Industrial mechanical arm

Abstract

The industrial mechanical arm comprises a transmission arm used for achieving the form change of the industrial mechanical arm; the end effector is used for realizing the grabbing and / or instruction operation of the industrial mechanical arm on a target object; the working visual feedback part is arranged close to the end effector, is used for obtaining working area environment feedback and comprises a structured light depth camera for obtaining depth data in a first visual field range; and the safety visual feedback part is arranged on the industrial mechanical arm, is used for acquiring operating area environment feedback and comprises a dToF depth camera for acquiring depth data in a second visual field range. A single wide-angle and sensitive'safety eye 'is additionally provided for an Eye-in-Hand 3D visual system except a'working eye', so that personnel mistakenly entering the working area of the mechanical arm can be identified more timely and accurately in a larger range, and alarming and braking are performed, so that stable safety protection of the whole working area is realized.

Description

Technical Field

[0001] This disclosure relates to the field of industrial automation, and more particularly to an industrial robotic arm. Background Technology

[0002] Industrial robotic arms are robotic devices used for automated production, precision manufacturing, assembly, and material handling, and are widely used in industries such as automotive manufacturing, electronics assembly, medical, and logistics. With the trend of intelligentization and computer vision development, eye-in-hand 3D vision systems (i.e., depth cameras mounted at the end of the robotic arm) are widely used to improve the robotic arm's environmental perception, target recognition, and adaptive operation capabilities.

[0003] In industrial automation scenarios, employees accidentally entering the robotic arm's operating area can lead to serious safety accidents. While the probability of such accidents can be reduced through management regulations and personnel training, it is still desirable to fundamentally prevent them from occurring through the system's own technical safeguards.

[0004] Therefore, an improved industrial robotic arm is needed. Utility Model Content

[0005] One technical problem this disclosure aims to solve is to provide an industrial robotic arm that, in addition to utilizing a structured light camera to provide high-precision depth vision, is separately equipped with a wide-angle dToF depth camera to detect the presence of people in the work area and to issue timely alarms and brakes, thereby fundamentally eliminating the possibility of safety accidents.

[0006] According to one aspect of this disclosure, an industrial robotic arm is proposed, comprising: a transmission arm for realizing shape changes of the industrial robotic arm; an end effector for realizing the industrial robotic arm grasping and / or commanding operations on a target object; a working visual feedback unit arranged near the end effector for acquiring environmental feedback in the working area and including a first depth camera for acquiring depth data within a first field of view; and a safety visual feedback unit arranged on the industrial robotic arm for acquiring environmental feedback in the operating area and including a second depth camera for acquiring depth data within a second field of view, wherein the first depth camera includes a structured light depth camera, and the second depth camera includes a dToF depth camera.

[0007] Optionally, the industrial robotic arm may further include an alarm device connected to the safety visual feedback unit, for issuing an alarm in response to the second depth camera detecting the presence of a person within a first area of ​​the second field of view.

[0008] Optionally, the industrial robotic arm may further include a braking device connected to the safety vision feedback unit, for braking the operating industrial robotic arm in response to the second depth camera detecting the presence of a person in a second area of ​​the second field of view.

[0009] Optionally, the working visual feedback unit is connected to the alarm device and the braking device, and includes: a personnel reporting device for reporting to the alarm device and / or the braking device in response to the first depth camera detecting the presence of a person within the first field of view.

[0010] Optionally, the second depth camera includes at least one of the following: two dToF depth cameras arranged on both sides of the structured light depth camera; a dToF depth camera arranged outside the structured light depth camera; a dToF depth camera arranged on the upper side of the structured light depth camera; and a dToF depth camera arranged on the transmission arm base.

[0011] Optionally, the first depth camera includes: a structured light projection device for projecting structured light into a first field of view; and first and second image sensors having a predetermined relative spatial relationship for imaging the first field of view onto which the structured light is projected to obtain first and second two-dimensional images.

