Bolt tightening device and battery assembly line
By integrating a bolt tightening device with a collaborative robot and a multi-axis robotic arm, a vision camera and tightening shaft are integrated. This solves the problem of difficult production line modification caused by the need for mechanical positioning mechanisms in the fixed torque system. It achieves mobile positioning and efficient tightening, reduces the misjudgment rate, and is suitable for battery assembly lines.
Patent Information
- Application Number
- CN202520356130.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-28
- Publication Date
- 2026-01-27
- Estimated Expiration
- 2035-02-28
AI Technical Summary
Existing torque-fixing systems require mechanical positioning mechanisms, which makes production line modifications difficult and costly. Furthermore, the camera positioning mechanism cannot be moved, limiting the application scope of automatic torque-fixing systems.
An integrated bolt tightening device is adopted, which integrates a first vision camera, a second vision camera, and a tightening shaft using a collaborative robot and a multi-axis robotic arm to achieve mobile positioning and multi-feature recognition, reduce the false judgment rate, and provide a foundation for AI recognition.
It achieves mobile positioning without the need for mechanical positioning mechanisms, simplifies the structure, reduces the misjudgment rate, and improves the accuracy and efficiency of bolt tightening. It is suitable for non-automatic workstations without the need for large-scale modifications.
Smart Images

Figure CN223833920U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the technical field, and more specifically to a bolt tightening device and a battery assembly line. Background Technology
[0002] With the continuous development of the power battery industry, labor costs in manufacturing remain high, and whether a battery product has a cost advantage is a crucial factor in determining its competitiveness.
[0003] Existing torque-fixing systems all use mechanical positioning combined with secondary guidance positioning via camera imaging to position bolts. For example, patent document CN115302236A discloses a method of lifting and positioning the product using a lifting cylinder and positioning pin. Therefore, when a new production line is put into operation, a mechanical positioning mechanism needs to be designed and implemented simultaneously. However, for production lines that are already in operation, there is no corresponding positioning mechanism. If the torque-fixing system is implemented, the production line needs to be modified accordingly. The modification of the production line is difficult and costly, which limits the application scope of the automatic torque-fixing system.
[0004] Patent document CN 116673715 A discloses a vision-guided robot for screw fastening and a screw fastening quality inspection device and control method in the field of vision screw fastening technology. The device includes a machine frame, a double-speed wheel conveyor system, a product positioning system, a screw fastening system, and a machine control system. The machine frame is connected to both ends of the production line. The product to be processed is automatically conveyed to the blocking mechanism via the double-speed wheel conveyor system. A sensor is installed on the blocking mechanism side to detect the product's arrival. The product positioning system includes a lifting mechanism and a positioning mechanism.
[0005] However, its camera is used for taking pictures and positioning, and it can only be located at a fixed workstation and cannot be moved, which makes it inconvenient to use. Utility Model Content
[0006] The technical problem to be solved by this utility model is how to provide a movable integrated bolt tightening device.
[0007] This utility model solves the above-mentioned technical problems through the following technical means: a bolt tightening device, including a collaborative robot, the collaborative robot including a multi-axis robotic arm, the execution end of the multi-axis robotic arm is fixedly connected to a first vision camera, a second vision camera and a tightening shaft, the shooting ends of the first vision camera and the second vision camera and the tightening end of the tightening shaft are all arranged facing the part to be tightened, and the tightening shaft is arranged side by side with the first vision camera and the second vision camera.
[0008] By integrating the first vision camera, the second vision camera, and the tightening axis onto the execution end of the multi-axis robotic arm of the collaborative robot, the movement of the first vision camera, the second vision camera, and the tightening axis is facilitated. At the same time, the tightening axis is integrated with the first vision camera and the second vision camera side by side, which simplifies the structure.
[0009] As a preferred technical solution, the first vision camera, the second vision camera, and the tightening shaft are arranged sequentially along the conveying direction of the part to be tightened. The first vision camera takes pictures of the battery pack, and the second vision camera takes pictures of the bolts on the battery pack. By taking pictures in sections for positioning, the accuracy of subsequent recognition is improved, and a hardware foundation is provided for subsequent AI recognition or automatic recognition.
[0010] As a preferred technical solution, the tightening shaft is capable of rotating around its axis.
