Robot safety interaction method based on vision

Abstract

The invention discloses a vision-based robot safety interaction method, which relates to the technical field of image processing, and comprises the following steps: an illumination acquisition module acquires the positions of light spots and the size and intensity of halo on the surfaces of a head camera and an external camera; light spots on the surfaces of the head camera and the external camera are optimized and processed through a hardware adjusting module, and in the process of optimizing and processing the light spots on the surfaces of the head camera and the external camera, light is automatically supplemented in the shooting process of the head camera and the external camera through a software configuration optimization module; compared with a traditional visual robot, the visual robot, the head camera and the external camera have the advantages that the images shot by the visual robot, the head camera and the external camera are clear, light spot blind areas on the surfaces of the head camera and the external camera are avoided, the visual robot is accurate in judgment, normal operation of the visual robot is further guaranteed, and the visual robot is convenient to use. And the injury to the personnel and the robot is also prevented.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly relates to a vision-based robot safety interaction method. Background Art

[0002] A robot is a machine that can perform tasks such as operations or movements through programming and automatic control. Robotics has been widely applied in fields such as industry, medicine, agriculture, service industry, construction, and even military. Robots are divided into indoor and outdoor robots.

[0003] Chinese Patent Application No. CN202410588140.9 discloses a vision-based robot safety interaction method, including: acquiring a head camera image and an external camera image; extracting human key points according to the head camera image and the external camera image, and detecting the distance between the human body and the robot according to the human key points; detecting whether the distance between the human body and the robot is less than a safety distance; driving the robot to perform corresponding obstacle avoidance actions according to the detection result, and controlling the robot to maintain the safety distance and continue to work.

[0004] In the above technical solution, the robot is always controlled to maintain a safe distance from the human body and continue to work, improving the safety of robot operations. Vision robots can obtain environmental information through cameras and make intelligent judgments and decisions, which has received wide attention. However, existing vision robots still have some problems in practical applications. Especially under lighting conditions, light spots are likely to appear on the surfaces of the head camera and the external camera. The positions of these light spots will affect the captured images of the head camera and the external camera, thereby causing blind spots in the captured images, which in turn affects the misjudgment of the vision robot, and may cause harm to people, or damage to physical objects or the robot itself. Summary of the Invention

[0005] The main purpose of the present invention is to provide a vision-based robot safety interaction method to solve the problems raised in the above background.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] A vision-based robot safety interaction method includes the following steps:

[0008] S1: The light collection module collects the positions of light spots, the sizes and intensities of halos on the surfaces of the head camera and the external camera, and then the hardware adjustment module optimizes and processes the light spots on the surfaces of the head camera and the external camera.

[0009] S2: During the process of optimizing and processing the light spots on the surfaces of the head camera and the external camera, the software configuration optimization module automatically compensates for the light during the shooting process of the head camera and the external camera;

[0010] S3: Collect the image frames captured by the head camera and the external camera;

[0011] S4: The receiving module receives the image frames captured by the head camera and the external camera, and the image overlapping module overlaps and displays the image frames captured by the head camera and the external camera at the same time and at the same location;

[0012] S5: The comprehensive judgment module accurately judges the obstacles within the running path of the robot, and the safety distance module judges the safety distance between the obstacles and the running path of the robot.

[0013] Preferably, the light collection module collects the positions of the light spots, the sizes and intensities of the halos on the surfaces of the head camera and the external camera, and transmits them to the hardware adjustment module and the software configuration optimization module in the form of electrical signals.

[0014] Preferably, the hardware adjustment module includes a light blocking module and an execution module I. The light blocking module receives the electrical signal instructions sent by the hardware adjustment module, processes the electrical signals, and then transmits the electrical signals to the execution module I.

[0015] Preferably, the execution module I includes a baffle unit and a cleaning unit. The baffle unit receives the execution instructions sent by the execution module I, and the baffle unit works to block the light, reducing and eliminating the sizes and intensities of the halos on the surfaces of the head camera and the external camera. The cleaning unit receives the execution instructions sent by the execution module I and cleans the surfaces of the head camera and the external camera.