[0012] Optionally, the first depth camera further includes a visible light image sensor, used to image the first field of view to obtain a visible light image when the structured light projection device does not project the structured light.

[0013] Optionally, the second depth camera includes: a light source module for projecting area array light into a second field of view; and a dToF sensor for generating a sensing signal characterizing distance based on the reception time of the returned light from the projected area array light.

[0014] Optionally, the dToF sensor includes: an area array sensor composed of multiple avalanche photoelectric sensors, and the light source module includes: a light-emitting device for generating a laser beam; and a diffuser arranged in the propagation path of the beam to convert the laser beam into a wide-angle area array light.

[0015] Optionally, the transmission arm is a multi-axis transmission arm, and the working visual feedback unit and / or the safety visual feedback unit are arranged on one end of the multi-axis transmission arm near the end effector.

[0016] Optionally, the industrial robotic arm may further include: a connection portion for connecting the end effector to the drive arm, and the work vision feedback portion and / or the safety vision feedback portion are arranged on the connection portion.

[0017] This disclosure reduces the field of view requirement of the "working eye" by providing an additional wide-angle and sensitive "safety eye" in addition to the "working eye" for the Eye-in-Hand 3D vision system, allowing it to focus more on high-precision imaging within the work area. At the same time, it can identify personnel who have accidentally entered the robotic arm's work area in a wider range and more timely and accurate manner, and issue alarms and brakes, thereby achieving reliable safety protection throughout the entire work area. Attached Figure Description

[0018] The above and other objects, features and advantages of this disclosure will become more apparent from the more detailed description of exemplary embodiments thereof taken in conjunction with the accompanying drawings, wherein like reference numerals generally denote like parts.

[0019] Figure 1 A schematic diagram of the appearance of an industrial robotic arm according to an embodiment of the present invention is shown.

[0020] Figure 2 A schematic diagram of the field of view of the first and second depth cameras is shown.

[0021] Figure 3 This illustrates a typical working scenario for an industrial robotic arm.

[0022] Figure 4 A schematic diagram of the structured light depth camera used in industrial robotic arms is shown.

[0023] Figure 5 A schematic diagram of the composition of a dToF depth camera used in an industrial robotic arm is shown. Detailed Implementation

[0024] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0025] Industrial robotic arms are robotic devices used for automated production, precision manufacturing, assembly, and material handling, widely applied in industries such as automotive manufacturing, electronics assembly, medical, and logistics. Driven by the trends of intelligentization and computer vision development, eye-in-hand vision systems are widely used to improve the robotic arm's environmental perception, target recognition, and adaptive operation capabilities. Here, eye-in-hand refers to a camera mounted at the end of the robotic arm, moving with it to acquire visual information. This structure allows the robot to perceive the position, shape, depth, and other information of target objects in real time to adjust its operational path. Eye-in-hand robotic arms combine computer vision, deep learning, and motion control technologies to achieve intelligent operation. While in some implementations, robotic arms can use ordinary cameras to provide 2D vision for detecting object shape, color, barcodes, etc., in broader applications, eye-in-hand systems provide 3D vision through depth cameras, thereby enabling more complex functions such as object grasping, target tracking, and high-precision assembly.

[0026] In industrial automation scenarios, employees accidentally entering the robotic arm's operating area can lead to serious safety accidents. While the probability of such accidents can be reduced through management regulations and personnel training, it is still desirable to fundamentally prevent them from occurring through the system's own technical safeguards.

[0027] To this end, this disclosure proposes an industrial robotic arm that, in addition to utilizing a structured light camera to provide high-precision depth vision, is separately equipped with a wide-angle dToF depth camera to detect the presence of people in the work area and to issue timely alarms and brakes, thereby fundamentally eliminating the possibility of safety accidents.

[0028] Figure 1 A schematic diagram of the appearance of an industrial robotic arm according to an embodiment of the present invention is shown. As shown, the industrial robotic arm 10 can be arranged on a workbench (or on the ground, or in other suitable locations), and has a drive arm 110, an end effector 120, a work vision feedback unit 130, and a safety vision feedback unit 140.