[0011] As a preferred technical solution, it also includes a collaborative robot, wherein the execution end of the multi-axis robotic arm is fixedly connected to a connection platform, and the first vision camera, the second vision camera, and the tightening shaft are fixed to the connection platform.
[0012] As a preferred technical solution, the connection platform is also connected to a light source, which is located outside the shooting ends of the first vision camera and the second vision camera. The light source has a through hole through which light from the first vision camera and the second vision camera can pass.
[0013] As a preferred technical solution, the part to be twisted includes a bolt, and the bolt is provided with an outer circle, a hexagonal drive head, a chamfered circle and a cross groove in sequence.
[0014] As a preferred technical solution, the outer circle dimension is larger than the hexagonal drive head dimension, the hexagonal drive head dimension is larger than the chamfer circle dimension, and the chamfer circle dimension is larger than the cross groove dimension.
[0015] As a preferred technical solution, the tightening shaft is configured as a straight handle.
[0016] The battery assembly line includes at least one set of the bolt tightening device and a conveyor line, wherein the bolt tightening device is located on the outside of the conveying surface of the conveyor line.
[0017] As a preferred technical solution, it includes two sets of bolt tightening devices, which are located on both sides of the conveyor line and are symmetrically arranged about the conveyor line. By setting bolt tightening devices on both sides of the conveyor line, the bolt tightening efficiency can be improved. Attached Figure Description
[0018] Figure 1 This is a schematic diagram of the overall structure provided in Embodiment 1 of the present utility model;
[0019] Figure 2 A schematic diagram of prior art camera recognition features provided for Embodiment 1 of this utility model;
[0020] Figure 3 This is a schematic diagram of camera recognition features provided in Embodiment 1 of this utility model;
[0021] Figure 4 This is a schematic diagram of the outer circular structure provided in Embodiment 1 of this utility model;
[0022] Figure 5 This is a schematic diagram of the chamfered round structure provided in Embodiment 1 of this utility model;
[0023] Figure 6 This is a schematic diagram of the abnormal surface contamination structure of the bolt provided in Embodiment 1 of this utility model;
[0024] Figure 7 This is a schematic diagram of the cross bolt structure provided in Embodiment 1 of this utility model;
[0025] Figure 8 This is a schematic diagram of the overall structure provided for Embodiment 2 of the present utility model;
[0026] Figure 9 Provided for Embodiment 2 of this utility model Figure 8 A partially enlarged structural diagram;
[0027] Figure 10 This is a schematic diagram of bolt identification and comparison provided in Embodiment 1 of this utility model;
[0028] Reference numerals: 1. Base; 11. Battery pack; 111. Phillips head bolt; 112. Outer circle; 113. Hexagonal drive head; 114. Chamfered circle; 115. Phillips head groove; 12. Conveyor line; 13. Battery pack tray; 2. Collaborative robot; 3. Tightening shaft; 4. Vision device; 41. Connecting platform; 42. First vision camera; 43. Second vision camera; 44. Light source. Detailed Implementation
[0029] To make the objectives, technical solutions, and advantages of the embodiments of this utility model clearer, the technical solutions of the embodiments of this utility model will be clearly and completely described below in conjunction with the embodiments of this utility model. Obviously, the described embodiments are only some embodiments of this utility model, not all embodiments. Based on the embodiments of this utility model, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this utility model.
[0030] Example 1
[0031] See Figure 1A bolt tightening device includes a base 1, a collaborative robot 2, a tightening shaft 3, and a vision device 4. The shooting ends of the first vision camera 42 and the second vision camera 43 are both set facing the part to be tightened. The fixed end of the collaborative robot 2 is fixedly connected to the top of the base 1. The execution end of the collaborative robot 2 is connected to the vision device 4. The tightening shaft 3 is fixed on the vision device 4 and can follow the collaborative robot 2 to move, that is, follow the part to be tightened or move to the position of the part to be tightened. In this embodiment, the part to be tightened is the cross bolt 111 on the battery pack 11. The collaborative robot 2 is a commercially available multi-axis robotic arm. The tightening shaft 3 is set in the shape of a straight handle.