[0016] Preferably, the software configuration optimization module includes a light optimization module and an execution module II. The light optimization module receives the instructions sent by the software configuration optimization module. The light optimization module works to perform light compensation on the surfaces of the head camera and the external camera. The light optimization module processes the electrical signals and sends the signal instructions to the execution module II;

[0017] The light optimization module collects the brightness and darkness on the surfaces of the head camera and the external camera. The light optimization module calculates the difference based on the clarity of the images captured by the camera itself, and uses a fixed difference as the brightness difference threshold, which is set to 0.5. The light optimization module collects the brightness and darkness of the camera and calculates the difference. When the brightness and darkness difference > 0.5, the light optimization module fills in the light at the dark positions on the surfaces of the head camera and the external camera. When the brightness and darkness difference ≤ 0.5, the light optimization module does not fill in the light on the surfaces of the head camera and the external camera. The calculation formula for the brightness and darkness difference on the camera surface is as follows:

[0018]

[0019] N is the number of pixels in the image, and Y i is the luminance value of the i-th pixel, and Y avg is the average luminance value of the image, which can be calculated by summing the luminance values of all pixels and dividing by the number of pixels. The calculation formula is as follows:

[0020]

[0021] Preferably, the second execution module includes an anti-vertigo unit and a reverse fill light unit. The anti-vertigo unit and the reverse fill light unit are electrically connected to the second execution module, and the anti-vertigo unit and the reverse fill light unit simultaneously receive the signal instructions sent by the second execution module.

[0022] By adopting the above technical solution, the anti-vertigo unit effectively blocks the direct irradiation of strong light sources onto the lens, reduces the reflection and diffusion of light inside the lens. The anti-vertigo unit reduces the exposure value, reduces overexposure phenomena, balances the luminance difference between the bright and dark parts in the picture, and improves clarity. The reverse fill light unit fills light during the shooting process of the head camera and the external camera.

[0023] Preferably, the receiving module includes a processor, and the processor processes the pictures taken by the head camera and the external camera;

[0024] The processor includes an image overlapping module, a comprehensive judgment module, and a safety distance module. The image overlapping module overlaps and compares the pictures taken by the head camera and the external camera. The image overlapping module sends the signal instructions to the comprehensive judgment module. The comprehensive judgment module judges whether there are obstacles in the running path of the robot. The comprehensive judgment module sends the signal instructions to the safety distance module. The safety distance module judges the safety distance between the obstacle and the running path of the robot.

[0025] Preferably, the image overlapping module includes a same-time comparison unit and a same-location comparison unit. The same-time comparison unit aligns the pictures taken by the head camera and the external camera in the forward direction in chronological order, and trims the pictures taken at the same time point. The same-location comparison unit marks the trimmed pictures at the same location.

[0026] Preferably, the processing methods of the same-time comparison unit and the same-location comparison unit for the pictures taken by the head camera and the external camera are as follows:

[0027] Step 1: The same-time comparison unit aligns the pictures taken by the head camera and the external camera in the forward direction at the same time point;

[0028] Step 2: The same-time comparison unit arranges the picture images taken by the head camera and the external camera in the order of time points, and at the same time crops the picture images taken by the head camera and the external camera at the same time point, and arranges them in sequence;

[0029] Step 3: The same-location comparison unit then marks the two same-location pictures cropped at the same time point with the same number.