[0029] The transmission arm 110 is used to realize the shape changes of the industrial robotic arm. It is a key component in the industrial robotic arm responsible for transmitting motion and force, and enables the robotic arm to perform various rotation, swing, and extension operations. The transmission arm typically includes multiple joints and links. Joints are connection points that provide selective or linear motion, while links are usually rigid structures used to connect the joints and determine the length and range of motion of the robotic arm. In addition, although not shown in the figure, the transmission arm 110 also needs to include a drive system (e.g., motor drive, hydraulic drive, or pneumatic drive system) and control the shape of the joints and links through a transmission mechanism (e.g., gears, lead screws, or timing belts). As shown in the figure, the transmission arm 110 can be a multi-axis transmission arm. A multi-axis transmission arm refers to a robotic arm with multiple rotational or linear motion axes, which realizes complex spatial motion through multiple degrees of freedom, thereby realizing the position changes of the end effector connected to it in multiple directions.

[0030] The end effector 120 is used to enable the industrial robotic arm to grasp and / or perform command operations on target objects. An end effector is a device installed at the end of a robotic arm, responsible for directly contacting external objects and performing specific tasks. It is analogous to the robot's "hand," determining the robotic arm's final function, such as grasping, welding, applying glue, or grinding. In the illustrated example, the end effector 120 is a suction cup for grasping objects. The suction cup utilizes vacuum to adsorb items and is suitable for objects with flat, smooth surfaces. In other implementations, the end effector can also have other forms, such as mechanical grippers for holding, transporting, and grasping objects, welding tools for welding, or nozzles for spraying paint, glue, or adhesives, etc.

[0031] To achieve eye-in-hand 3D vision, the robotic arm 10 also includes a working vision feedback unit 130 located near the end effector 120 for acquiring environmental feedback in the working area. Specifically, the working vision feedback unit 130 may include a first depth camera for acquiring depth data within a first field of view. When the working vision feedback unit 130 includes only one first depth camera, the working area of ​​the working vision feedback unit 130 corresponds to the first field of view of the first depth camera. When the working vision feedback unit 130 is equipped with multiple first depth cameras, the working area of ​​the working vision feedback unit 130 corresponds to the sum of the first fields of view of the multiple first depth cameras (in this case, calibration is required between the multiple first depth cameras). In this disclosure, the first depth camera can be implemented as a structured light depth camera, that is, acquiring depth data within the first field of view by actively projecting structured light and performing two-dimensional image capture and transformation. The following will combine... Figure 4 Describe in detail the composition and imaging principle of a structured light depth camera.

[0032] The first depth camera acts as the "eye" in Eye-in-Hand 3D vision, acquiring environmental feedback from the work area to guide the movement of the drive arm and the specific actions of the end effector. To achieve precise operation, the field of view of the first depth camera is typically not very large (the projected structured light needs to cover the field of view; if the field of view is too large, the structured light is usually more complex and requires higher projection power). For ease of understanding, Figure 1 The first field of view of the first depth camera is shown in the form of a pale red light cone.

[0033] In this disclosure, the working area of ​​the robotic arm refers to the precise operating area where the end effector actually performs its tasks, typically requiring high-precision vision support. However, in actual operation, the entire range of motion of the robotic arm (i.e., the operating area) encompasses all possible motion trajectories. In other words, the operating area of ​​the robotic arm is larger than its working area. High-precision imaging of the working area by a first depth camera ensures correct operation of the end effector, but it cannot achieve full-range imaging of the robotic arm's operating area. With only a first depth camera mounted on the robotic arm, a safety risk can easily arise if someone accidentally enters the operating area while the robotic arm is working.

