Method and system for sensing, identifying and quickly retreating sliding table saw

Abstract

The invention relates to the field of use safety of sliding table saws, in particular to a method and a system for sensing, identifying and quickly retreating a sliding table saw, which are characterized by comprising an identifying unit, a monitoring unit, a driving unit, a quick active retreating unit and a braking unit. Wherein the recognition unit recognizes whether a person gets close to the sliding table saw is a hand or not through a visual sensor in real time, and through double insurance of the visual sensor and an infrared sensor of the monitoring unit, when the safety distance of the hand is smaller than a set safety threshold value, the person is reminded to get close to the sliding table saw. The active rapid concession unit drives the lifting motor to pull the connecting rod through the ball screw to enable the saw blade shaft to achieve active concession and drives the braking unit to control the main motor to be powered off so as to decelerate the saw blade shaft, and the protection function of the system is achieved. According to the invention, real-time monitoring can be carried out in the cutting use environment of the industrial sliding table saw, and hands can be protected from being hurt.

Description

Technical Field

[0001] The present invention relates to the field of safety in the use of panel saws, and particularly to a method and system for sensing, identifying, and quickly retreating of a panel saw. Background Art

[0002] In the use of panel saws, it often happens that workers are cut because they cannot always notice the potential safety hazards of the high-speed rotating panel saw blades in real time, and even serious cutting incidents occur, making the personal safety of workers unable to be guaranteed. Therefore, it is very important to add protective measures to panel saws.

[0003] At present, with regard to the personal safety protection of workers, some solutions have also been proposed in existing patents. For example, Chinese Patent CN111660375A discloses a cut-resistant bench saw. This invention patent achieves the purpose of preventing cuts by adopting a special design of the external housing so that the workpiece being processed and the human hand cannot touch the saw blade simultaneously during autonomous operation. However, this working mode obviously does not conform to the processing environment of complex and large workpieces in industrial production. For example, Chinese Patent CN201821576243.X discloses a safety table saw sliding workbench. The table saw sliding workbench includes a table saw tabletop and a sliding tabletop that can move along the table saw tabletop. The saw blade is arranged in a transparent material pressing protection cover in a covering manner. The front and rear ends of the material pressing protection cover can move up and down along the front fixed plate and the rear push plate. The transparent material pressing protection cover is arranged outside the saw blade, which can effectively prevent the saw blade from hurting people. But it also does not conform to the processing environment of complex and large workpieces in industrial production. For example, Chinese Patent CN110900717A discloses a protection device for preventing accidental injury of a panel saw, which includes inner and outer ring infrared temperature sensors arranged around the blade in the panel saw, and there is a gap between the inner and outer ring infrared temperature sensors. When both the inner ring infrared temperature sensor and the outer ring infrared temperature sensor sense a human body, the control system controls the blade brake to brake the blade. Although this device protects the operator, a large amount of heat is also released during the processing of the workpiece, resulting in an increase in the ambient temperature, which may cause the inner and outer ring infrared temperature sensors to be unable to accurately determine whether it is a human body or a workpiece. For example, Chinese Patent CN201721164339.0 discloses an electronically controlled protective table saw based on the principle of capacitive touch. This invention decelerates the saw blade by simultaneously reversing the motor in a capacitive touch sensing manner. This method not only has an obvious deceleration effect but also has certain requirements for the hardware sensitivity and lacks consideration for safety. For example, Chinese Patent CN201810457214.X discloses a cut-resistant table saw that does not damage the saw blade. It includes a table saw installation shell, a saw blade, a driving motor, a second belt, and a gunpowder driving structure. During the working process, if people accidentally touch the saw blade with their fingers during use, it also uses a capacitive touch sensing method and uses gunpowder explosion to drive the action mechanism. It not only has high requirements for the hardware sensitivity and has certain dangers but also uses gunpowder as the driving energy source and lacks consideration for safety. These patents can all protect workers to a certain extent, but they all lack consideration in terms of safety or some aspects, thus there is a possibility of cuts occurring.