[0030] Preferably, the receiving module is used to receive the picture images taken by the head camera and the external camera, and send the information to the processor in the form of an electrical signal. After receiving the signal, the processor sends a signal instruction to the image overlapping module. The image overlapping module receives the signal and processes it, and sends an instruction signal to the same-time comparison unit and the same-location comparison unit. At this time, the same-time comparison unit and the same-location comparison unit process and compare the picture images taken by the head camera and the external camera. After the same-time comparison unit and the same-location comparison unit complete the processing, the processing results are sent to the comprehensive judgment module in the form of an electrical signal. The comprehensive judgment module receives the signals transmitted from the same-time comparison unit and the same-location comparison unit, and judges whether there are obstacles in the running track of the robot. After the comprehensive judgment module judges that there is an obstacle, it sends an electrical signal to the safety distance module, and the safety distance module judges whether the obstacle is within the safe running distance of the robot.

[0031] The present invention has the following beneficial effects:

[0032] 1. In the present invention, compared with traditional vision robots, the vision robot has clear picture images taken by the head camera and the external camera, avoiding the appearance of light spot blind areas on the surfaces of the head camera and the external camera, making the vision robot judge accurately, thus ensuring the normal operation of the vision robot and preventing harm to personnel and the robot itself.

[0033] 2. In the present invention, the same-time comparison unit aligns the picture images taken by the head camera and the external camera in a positive manner. The same-time comparison unit arranges the picture images taken by the head camera and the external camera in the order of time points, crops the picture images taken by the head camera and the external camera at the same time point into several pieces, and arranges them in sequence. The same-location comparison unit then marks the two same-location pictures cropped at the same time point with the same number, compares the pictures, and then the comprehensive judgment module judges whether there are obstacles, further improving the accuracy of the comprehensive judgment module in judging obstacles and further ensuring the normal operation of the vision robot. Description of the Drawings

[0034] Figure 1 It is a flowchart of a vision-based robot safety interaction method of the present invention;

[0035] Figure 2Flowchart of light collection and processing for a vision-based robot safety interaction method of the present invention;

[0036] Figure 3 Flowchart of processor processing for a vision-based robot safety interaction method of the present invention;

[0037] Figure 4 Flowchart of the image overlap module for a vision-based robot safety interaction method of the present invention;

[0038] Figure 5 Flowchart of safety distance judgment for a vision-based robot safety interaction method of the present invention. Detailed implementation manners

[0039] 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 present invention, rather than all the embodiments. Based on the embodiments of the present invention, 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.

[0040] Please refer to Figure 1 and Figure 5 , the present invention provides a technical solution, including the following steps:

[0041] S1: The light collection module collects the positions of light spots, the sizes and intensities of halos on the surfaces of the head camera and the external camera, and then the hardware adjustment module optimizes and processes the light spots on the surfaces of the head camera and the external camera;

[0042] S2: During the process of optimizing and processing the light spots on the surfaces of the head camera and the external camera, the software configuration optimization module automatically fills in light during the shooting process of the head camera and the external camera;

[0043] S3: Collect the image frames captured by the head camera and the external camera;

[0044] S4: The receiving module receives the image frames captured by the head camera and the external camera, and the image overlap module overlays and displays the image frames captured by the head camera and the external camera at the same time and in the same place;

[0045] S5: The comprehensive judgment module accurately judges the obstacles within the running path of the robot, and the safety distance module judges the safety distance between the obstacles and the running path of the robot.

[0046] Compared with traditional vision robots, the vision robot has clear images captured by the head camera and the external camera, avoiding the appearance of light spot blind areas on the surfaces of the head camera and the external camera, enabling the vision robot to make accurate judgments, thus ensuring the normal operation of the vision robot and preventing harm to personnel and the robot itself; through the simultaneous comparison unit, the images captured by the head camera and the external camera are aligned correctly. The simultaneous comparison unit arranges the image pictures captured by the head camera and the external camera in the order of time points, cuts the image pictures captured by the head camera and the external camera at the same time point into several pieces and arranges them in sequence. The same location comparison unit then marks the two same location pictures cut at the same time point with the same number, compares the pictures, and further determines whether there are obstacles through the comprehensive judgment module, improving the accuracy of the comprehensive judgment module in judging obstacles and further ensuring the normal operation of the vision robot.