[0034] Therefore, such as Figure 1 As shown, the industrial robotic arm 10 of this disclosure also includes a safety visual feedback unit 140 disposed on the industrial robotic arm, used to acquire environmental feedback of the operating area and including a second depth camera for acquiring depth data within a second field of view. In this disclosure, the second depth camera can be a dToF depth camera. Here, "dToF" corresponds to "direct time of flight," and the dToF depth camera calculates depth data by actively projecting a laser (implemented in this disclosure as an invisible area array) and based on the time it takes for the laser to return to the dToF sensor. Compared to structured light depth cameras, which typically require complex calculations, dToF depth cameras are more sensitive, have a larger field of view, and a simpler imaging principle, making them suitable for safety feedback of the operating area. The following will combine... Figure 5 Describe in detail the composition and imaging principle of a dToF depth camera.

[0035] Unlike a first depth camera that needs to be positioned at the end of the robotic arm (i.e., close to the end effector) to provide eye-in-hand functionality, a second depth camera for safety visual feedback can be positioned at any suitable location on the industrial robotic arm. Figure 1 In the example, the second depth camera is positioned close to the first depth camera and has a larger field of view than the first depth camera. Figure 2 A schematic diagram of the field of view of the first and second depth cameras is shown. Figure 2The second depth camera's second field of view is shown in the form of a deep red light cone. Thus, not only does the "working eye" move with the "hand," but it also has a wider field of view and is more sensitive to personnel intrusion; the "safety eye" also moves with the "hand."

[0036] In some embodiments, the robotic arm itself can react to anomalies detected by the second depth camera, for example, by issuing an alarm or braking. The robotic arm may include an alarm device connected to the safety visual feedback unit, for issuing an alarm in response to the second depth camera detecting the presence of a person within a first area of ​​the second field of view. The alarm may include an audible warning, such as a verbal alarm saying "Danger area, leave immediately" from a speaker on the robotic arm, or a buzzer sounding. The alarm may also include visual warnings, such as flashing red lights. In some embodiments, the alarm device may also report abnormal intrusion events to the production line and workshop's central control system for the elimination of current and subsequent hazardous actions.

[0037] The robotic arm may also include a braking device connected to the safety vision feedback unit, for braking the operating industrial robotic arm in response to the second depth camera detecting the presence of a person in a second area of ​​the second field of view. In some embodiments, both the alarm device and the braking device may be included within the operating area environmental feedback operating device. That is, the same device on the robotic arm can generate an alarm or braking signal based on a signal generated by the second depth camera. Furthermore, the "first area" of the second field of view may be the outer area of ​​the second field of view or an area far from the robotic arm, corresponding to an area within the field of view of the second depth camera that has not yet reached the robotic arm's operating area. Therefore, in this case, only an alarm may be issued without braking the robotic arm. Conversely, the "second area" of the second field of view may be the central area of ​​the second field of view or an area close to the robotic arm, corresponding to an area within the field of view of the second depth camera that has approached or even reached the robotic arm's operating area. In this case, forced braking of the robotic arm is required to prevent accidents. In some embodiments, the first and second areas may also have the same or similar range; for example, an alarm and braking may be triggered simultaneously upon detecting the approach of a person. In other embodiments, an alarm can be triggered and the operating speed of the robotic arm can be reduced when a person approaches, and a braking operation can be performed when the person approaches further, etc.

[0038] For ease of understanding, Figure 3This diagram illustrates a typical working scenario for an industrial robotic arm. As shown, the robotic arm needs to grasp goods from stack 1 and move them to stack 2. The robotic arm may include a central control unit (NCU) for incorporating visual feedback to control its movement and grasping operations. Specifically, under the guidance of a safety visual feedback unit, the NCU controls the robotic arm to accurately grasp the goods on top of stack 1. After grasping the goods, the robotic arm moves directly to stack 2, confirms the top of stack 2, and completes the placement of the grasped goods. Figure 3 In the scenario shown, if an employee accidentally enters the robotic arm's operating area, the movement path of the end effector grasping the goods, as indicated by the dotted line in the diagram, could cause injury. At this point, because the second depth camera in the safety vision feedback unit has a wider field of view (corresponding to the second field of view in the diagram), it can promptly detect the employee's intrusion into the operating area and determine that there is an obstruction in the movement path of the robotic arm. The robotic arm's operation can then be braked and an alarm can be issued to prevent personal injury.

