (5), a motor (14), a slider (12), and a helical gear (13);

[0077] In this embodiment, the six-axis robot (10) is mounted on the ground rail (7) via the mounting base (11) to complete the scanning of the workpiece (18) to be ground and the axial sliding of different weld positions; the tool magazine (5) is installed next to the six-axis robot (10), and there is also a dust removal bracket (14) on the platform to support the dust removal hood (3); the laser scanning camera (16) and the floating electric spindle (17) are installed at the end of the six-axis robot (10);

[0078] In this embodiment, the intelligent algorithm unit includes a dual-robot task collaborative planning algorithm. The system obtains three-dimensional point cloud data of multiple weld seams through laser vision scanning, extracts geometric parameters and calculates the grinding amount of each weld seam. According to the anti-interference mechanism, the six-axis robot (left) grinds the weld seams one by one from the initial position at the left end of the workpiece (18) to the center of the workpiece (18), and the six-axis robot (right) slides from the right end to the center of the workpiece (18) and grinds the weld seams one by one to the right. The system generates dual-machine paths based on B-spline curves to ensure that the distance between the two machines is ≥500mm at any time, and finally realizes efficient collaborative grinding of multiple weld seams.

[0079] In this embodiment, the dust removal system consists of a dust removal duct (6), a dust cover (3), and a dust purification box (4). The dust removal duct (6) is placed above the ground rail (7) and is used to transport the dust generated during grinding. The dust cover (3) covers the six-axis robot (10) to prevent the dust generated during grinding from damaging the surface of the workpiece (18). The dust purification box (4) is used to purify the dust transported by the dust removal duct (6).

[0080] In this embodiment, the visual scanning and feature extraction system consists of a laser line scanning camera (16) and an industrial control computer (1). The laser scanning camera (16) is mounted on the laser mounting plate at the end of the robot. The robot can drive the scanning camera (16) to complete the acquisition of information such as the position, length, width and thickness of all welds. The information is transmitted to the industrial control computer (1) via the bus, and the three-dimensional contour of the weld can be reconstructed.

[0081] In this embodiment, the grinding system consists of a floating electric spindle (17) and a cutting head (15). The grinding system is installed at the end of a six-axis robot (10). The control system of the floating electric spindle (17) is paired with the main control system through a bus to realize data interaction and remote control. A high-precision displacement sensor is built in to monitor the gap change between the grinding tool and the workpiece surface in real time and transmit the collected gap data to the intelligent control system.

[0082] Combination Figure 6 and 7 The basic execution flow of the method for autonomous identification and trajectory planning of complex components in irregular spaces is as follows:

[0083] The first step is to load the workpiece. The workpiece model is manually selected and the workpiece to be ground (18) is hoisted onto the V-shaped support roller frame (20) by a crane. The semi-circular rib plate (19) is placed on the roller to ensure the stability of the workpiece's center of gravity.

[0084] The second step is to fix the workpiece. The cylinder clamping block (21) on the fixing fixture clamps and fixes the straight rib plate (22) below the workpiece to ensure that the workpiece (18) does not move during the grinding process.

[0085] In the third step, after the workpiece (18) is placed and fixed, the robot grinding platform (left) drives the helical gear (13) through the servo motor (14) to move along the ground rail (7) at a constant speed. At the same time, the laser scanning camera (16) at the end of the robot is activated to perform high-precision scanning on the surface of the workpiece (18) to obtain the three-dimensional point cloud data of the weld, including the position, length, width, thickness and curvature of the weld, and transmit the data to the control unit in real time.

[0086] Fourth, after the robot control unit obtains the weld information, the dual-robot task collaboration algorithm first calculates the grinding amount of each weld based on the weld geometric features (such as height, width, and curvature), then, combined with the anti-interference mechanism, the grinding task is evenly distributed to the two robots according to the principle of equal distribution. Finally, the task is adjusted according to the real-time situation through dynamic allocation.

[0087] Fifth step, according to the task allocation result, the robot (left) grinds the weld seam from the initial position at the left end of the workpiece (18) to the center of the workpiece (18) one by one, and the robot (right) slides from the right end to the center of the workpiece (18) and grinds the weld seam to the right one by one; the encoder of the motor (14) will collect the torque and speed of the motor (14) in real time and convert it into pulse signals. These pulse signals are transmitted to the industrial control computer (1) through the bus and converted into the actual position coordinates of the robot grinding platform on the ground rail (7). Based on the position coordinates of the two robot grinding platforms, the distance between them is calculated in real time to ensure that the distance between the two machines is ≥3000mm at any time to avoid collision;

[0088] The sixth step involves starting the dust removal system during the grinding process. The dust is collected by a vacuum cleaner to ensure a clean working environment that meets safety standards.