[0047] The light collection module collects the light spot positions, the sizes and intensities of the halos on the surfaces of the head camera and the external camera, and transmits them to the hardware adjustment module and the software configuration optimization module in the form of electrical signals.

[0048] The hardware adjustment module includes a light shielding module and an execution module 1. The light shielding module receives the electrical signal instructions sent by the hardware adjustment module, processes the electrical signals, and then transmits the electrical signals to the execution module 1.

[0049] The execution module 1 includes a baffle unit and a cleaning unit. The baffle unit receives the execution instructions sent by the execution module 1, and the baffle unit works to block the light, reducing and eliminating the sizes and intensities of the halos on the surfaces of the head camera and the external camera. The cleaning unit receives the execution instructions sent by the execution module 1 and cleans the surfaces of the head camera and the external camera.

[0050] The software configuration optimization module includes a light optimization module and an execution module 2. The light optimization module receives the instructions sent by the software configuration optimization module, and the light optimization module works to perform light compensation on the surfaces of the head camera and the external camera. The light optimization module processes the electrical signals and sends the signal instructions to the execution module 2;

[0051] The light optimization module collects the brightness and darkness on the surfaces of the head camera and the external camera. The light optimization module calculates the difference based on the clarity of the images captured by the camera itself, uses a fixed difference as the brightness difference threshold, and sets this threshold to 0.5. The light optimization module collects the brightness and darkness of the camera for difference calculation. When the brightness and darkness difference > 0.5, the light optimization module fills light on the dark areas of the surfaces of the head camera and the external camera. When the brightness and darkness difference ≤ 0.5, the light optimization module does not fill light on the surfaces of the head camera and the external camera. The calculation formula for the brightness and darkness difference on the camera surface is as follows:

[0052]

[0053] N is the number of pixels in the image, and Y i is the luminance value of the i-th pixel, and Y avg is the average luminance value of the image, which can be calculated by summing up the luminance values of all pixels and dividing by the number of pixels. The calculation formula is as follows:

[0054]

[0055] Execution module two includes an anti-vertigo unit and an inverse fill light unit. The anti-vertigo unit and the inverse fill light unit are electrically connected to execution module two. The anti-vertigo unit and the inverse fill light unit simultaneously receive the signal instructions sent by execution module two. The anti-vertigo unit effectively blocks the direct irradiation of strong light sources onto the lens, reduces the reflection and diffusion of light inside the lens, the anti-vertigo unit reduces the exposure value and the overexposure phenomenon, and the anti-vertigo unit balances the luminance difference between the bright and dark parts in the picture and improves the clarity. The inverse fill light unit fills light during the shooting process of the head camera and the external camera.

[0056] The receiving module includes a processor, which is a conventional processor in the prior art. The processor processes the pictures taken by the head camera and the external camera;

[0057] The processor includes an image overlap module, a comprehensive judgment module, and a safety distance module. The image overlap module overlaps and compares the pictures taken by the head camera and the external camera, and the image overlap module sends the signal instructions to the comprehensive judgment module. The comprehensive judgment module judges whether there are obstacles in the running path of the robot, and the comprehensive judgment module sends the signal instructions to the safety distance module. The safety distance module judges the safety distance between the obstacle and the running path of the robot.

[0058] The image overlap module includes a same-time comparison unit and a same-location comparison unit. The same-time comparison unit aligns the pictures taken by the head camera and the external camera in the positive direction in chronological order, and crops the pictures taken at the same time point. The same-location comparison unit marks the same location for the cropped pictures taken.

[0059] The methods for the same-time comparison unit and the same-location comparison unit to process the pictures taken by the head camera and the external camera are as follows:

[0060] Step 1: The same-time comparison unit aligns the pictures taken by the head camera and the external camera in the positive direction at the same time point;

[0061] Step 2: The same-time comparison unit arranges the pictures taken by the head camera and the external camera in the order of time points, and at the same time crops the pictures taken by the head camera and the external camera at the same time point and arranges them in order;

[0062] Step 3: The same-location comparison unit then marks the two same-location images cropped at the same time point with the same number, and uses letters A, B, C, D... A1, B1, C1, D1... A2, B2, C2, D2, etc. for marking.