[0039] As described above, the alarm and braking devices in this disclosure are connected to a safety visual feedback unit and can issue an alarm and / or brake based on the detection of a person by the second depth camera (e.g., based on head and shoulder detection) or when a person is detected to be sufficiently close. In some embodiments, the first depth camera can also detect the entry of a person while providing eye-in-hand vision guidance to the robotic arm. In this case, the working visual feedback unit can also be connected to the alarm device and the braking device, and includes a person reporting device for reporting to the alarm device and / or the braking device in response to the first depth camera detecting the presence of a person within the first field of view. Thus, the safe operation of the robotic arm can be ensured in extreme cases, such as when the safety visual feedback unit fails, preventing accidents.

[0040] exist Figure 1-3 In the example, the second depth camera is implemented as a dToF depth camera positioned outside the structured light depth camera; that is, the first and second depth cameras are arranged adjacent to each other and do not undergo relative displacement during robotic arm operation. Alternatively or supplementarily, the dToF depth camera can be positioned on either side (i.e., two dToF depth cameras), above, or below the structured light depth camera. The dToF depth camera can also be positioned on the base of the drive arm; in this case, the dToF depth camera typically does not shift during robotic arm operation, but its relative position to the structured light depth camera changes. The advantage of this configuration is that the dToF depth camera provides a relatively fixed field of view, without any shift in the field of view during robotic arm operation, thus enabling safe detection of the operating area with less computational effort.

[0041] like Figure 1As shown, the transmission arm 110 can be a multi-axis transmission arm, and the end effector 120 is connected to one end of the multi-axis transmission arm via a connecting part. Figure 1 In one example, both the work vision feedback unit 130 and the safety vision feedback unit 140 are arranged on the end of the multi-axis drive arm 110 near the end effector 120. In other examples, the work vision feedback unit and / or the safety vision feedback unit may also be arranged on the connecting part. In any implementation, in order to achieve hand-eye coordination, the work vision feedback unit 130 and the base of the end effector 120 need to maintain a constant relative position during the operation of the robot arm, while the safety vision feedback unit 140 does not have this strict requirement.

[0042] In this invention, a depth camera capable of actively projecting structured light is used as the first depth camera 130 to acquire depth information of the field of view. Figure 4 A schematic diagram of the structured light depth camera used in industrial robotic arms is shown. Figure 1-3 In the example shown, the light inlet / outlet window can be angled downwards to capture images of the work area. In other examples, the orientation of the structured light depth camera can vary depending on the specific task at hand.

[0043] As shown in the figure, the first depth camera 130 includes a structured light projection device 131 for projecting structured light into the working area. To distinguish it from ambient light, the structured light projection device 131 projects invisible light, such as infrared or ultraviolet structured light. Here, the structured light can be speckle structured light or striped structured light. The first depth camera 130 can use an image sensor to image the projected structured light and calculate the depth information of the field of view from the captured two-dimensional image. Figure 4 In the example, the first depth camera 130 adopts a binocular imaging structure, that is, it is equipped with two image sensors, a first image sensor 132 and a second image sensor 133. The two image sensors have a predetermined relative spatial position relationship, and each of them images the target area on which the structured light is projected to obtain first and second two-dimensional images. Subsequently, the three-dimensional distribution depth information of the goods within the field of view can be obtained by pixel window matching and triangulation calculation, that is, the height z value of each point on the xy distribution plane of the structured light in the measured space (target area) is calculated, and the three-dimensional information of the current goods in the target area is obtained thereby.