[0032] The vision device 4 includes a connecting platform 41, a first vision camera 42, a second vision camera 43, and a light source 44. The first vision camera 42, the second vision camera 43, and the tightening shaft 3 are fixed side by side on the connecting platform 41. The light source 44 is fixedly connected to the end of the connecting platform 41 facing the part to be tightened, i.e., the bottom of the connecting platform 41. The light source 44 is located outside the shooting end of the first vision camera 42 and the second vision camera 43, and has a through hole through which the light from the first vision camera 42 and the second vision camera 43 can pass.
[0033] See Figure 1 A drive component that drives the tightening shaft 3 to rotate is fixedly connected to the connecting platform 41. The drive component can drive the tightening shaft 3 to rotate around its axis. The drive component can be a commercially available motor.
[0034] The head of the cross bolt 111 is provided with an outer circle 112, a hexagonal drive head 113, a chamfered circle 114 and a cross countersunk groove 115 from the outside to the inside. The outer circle 112 is larger than the hexagonal drive head 113, the hexagonal drive head 113 is larger than the chamfered circle 114, and the chamfered circle 114 is larger than the cross countersunk groove 115.
[0035] See Figure 3 This is a traditional camera recognition system that extracts only a single feature. Compared to traditional camera image recognition, which only extracts the outer circle 112 feature and sets a corresponding threshold, if this point matches the outer circle 112 feature, it is considered a bolt and the position coordinates are output. However, in actual production, misjudgment often occurs. For example, other round holes, round blocks, or dirt on the outer circle contour within the field of view may cause misjudgment or failure to recognize the bolt, resulting in tightening failure.
[0036] See Figures 2-7The product in this embodiment provides the hardware foundation for subsequent AI recognition. The AI recognition mentioned in this embodiment refers to the feature recognition module. The first vision camera 42 and the second vision camera 43 are both electrically or communicatively connected to the feature recognition module. The feature recognition module can recognize multiple features of the cross bolt 111. The feature recognition module compares the similarity with the reference features and extracts the following features through machine learning of a large number of samples: similarity of outer circle 112, similarity of hexagonal drive head 113, similarity of chamfer circle 114, and similarity of cross groove 115. By combining the above features to determine whether the conditions are met, it can be determined whether this is a bolt and output the corresponding bolt position coordinates.
[0037] Specifically, the first vision camera 42 can take pictures of the battery pack location and extract features through the feature extraction module to pick up the battery pack location. The second vision camera 43 can take pictures of the cross bolt 111 on the battery pack and determine the position of the cross bolt 111 based on the relative position of the bolt and the MARK location through the feature extraction module. The false judgment rate is reduced by the two feature extractions.
[0038] The input conditions are: similarity of outer circle 112, similarity of hexagonal drive head 113, similarity of chamfer circle 114, similarity of cross groove 115, and the relative positions of each feature;
[0039] The output results are: NG, OK, N / A (to be determined), and position coordinates, such as the coordinates of battery pack 11 and cross bolt 111.
[0040] The algorithms, technical models, technical conditions, thresholds, etc. designed herein are all existing technologies and do not constitute innovations of this embodiment.
[0041] In this case, the relevant data calculation formulas during the verification process are as follows:
[0042] Success rate of bolt recognition = (Number of bolts successfully recognized and tightened / Total number of bolts) * 100%
[0043] Similarity feature misclassification rate = (Number of bolts successfully identified and tightened / Total number of similar bolts) * 100%
[0044] Dirty bolt false positive rate = (Number of bolts successfully identified and tightened) / (Total number of dirty bolts) * 100%
[0045] The recognition success rate statistics are derived from actual production, and the recognition success rate is statistically analyzed on-line, with a sample size of 1000 groups;
[0046] False positive rate for similar features: Set two similar features and verify repeatedly;
[0047] See Figure 4 A circular block 116 is placed next to bolt 111, with a sample size of 500 groups;
[0048] See Figure 5 Next to bolt 111, place hex bolt 117 (non-Phillips head), with a sample size of 500 groups;
[0049] See Figure 6 1. Bolt surface contamination verification case: 118 types of contamination features were set, and the outer circle contour was painted black with a paint pen. 500 bolt samples were collected.