[0063] The receiving module is used to receive the images captured by the head camera and the external camera, and send the information to the processor as an electrical signal. After receiving the signal, the processor sends a signal instruction to the image overlapping module. The image overlapping module receives the signal and processes it, and sends an instruction signal to the same-time comparison unit and the same-location comparison unit. At this time, the same-time comparison unit and the same-location comparison unit process and compare the images captured by the head camera and the external camera. After the same-time comparison unit and the same-location comparison unit complete the processing, the processing results are sent to the comprehensive judgment module as an electrical signal. The comprehensive judgment module receives the signals transmitted by the same-time comparison unit and the same-location comparison unit, and judges whether there are obstacles in the running track of the robot. After the comprehensive judgment module judges that there is an obstacle, it sends an electrical signal to the safety distance module. The safety distance module measures the distance between the obstacle and the robot (when the safety distance module measures the distance, the robot is used as the coordinate origin to measure the distance between the obstacle and the robot. The safety distance module measures the safety distance by using the conventional infrared measurement in the existing technology. When the infrared measurement of the distance between the person and the robot is greater than the maximum travel distance of the robot's robotic arm, it is determined at this time that the person is within the safety distance range). After the image overlapping module scans the two marked images at the same time point and the same location, it sends a scanning signal to the comprehensive judgment module, and then the comprehensive judgment module judges whether there is an obstacle. The judgment method is that when the image overlapping module scans that there are obstacles in both pictures, at this time the comprehensive judgment module judges that there is an obstacle at this position. When the image overlapping module scans that there is an obstacle in one picture and no obstacle in the other picture, the comprehensive judgment module judges that there is an obstacle at this position. When the image overlapping module scans that there are no obstacles in both pictures, at this time the comprehensive judgment module judges that there is no obstacle at this position. The judgment table is as follows:

[0064]