[0044] In some embodiments, the first depth camera 130 may further include a visible light image sensor 134, preferably an RGB sensor. The RGB sensor 134 can capture images at times other than when the structured light is projected to obtain visible light images of the target area, which can be combined with previously acquired depth information to obtain more accurate cargo volume and location information.

[0045] Although Figure 4 The image shows a depth camera (3D camera) based on binocular imaging with visible light imaging capabilities. However, it should be understood that in other embodiments, depth imaging can also be achieved based on a comparison of a monocular image with a reference image, and visible light imaging capabilities are optional.

[0046] After the first depth camera 130 captures the images required for calculating depth / 3D information, a computing device can be used to determine the 3D distribution depth information of the target object within the work area based on the first and second 2D images (and optionally, visible light images). The computing device can be located within the first depth camera 130, in which case the first depth camera 130, in addition to its measurement function, also has a depth data generation function. In other embodiments, the computing device can be located outside the first depth camera 130, for example, within a separately configured control device of the industrial robotic arm. Regardless of the configuration, the first depth camera 130 needs to communicate with the outside (e.g., wired or wireless communication) to send out the captured image information and / or depth calculation results for controlling the operation of the robotic arm 10.

[0047] As mentioned earlier, the Eye-in-Hand 3D vision system is a visual servoing system. A depth camera is mounted on the end effector of a robotic arm and moves with the arm, enabling the robot to perceive its environment and adjust its operations in real time. The core of the Eye-in-Hand 3D vision system lies in 3D visual feedback control, ensuring that the end effector accurately completes its tasks.

[0048] In the hand-eye coordination process of the industrial robotic arm of this invention, the first depth camera first acquires image information (the acquired image is a two-dimensional image of projected invisible structured light, but depth information can be obtained from it through calculation; alternatively, when the first depth camera includes a visible light sensor, it can also acquire conventional two-dimensional image information, such as RGB image information), and identifies the position, posture, and size of the target object through computer vision algorithms. Subsequently, the system converts the visual information into motion coordinates that the robotic arm can understand and performs path planning to ensure that the robotic arm can efficiently reach the target position. During execution, the robotic arm continuously receives visual feedback and relies on position servo (PBVS) to complete dynamic control and make real-time adjustments to improve operational accuracy and adaptability. Specifically, the system generates joint motion commands based on the real-time comparison of the deviation between the current three-dimensional coordinates of the target object and the desired coordinates. At this point, image processing, error calculation, and motion command generation typically need to be completed in a very short time, usually requiring a closed-loop control cycle of less than 50 milliseconds to avoid motion lag caused by visual delay, thereby enabling the robotic arm to complete precise operations under visual guidance, for example... Figure 3The robot arm can be used for various tasks, including picking up goods, grasping and assembling moving parts, or continuously processing irregular surfaces. The entire process forms a closed-loop control chain of "perception-computation-action-re-perception," and through continuous iterative optimization, it maintains high precision and strong adaptability in dynamic environments.

[0049] Furthermore, as mentioned above, in this invention, a dToF depth camera is used as a second depth camera 140 to acquire depth information of the second field of view. Figure 5 A schematic diagram of the components of a dToF depth camera used in an industrial robotic arm is shown. As shown, the second depth camera 140 may include: a light source module 141 for projecting an area array of light into a second field of view; and a dToF sensor 142 for generating a sensing signal characterizing distance based on the reception time of the returned light from the projected area array. Here, the light source module 141 may be used to project an invisible area array of light (e.g., an infrared area array) into the second field of view, and may include: a light-emitting device for generating a laser beam; and a diffuser arranged in the propagation path of the beam to convert the laser beam into a wide-angle area array of light. Accordingly, the dToF sensor 142 may also be an area array sensor, such as an area array sensor composed of multiple avalanche photoelectric sensors. Similar to the first depth camera 130, the second depth camera 140 may also include a computing device for generating depth information based on the sensing signal and thereby identifying a person, or the acquired sensing signal may be sent to an external computing device for calculation. In either case, when the sensing signal indicates that someone has entered, an alarm and braking device can be triggered with the appropriate alarm or braking operation. In one embodiment, intrusion can be detected by head and shoulder recognition; in another embodiment, intrusion by motion detection can be used to detect intrusion by a person or other object. Furthermore, since the first and second depth cameras need to operate simultaneously, the invisible light projected by each depth camera needs to have different wavelengths, for example, different wavelengths of infrared light.