[0050] See Figure 10 After verification, the AI technology was used to extract and analyze the features of the cross bolt 111. After learning and verification from a large number of samples, the corresponding threshold was optimized. Compared with traditional photo recognition, it has obvious advantages.
[0051] The success rate of identification has been greatly improved; it can accurately identify and judge anomalies such as dirt and similar features that occur in actual production.
[0052] Therefore, automation solutions can be provided for non-automatic workstations without the need for new mechanical positioning or large-scale modifications to pallets and production lines;
[0053] Compared to traditional camera systems, AI camera recognition systems can simultaneously extract and analyze multiple cross bolt features, which can significantly improve recognition accuracy, significantly reduce false judgment rate, and enhance error prevention capabilities.
[0054] It should be noted that the collaborative robot 2 is 600mm high, weighs 20kg, has an arm span of 1.6m, and has a tightening shaft 3 torque specification of 5-15Nm. The first vision camera 42 and the second vision camera 43 are both commercially available 2D cameras with 5 megapixels and a resolution of 2448×2048.
[0055] Example 2
[0056] See Figure 8 , Figure 9 The difference between this embodiment and embodiment 1 is that this embodiment provides a battery assembly production line including at least one set of bolt tightening devices and conveyor line 12 as in embodiment 1. In this embodiment, two sets of bolt tightening devices are used as an example. They are respectively set on both sides of the conveyor line 12 and are arranged symmetrically to improve tightening efficiency. The conveyor line 12 is a commercially available roller conveyor, which is responsible for carrying and transporting battery trays. The conveyor line 12 is provided with multiple battery pack trays 13 for carrying battery packs 111.
[0057] The above embodiments are only used to illustrate the technical solutions of this utility model, and are not intended to limit it. Although this utility model has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this utility model.
Claims
1. A bolt tightening device, characterized in that, The system includes a collaborative robot (2), which includes a multi-axis robotic arm. The execution end of the multi-axis robotic arm is fixedly connected to a first vision camera (42), a second vision camera (43), and a tightening shaft (3). The shooting ends of the first vision camera (42) and the second vision camera (43), and the tightening end of the tightening shaft (3) are all set towards the part to be tightened. The tightening shaft (3) is arranged side by side with the first vision camera (42) and the second vision camera (43).
2. The bolt tightening device according to claim 1, characterized in that, The first vision camera (42), the second vision camera (43), and the tightening shaft (3) are arranged sequentially along the conveying direction of the part to be twisted.
3. The bolt tightening device according to claim 1, characterized in that, The tightening shaft (3) is capable of rotating around its axis.
4. The bolt tightening device according to claim 1, characterized in that, The execution end of the multi-axis robotic arm is fixedly connected to a connection platform (41), and the first vision camera (42), the second vision camera (43), and the tightening shaft (3) are fixed on the connection platform (41).
5. A bolt tightening device according to claim 4, characterized in that, The connection platform (41) is also connected to a light source (44), which is located outside the shooting ends of the first visual camera (42) and the second visual camera (43). The light source (44) has a through hole through which light from the first visual camera (42) and the second visual camera (43) can pass.
6. A bolt tightening device according to claim 1, characterized in that, The part to be twisted includes a bolt, which is provided with an outer circle (112), a hexagonal drive head (113), a chamfered circle (114), and a cross countersunk groove (115) in sequence.
7. A bolt tightening device according to claim 6, characterized in that, The outer circle (112) is larger than the hexagonal drive head (113), the hexagonal drive head (113) is larger than the chamfer circle (114), and the chamfer circle (114) is larger than the cross groove (115).
8. A bolt tightening device according to claim 1, characterized in that, The tightening shaft (3) is configured as a straight handle.
9. A battery assembly line, characterized in that, It includes at least one set of bolt tightening devices as described in any one of claims 1-8 and a conveyor line (12), wherein the bolt tightening devices are located on the outside of the conveying surface of the conveyor line (12).
10. The battery assembly line according to claim 9, characterized in that, It includes two sets of bolt tightening devices, which are located on both sides of the conveyor line (12) and are symmetrically arranged about the conveyor line (12).
Citation Information
Patent Citations
Automatic screw tightening machine
CN115302236A
Visual guidance robot screw locking and locking quality detection equipment and control method
CN116673715A