[0065] In the present invention, a vision-based robot safety interaction method is provided. Under the irradiation of external sunlight or light, light spots appear on the surfaces of the head camera and the external camera. At this time, the light acquisition module collects the positions of the light spots, the sizes and intensities of the halos on the surfaces of the head camera and the external camera, and then sends the signals to both the hardware adjustment module and the software configuration optimization module simultaneously. The hardware adjustment module receives and processes the signals, sends the signals to the light blocking module, and then transmits the electrical signals to the first execution module. At this time, the first execution module controls the baffle unit to block the light, reducing and eliminating the sizes and intensities of the light spots on the surfaces of the head camera and the external camera. The first execution module controls the cleaning unit to clean the surfaces of the head camera and the external camera, ensuring the cleanliness of the surfaces of the head camera and the external camera. The software configuration optimization module receives and processes the signals, sends the signals to the light optimization module, and then transmits the electrical signals to the second execution module. The second execution module controls the anti-vertigo unit to work. The anti-vertigo unit can effectively block the direct irradiation of strong light sources onto the lens, reducing the reflection and diffusion of light inside the lens. The anti-vertigo unit also has the function of reducing the exposure value and reducing the overexposure phenomenon. The second execution module controls the inverse fill light unit to work. According to the intensity of the light, when the difference in brightness and darkness > 0.5, the light optimization module fills light at the dark positions on the surfaces of the head camera and the external camera. When the difference in brightness and darkness ≤ 0.5. The light optimization module does not supplement light to the surfaces of the head camera and the external camera. The inverse fill light unit automatically performs backlight compensation on the head camera and the external camera. The inverse fill light unit greatly reduces the subjective brightness difference between the background image and the main image, improving the visibility of the entire field of view. After the head camera and the external camera capture images, the receiving module receives the captured images of the head camera and the external camera and sends the information as an electrical signal to the processor. After receiving the signal, the processor sends a signal instruction to the image overlapping module. The image overlapping module receives the signal and processes it, and sends an instruction signal to the same-time comparison unit and the same-location comparison unit. At this time, the same-time comparison unit aligns the two captured images of the head camera and the external camera positively. The same-time comparison unit arranges the captured images of the head camera and the external camera in the order of time points, cuts the captured images of the head camera and the external camera at the same time point into several pieces, and arranges them in sequence. The same-location comparison unit then marks the two same-location images cut at the same time point with the same number. The image overlapping module scans and compares the images, and then the comprehensive judgment module determines whether there are obstacles. If there are no obstacles, the vision robot runs. When an obstacle is detected, at this time, the comprehensive judgment module sends a signal to the safety distance module. The safety distance module determines whether the obstacle is on the running path of the robot. When the safety distance module measures the distance, with the robot as the coordinate origin, it measures the distance between the obstacle and the robot. The safety distance module measures the safety distance using the conventional infrared measurement in the existing technology. When the distance measured by the infrared between the person and the robot is greater than the maximum travel distance of the robot's robotic arm, it is determined at this time that the person is within the safe distance range. If the obstacle meets the safety distance, the robot runs; otherwise, the vision robot stops running. Compared with traditional vision robots, for this vision robot, the captured images of the head camera and the external camera are clear, avoiding the appearance of light spot blind areas on the surfaces of the head camera and the external camera, making the judgment of the vision robot accurate, thus ensuring the normal operation of the vision robot and preventing harm to personnel and the robot itself; through the positive alignment of the images captured by the head camera and the external camera by the same-time comparison unit, the same-time comparison unit arranges the captured images of the head camera and the external camera in the order of time points, cuts the captured images of the head camera and the external camera at the same time point into several pieces, and arranges them in sequence. The same-location comparison unit then marks the two same-location images cut at the same time point with the same number. The image overlapping module scans and compares the images, and then the comprehensive judgment module determines whether there are obstacles, further improving the accuracy of the comprehensive judgment module in determining obstacles and further ensuring the normal operation of the vision robot.

[0066] It should be noted that, in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0067] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by.

Claims

1. A vision-based robot safety interaction method, characterized in that: The following steps are involved: S1: The illumination acquisition module collects the light spot position, halo size and intensity on the surface of the head camera and the external camera, and then optimizes and processes the light spots on the surface of the head camera and the external camera through the hardware adjustment module; S2: During the optimization and processing of the light spots on the surface of the head camera and the external camera, the software configuration optimization module is used to automatically fill in the light during the shooting process of the head camera and the external camera; S3: Collect images taken by the head camera and external cameras; S4: The receiving module receives the images taken by the head camera and the external camera, and the image overlapping module displays the images taken by the head camera and the external camera at the same time and place in an overlapping manner; S5: The comprehensive judgment module accurately determines the obstacles within the robot's running path, and the safety distance module determines the safety distance between the obstacles and the robot's running path.

2. The vision-based robot safe interaction method according to claim 1, characterized in that: The illumination acquisition module acquires the light spot positions, the size and the intensity of the halo on the surface of the head camera and the external camera, and transmits them to the hardware adjustment module and the software configuration optimization module in the form of electrical signals.

3. The vision-based robot safe interaction method according to claim 2, characterized in that: The hardware adjustment module includes a light shielding module and an execution module 1. The light shielding module receives the electrical signal instruction sent by the hardware adjustment module and processes the electrical signal, and then transmits the electrical signal to the execution module 1.

4. The vision-based robot safe interaction method according to claim 3, characterized in that: The execution module 1 includes a baffle unit and a cleaning unit. The baffle unit receives the execution instruction sent by the execution module 1. The baffle unit works to block the light, reduce and eliminate the size and intensity of the halo on the surface of the head camera and the external camera. The cleaning unit receives the execution instruction sent by the execution module 1 and cleans the surface of the head camera and the external camera.