[0050] The industrial robotic arm according to this utility model has been described in detail above with reference to the accompanying drawings. This disclosure provides an additional wide-angle "safety eye" to the Eye-in-Hand 3D vision system in addition to the "working eye," enabling the identification of personnel who have mistakenly entered the robotic arm's working area over a wider range and with greater timeliness and accuracy, and triggering alarms and braking, thereby achieving reliable safety protection throughout the entire working area.

[0051] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. An industrial robot arm, characterized by, include: A transmission arm is used to realize the shape changes of the industrial robotic arm; An end effector is used to enable the industrial robotic arm to grasp and / or perform command operations on a target object; A working visual feedback unit arranged near the end effector is used to acquire environmental feedback in the working area and includes a first depth camera for acquiring depth data within a first field of view. as well as A safety vision feedback unit, arranged on the industrial robotic arm, is used to acquire environmental feedback in the operating area and includes a second depth camera for acquiring depth data within a second field of view. The first depth camera includes a structured light depth camera, and the second depth camera includes a dToF depth camera.

2. The industrial robot arm of claim 1, wherein, Also includes: An alarm device connected to the safety visual feedback unit is used to issue an alarm in response to the second depth camera detecting the presence of a person in a first area of ​​the second field of view.

3. The industrial robot arm of claim 2, wherein, Also includes: A braking device connected to the safety vision feedback unit is used to brake the operating industrial robotic arm in response to the second depth camera detecting the presence of a person in a second area of ​​the second field of view.

4. The industrial robot arm of claim 3, wherein, The visual feedback unit is connected to the alarm device and the braking device, and includes: A personnel reporting device is used to report to the alarm device and / or the braking device in response to the first depth camera detecting the presence of a person within the first field of view.

5. The industrial robotic arm of claim 1, wherein, The second depth camera includes at least one of the following: Two dToF depth cameras are arranged on both sides of the structured light depth camera; A dToF depth camera is arranged outside the structured light depth camera; A dToF depth camera is arranged above the structured light depth camera; as well as A dToF depth camera is mounted on the base of the transmission arm.

6. The industrial robotic arm of claim 1, wherein, The first depth camera includes: Structured light projection device for projecting invisible structured light into a first field of view; and First and second image sensors having a predetermined relative spatial relationship are used to image the first field of view through which the structured light is projected to obtain first and second two-dimensional images.

7. The industrial robot arm of claim 6, wherein, The first depth camera also includes: A visible light image sensor is used to image the first field of view to obtain a visible light image when the structured light projection device does not project the structured light.

8. The industrial robotic arm of claim 1, wherein, The second depth camera includes: The light source module is used to project area array light into the second field of view; The dToF sensor generates a sensing signal characterizing distance based on the reception time of the returned light from the projected area array light.

9. The industrial robotic arm of claim 8, wherein, The dToF sensor includes: A planar array sensor composed of multiple avalanche photoelectric sensors is used, and The light source module includes: Light-emitting device for generating laser beam; and A diffuser is arranged in the propagation path of the laser beam to convert the laser beam into a wide-angle area array beam.

10. The industrial robotic arm of claim 1, wherein, The drive arm is a multi-axis drive arm, and the working visual feedback unit and / or the safety visual feedback unit are arranged on the end of the multi-axis drive arm near the end effector.

11. The industrial robotic arm of claim 1, wherein, Also includes: The end effector is connected to the transmission arm at a connection portion, and the work visual feedback portion and / or the safety visual feedback portion is arranged on the connection portion.

Images