[0004] The present invention uses camera machine vision deep learning to recognize human hands and an infrared sensor for distance measurement. Under the dual insurance verification of camera machine deep learning and infrared ranging, when the machine vision deep learning first recognizes that the object approaching the saw blade is a human hand rather than a workpiece, and when the machine vision or infrared ranging detects that the distance between the human hand and the saw blade is less than the dangerous threshold, the lifting motor starts to work, causing the ball screw connected to the saw blade shaft to quickly pull down the saw blade shaft, completing a quick and active retraction, and cutting off the power supply of the drive motor to decelerate the saw blade shaft. The human-machine interaction control panel is provided with a one-key reset button. When an abnormal safety retraction occurs, the one-key reset button needs to be pressed first before the start button can be pressed to raise the saw blade shaft to the workbench surface. This utility model design has rapid perception and recognition by machine vision deep learning, sufficient hardware response time, low performance requirements, reduced costs, and higher safety performance. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for a push table saw to sense, recognize, and quickly retract, aiming at the deficiencies of the existing technology. By adding machine deep learning and an infrared sensor module, it can automatically sense and recognize whether the object approaching the saw blade is a human hand, and measure the relative distance between the human hand and the saw blade in real time. When the relative distance between the human hand and the saw blade is less than the safety threshold, the driving lifting motor pulls down the saw blade quickly through a ball screw, realizing a quick and active retraction, and cutting off the power supply of the drive motor to achieve deceleration and braking of the saw blade. It has the characteristics of intelligent sensing and recognition, quick and active retraction control, etc. in the field of real-time anti-cutting during industrial cutting.

[0006] The technical problems to be solved by the present invention are realized by the following technical solutions.

[0007] A method and system for a push table saw to sense, recognize, and quickly retract, including an identification unit, a monitoring unit, a driving unit, an active retraction control unit, and a braking unit.

[0008] Preferably, the identification unit includes a vision sensor, and the vision sensor is fixed to the bracket by bolts. The bracket is connected to the table board by bolts.

[0009] For the identification unit system, according to machine vision deep learning, by extracting pixel point features, the method for the human hand identification unit through the vision sensor includes the following steps:

[0010] Step 1: Train a human hand data set through a deep learning model;

[0011] Step 2: Test the trained model with real-time images from the vision sensor;

[0012] Step 3: Use deep learning training to take the real-time image of the vision sensor as the input and the target detection result as the output, establish the corresponding hardware control relationship, and complete the identification unit.

[0013] Preferably, the monitoring unit includes an infrared sensor and a vision sensor. The infrared sensor is fixed on the working plane by bolts, connected to the perception and recognition system control, and connected to the hardware control system through wires with the vision sensor. When the vision sensor or the infrared sensor detects that the distance between the human hand and the saw blade is less than the safety threshold, it controls the lifting motor to drive.

[0014] Preferably, the driving unit includes a main motor, a first pulley, a belt, a second pulley, a saw blade shaft and a saw blade. It is characterized in that: the main motor is connected to the first pulley through the motor rotating shaft, the first pulley is connected to the second pulley through the belt, the saw blade shaft is installed on the rotating shaft of the second pulley, and the saw blade can be installed on the saw blade shaft and replaced reliably.

[0015] Preferably, the active retreat control unit includes a lifting motor, a connecting rod and a ball screw. It is characterized in that: the lifting motor is connected to the ball screw, and the ball screw is connected to the saw blade shaft through a connecting rod bolt.

[0016] The braking unit includes a main motor and a hardware control system. The hardware control system is connected to the main motor through wires.

[0017] Compared with the existing technology, the effective effect of the present utility model is:

[0018] Through the perception, recognition and rapid active retreat module, under the dual-reliability monitoring of machine vision and infrared sensors, the safety threshold between the human hand and the saw blade is perceived, recognized and monitored in real time. When the worker is feeding and cutting, the vision sensor recognition unit recognizes that the object close to the saw blade is a human hand through deep learning object detection. When the monitoring unit's infrared sensor and vision sensor detect that the human hand reaches the safety threshold and the distance between the human hand and the saw blade is less than the safety threshold, the lifting motor drives the ball screw to pull down the saw blade shaft, realizing rapid active retreat control, and cutting off the power of the first motor to decelerate the saw blade to achieve braking. This method can effectively ensure the safety of the operator during the operation and can protect the saw blade to the greatest extent. Description of the Drawings

[0019] The present invention will be further described below with reference to the drawings.

[0020] Figure 1 It is the overall structure design drawing of the present invention;

[0021] Figure 2 It is the structure design drawing of the movement core component of the present invention;

[0022] Figure 3 It is the step diagram of the human hand target detection training model of the present invention;

[0023] Figure 4 It is the operation flow chart of the present invention;

[0024] Among them, the reference numerals in the figure are as follows:

[0025] 1. Connecting rod; 2. Main motor; 3. Lifting motor; 4. Ball screw; 5. Saw blade shaft; 6. Bracket; 7. First pulley; 8. Second pulley; 9. Workbench; 10. Infrared sensor; 11. Vision sensor; 12. Shield assembly; 13. Movement core assembly; 14. Keyboard touch screen. Embodiment

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] Please refer to the accompanying drawings of the specification. A method and system for perception, recognition and rapid retraction of a panel saw includes an identification unit, a monitoring unit, a driving unit, an active retraction control unit and a braking unit. The identification unit includes a vision sensor, and the vision sensor is fixed to the bracket by bolts, and the bracket is connected to the workbench surface by bolts.