5. The vision-based robot safe interaction method according to claim 1, characterized in that: The software configuration optimization module includes a light optimization module and an execution module 2, the light optimization module receives instructions sent by the software configuration optimization module, the light optimization module works to perform light compensation on the head camera and the external camera surface, the light optimization module processes electrical signals, and sends signal instructions to the execution module 2; The light optimization module collects the surface brightness and darkness of the head camera and the external camera. The light optimization module calculates the difference according to the clarity of the camera's own shooting, and uses a fixed difference as the brightness difference threshold, which is set to 0.

5. The light optimization module collects the camera brightness and darkness for difference calculation. When the brightness difference is greater than 0.5, the light optimization module fills in the light at the dark position of the head camera and the external camera surface. When the brightness difference is less than or equal to 0.5, the light optimization module does not fill in the light at the head camera and the external camera surface. The camera surface brightness and darkness difference calculation formula is as follows: N is the number of pixels in the image, Y i is the brightness value of the i-th pixel, Y avg is the average brightness value of the image, which can be calculated by summing the brightness values ​​of all pixels and dividing by the number of pixels. The calculation formula is as follows:

6. A vision-based robot safe interaction method according to claim 5, characterized in that: The execution module 2 includes an anti-dizziness unit and a reverse fill light unit, the anti-dizziness unit and the reverse fill light unit are electrically connected to the execution module 2, and the anti-dizziness unit and the reverse fill light unit simultaneously receive signal instructions sent by the execution module 2.

7. The vision-based robot safe interaction method according to claim 1, characterized in that: The receiving module includes a processor, and the processor processes the images captured by the head camera and the external camera; The processor includes an image overlapping module, a comprehensive judgment module and a safety distance module. The image overlapping module overlaps and compares the images taken by the head camera and the external camera. The image overlapping module sends a signal instruction to the comprehensive judgment module. The comprehensive judgment module judges whether there is an obstacle in the robot's running path. The comprehensive judgment module sends a signal instruction to the safety distance module. The safety distance module judges the safety distance between the obstacle and the robot's running path.

8. The vision-based robot safe interaction method according to claim 7, characterized in that: The image overlapping module includes a same-time comparison unit and a same-location comparison unit. The same-time comparison unit aligns the images taken by the head camera and the external camera in time sequence and crops the images taken at the same time point. The same-location comparison unit marks the cropped images at the same location.

9. A vision-based robot safe interaction method according to claim 8, characterized in that: The same-time comparison unit and the same-location comparison unit process the images taken by the head camera and the external camera as follows: Step 1: The same-time comparison unit aligns the images captured by the head camera and the external camera at the same time point; Step 2: The same-time comparison unit arranges the pictures taken by the head camera and the external camera in sequence according to the time points, and at the same time crops the pictures taken by the head camera and the external camera at the same time points, and arranges them in sequence; Step 3: The same-location comparison unit marks the two same-location images cut at the same time point with the same number.

10. The vision-based robot safe interaction method according to claim 7, characterized in that: The receiving module is used to receive the images taken by the head camera and the external camera, and send the information to the processor in the form of an electrical signal. After receiving the signal, the processor sends a signal instruction to the image overlapping module. The image overlapping module receives the signal and processes it, and sends an instruction signal to the same-time comparison unit and the same-location comparison unit. At this time, the same-time comparison unit and the same-location comparison unit process and compare the images taken by the head camera and the external camera. After the same-time comparison unit and the same-location comparison unit complete the processing, the processing result is sent to the comprehensive judgment module in the form of an electrical signal. The comprehensive judgment module receives the signal from the same-time comparison unit and the same-location comparison unit to determine whether there is an obstacle in the robot's running trajectory. After the comprehensive judgment module determines that there is an obstacle, it sends an electrical signal to the safety distance module. The safety distance module determines whether the obstacle is within the robot's running safety distance.

Images