[0089] Step 7: After the grinding is completed, the system will perform a pre-test and effect evaluation. If there are no problems, the system will automatically reset, the six-axis robot (10) will return to the initial position, the ground rail (7) will stop moving, and the display terminal will prompt that the operation is completed. The operator can view the grinding quality report through the human-machine interface and remove the workpiece (18) to carry out the next round of operation.

Claims

1. An autonomous identification and trajectory planning system for complex, irregularly shaped spatial components, characterized in that: The system includes a visual scanning and feature extraction system, a robot grinding platform, an intelligent algorithm unit, a grinding system, a dust removal system, and a workpiece fixing fixture. First, the workpiece (18) is placed on the fixing fixture. Then, the laser scanning camera (16) installed at the end of one of the six-axis robots (10) follows the base (11) and moves along the workpiece (18) on the ground rail (7) via the slider (12) and helical gear (13). The laser camera (16) begins to collect information on the position, length, width, and thickness of all welds on the workpiece (18) and transmits the information to the industrial control computer (1) via the bus. Finally, the dual-robot task collaboration algorithm calculates the grinding amount based on the collected weld information and allocates multiple weld tasks based on the anti-interference mechanism. The six-axis robot (10) drives the floating electric spindle (17) at the end to complete the grinding operation. The dust cover (3) and the dust removal duct (6) remove dust during the grinding operation. The aforementioned visual scanning and feature extraction system is used to collect all weld geometric feature information on the workpiece (18) and provide data for grinding operations. It consists of a large-range laser scanning camera (16) and an industrial control computer (1). The laser scanning camera (16) is mounted on the laser mounting plate at the end of the six-axis robot (10). The six-axis robot (10) can drive the scanning camera (16) to complete the acquisition of all weld position, length, width and thickness information, and transmit it to the industrial control computer (1) through the bus to reconstruct the three-dimensional contour of the weld. The intelligent algorithm unit includes a dual-robot task collaborative planning algorithm. The system uses a laser scanning camera (16) to scan and acquire three-dimensional point cloud data of multiple weld seams, extracts geometric parameters, and calculates the grinding amount of each weld seam. According to the anti-interference mechanism, the six-axis robot (10) on the left grinds the weld seams one by one from the initial position on the left end of the workpiece (18) to the center of the workpiece (18). The six-axis robot (10) on the right slides from the right end to the center of the workpiece (18) and grinds the weld seams one by one to the right. The encoders of the motors (14) of the two robot grinding platforms will collect the torque and speed of the motors (14) in real time and convert them into pulse signals. These pulse signals are transmitted to the industrial control computer (1) via the bus and converted into the actual position coordinates of the robot grinding platform on the ground rail (7). Based on the position coordinates of the two robot grinding platforms, the distance between them is calculated in real time to ensure that the distance between the two machines is ≥3000mm at any time to avoid collisions. When allocating tasks, the total workload of each weld seam is calculated first, and then the weld seam tasks are equally allocated to the two six-axis robots (10) according to the principle of equalization. Finally, the tasks are adjusted according to the real-time situation through dynamic allocation. After the grinding is completed, the system will perform a pre-warning detection and effect evaluation.

2. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 1, characterized in that: The robotic grinding platform consists of a six-axis robot (10), a tool magazine (5), a motor (14), a base (11), a slider (12), a dust removal bracket (9), and a helical gear (13). The six-axis robot (10) is mounted on the ground rail (7) via a mounting base (11) and is used to scan the workpiece (18) to be ground and slide axially at different weld positions. A laser scanning camera (16) and a floating electric spindle (17) are installed at its end. The motor (14) is used to rotate the helical gear (13), and there is a rack on the ground rail (7) that meshes with it, so that the robotic grinding platform can slide along the slider (12) on the ground rail (7). The tool magazine (5) is installed next to the six-axis robot (10), and the tool magazine (5) contains grinding heads (15). When the head (15) is worn to a certain extent, the head (15) is automatically replaced. The dust removal bracket (9) supports the dust cover (3) and the dust removal pipe (23).

3. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 2, characterized in that: The tool magazine (5) is an automatic tool changing device for adaptive grinding and polishing tools. Two sliding doors are provided on one side of the tool magazine (5). When the tool wears down to the preset threshold, or when different specifications of tool heads (15) need to be replaced according to the specific requirements of the grinding process, the six-axis robot (10) can automatically replace the tool heads (15) placed in the tool magazine (5). Based on the visual scanning and feature extraction system, the workpiece is scanned to determine the different types of welds and different process treatment methods. The tool magazine (5) automatically replaces different tool heads (15) to complete the process treatment.

4. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 1, characterized in that: The grinding system consists of a floating electric spindle (17) and a cutting head (15); The grinding system is installed at the end of the six-axis robot (10). The control system of the floating electric spindle (17) is paired with the main control system through the bus to realize data interaction and remote control. The built-in high-precision displacement sensor monitors the gap change between the grinding tool and the workpiece (18) surface in real time, transmits the collected gap data to the intelligent control system, calculates the error and generates control commands, optimizes the spindle speed and feed speed, and ensures that the tool head (15) always maintains the best contact state with the workpiece (18) surface.

5. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 4, characterized in that: The grinding operation begins. The floating electric spindle (17) follows the path planned by the intelligent algorithm unit and moves to the grinding part of the workpiece (18) according to the path planned by the intelligent algorithm unit. When the grinding tool approaches the surface of the workpiece (18) by 1 mm, the displacement sensor starts the high-precision detection mode and transmits the gap data between the tool and the workpiece (18) to the intelligent control system in real time. The control system quickly calculates the radial and axial displacement that the electric spindle (17) needs to be adjusted according to the preset process requirements, and drives the floating module to adjust the action so that the grinding tool is precisely attached to the surface of the workpiece (18). During the grinding process, the operator can monitor the running status of the floating electric spindle (17) in real time through the host computer, including the rotation speed, vibration amplitude, and displacement compensation parameters.

6. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 1, characterized in that: The dust removal system consists of a dust removal duct (6), a dust cover (3), and a dust purification box (4). The grinding operation is carried out in the dust cover (3), which is supported by a dust removal bracket (9). The dust generated during the operation is purified through the dust removal pipe (23), the dust removal duct (6), and the dust purification box (4).

7. The autonomous identification and trajectory planning polishing system for complex irregular spatial components according to claim 1, characterized in that: The workpiece fixing fixture consists of a V-shaped support roller frame (20) and a cylinder clamping block (21); the workpiece (18) has straight ribs (22) and semi-circular ribs (19) on its sides; when the workpiece (18) is placed, the semi-circular ribs (19) are supported by the rollers in the V-shaped bracket (20), and at the same time, the cylinder clamping block (21) next to the bracket fixes the straight ribs (22) located below the workpiece (18).

8. A polishing method based on the system according to any one of claims 1-7, characterized in that, Includes the following steps: Step 1: Manually select the workpiece model and hoist the workpiece (18) to be ground onto the V-shaped support roller frame (20) by a crane. Place the semi-circular rib plate (19) on the roller. Step 2: The cylinder clamping block (21) on the fixed fixture clamps and fixes the straight rib plate (22) below the workpiece (18); Step 3: After the workpiece (18) is placed and fixed, the robot grinding platform on the left side drives the helical gear (13) to move along the direction of the workpiece (18) on the ground rail (7) via the motor (14). At the same time, the laser scanning camera (16) at the end of the six-axis robot (10) also starts to work, scanning all the weld positions, lengths, widths and thicknesses on the workpiece (18) and transmitting the information to the industrial control computer (1). Step 4: After the industrial control computer (1) obtains the weld information, the dual robot task collaboration algorithm first calculates the amount of grinding required for the weld, and then allocates the grinding task according to the anti-interference mechanism to ensure that all welds are reasonably allocated. Step 5: According to the task allocation in Step 4, both robot grinding platforms start to move. The six-axis robot (10) drives the floating electric spindle (17) at the end to perform grinding operations according to the preset path. Step 6: During the grinding operation, the dust removal system also starts working simultaneously to ensure the safety and cleanliness of the work area; Step 7: After the polishing is completed, the system will perform a pre-warning test and effect evaluation. If there are no problems, all equipment in the system will return to their initial positions. The six-axis robot (10) will return to its initial position and the display terminal will show that the polishing operation is completed and wait for manual removal of the parts.

Images