[0028] The monitoring unit includes an infrared sensor. The infrared sensor is fixed to the working plane by bolts, connected to the entire perception and recognition system, and connected to the vision sensor through a wire to the hardware control system.

[0029] The driving unit includes a main motor, a first pulley, a belt, a second pulley, a saw blade shaft and a saw blade. It is characterized in that: the main motor is connected to the first pulley through the motor rotating shaft, the first pulley is connected to the second pulley through the belt, and the saw blade shaft is installed on the rotating shaft of the second pulley. The main motor is connected to the power supply required by the motor, and the main motor is connected to the hardware control system through a wire.

[0030] The active retraction control unit includes a lifting motor, a connecting rod and a ball screw. It is characterized in that: the lifting motor is connected to the ball screw, and the ball screw is connected to the saw blade shaft through a connecting rod bolt.

[0031] The braking unit includes a main motor and a hardware control system. The hardware control system is connected to the main motor through a wire.

[0032] In use, power on the hardware control system, the main motor, and the lifting motor. The vision sensor uses machine deep learning to continuously identify whether there are human hands around the saw blade in real time. When the vision sensor and the infrared sensor simultaneously identify a human hand and sense and monitor the safety threshold between the human hand and the saw blade, the safety threshold is set at 50 mm in its embodiment. When the distance between the human hand and the saw blade shaft is less than the safety threshold, a control signal is output, and the lifting motor starts to work to quickly lower the saw blade shaft through the ball screw to complete the active quick retraction. At the same time, the control signal will cut off the power supply of the main motor to decelerate the main motor to achieve braking. And there is a one-key reset function on the human-machine interaction control panel. After an abnormal safety retraction occurs, it is necessary to press the one-key reset button to start working again to raise the saw blade shaft to the working surface, realizing the product function requirements of preventing human body from being cut in industrial cutting. At the same time, the hardware control reaction meets the fast response time, coordinates the work control stably without damaging the saw blade, ensures personnel safety, and reduces production costs.

[0033] As described above, only the preferred specific embodiments of the present invention are provided, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and these changes and improvements should be covered within the protection scope of the present invention.

Claims

1. A method and system for sensing, identifying and quickly retracting a panel saw, comprising an identification unit, a monitoring unit, a driving unit, an active quick retraction unit and a braking unit, characterized in that: The described recognition unit includes a vision sensor (11), the vision sensor (11) is fixed to a bracket (6) by bolts, the bracket (6) is connected to a workbench (9) by bolts, the vision sensor is connected to a hardware control system through a wire. The recognition unit system, according to machine vision deep learning, extracts its features through pixel point features. It is characterized in that the method for the hand recognition unit by the vision sensor includes the following steps: Step 1: Train a hand dataset through a deep learning model; Step 2: Test the trained model with real-time images from the vision sensor; Step 3: Use deep learning training to take the real-time image of the vision sensor as the input and the target detection result as the output, establish the corresponding hardware control relationship, and complete the recognition unit; The described monitoring unit includes an infrared sensor (10) and a vision sensor (11). It is characterized in that: the infrared sensor (10) is fixed to the workbench (9) by bolts, and the vision sensor (11) and the infrared sensor (10) are connected to a hardware control system through wires.

2. The method and system for perception, recognition and rapid retraction of a panel saw according to claim 1, characterized in that: The described driving unit includes a main motor (2), a first pulley (7), a second pulley (8), and a saw blade shaft (5). The main motor (2) and the first pulley (7) are connected by a spline under the workbench (9). The first pulley (7) and the second pulley (8) are connected by a belt. The second pulley (8) and the saw blade shaft (5) are connected by a spline. The saw blade shaft (5) is connected to a saw blade by a spline. The main motor (2) is connected to the power supply required for this motor, and the main motor (2) is connected to a hardware control system through a wire; The described active rapid retraction unit includes a lifting motor (3), a ball screw (4), a connecting rod (1), and a saw blade shaft (5). It is characterized in that: the lifting motor (3) is connected to the ball screw (4) by bolts, the ball screw is connected to the saw blade shaft (5) through the connecting rod (1), the lifting motor (3) is connected to the power supply required for this motor, and the lifting motor (3) is connected to a hardware control system through a wire.

3. The method and system for sensing, identifying, and quickly retracting a panel saw according to claim 2, characterized in that: The described braking system includes a main motor (2), and the main motor is connected to a hardware control system through a wire.

Images