1. The utility model provides an industrial robot visual detection and keep away barrier system which characterized in that: the robot comprises a robot body (1), a control module (2), a walking module (3), a learning module (4), an image acquisition module (5), a communication module (6), an obstacle avoidance module (7) and a database (8);

the control module (2) is used for intelligently controlling the robot body (1) and all components in the system;

the walking module (3) is used for driving the robot body (1) to move in the working direction;

the image acquisition module (5) is used for image acquisition, processing and transmission in the system; the image acquisition module (5) comprises an image pre-importing unit (501), a monitoring image acquisition unit (502) and a binocular image acquisition unit (503);

the image pre-importing unit (501) is used for importing various related images of preset working environments into the robot body (1) in advance, analyzing obstacles in the images, and importing the analyzed data into a database (8) in the robot body (1); the monitoring image acquisition unit (502) is used for connecting a database (8) of the robot body (1) with monitoring camera equipment in the actual working environment of the robot body through a communication module (6), analyzing and processing monitoring camera images in the actual working environment of the robot body (1) in real time and transmitting the images to the database (8); the binocular image acquisition unit (503) is used for acquiring related images through a binocular camera installed on the eyes of the robot body (1), and storing and updating the related images into the database (8) in real time;

the learning module (4) is used for autonomous deep learning and practical application of the robot body (1); the learning module (4) comprises an obstacle pre-avoiding behavior unit (401), a monitoring image learning unit (402) and a binocular shooting learning unit (403);

the obstacle pre-avoidance behavior unit (401) is used for generating a simulated obstacle avoidance behavior learning mode for the quantitative obstacle according to data information imported by the image pre-import unit (501) in the database (8), and generating simulated obstacle avoidance behavior training for the robot body (1) aiming at the quantitative obstacle; the monitoring image learning unit (402) is used for processing according to data information imported by the monitoring image acquisition unit (502) in the database (8), and updating and upgrading the simulated obstacle avoidance behavior learning mode in real time; the binocular camera shooting learning unit (403) is used for analyzing and learning image data acquired by a binocular camera in the database (8), updating and upgrading a simulated obstacle avoidance behavior learning mode in real time;

the communication module (6) is used for transmitting, receiving and processing data signals among all components in the system;

the obstacle avoidance module (7) is used for analyzing and processing the information of the obstacles and the surrounding environment thereof collected by the sensors and the data collector in the system, generating an obstacle avoidance behavior task, sending the obstacle avoidance behavior task to the control module (2) through the communication module (6), and controlling the robot body (1) to avoid the obstacles;

the database (8) is used for storing relevant data and configuration parameters in the system.

2. The visual inspection and obstacle avoidance system for an industrial robot of claim 1, wherein: the obstacle avoidance module (7) comprises an obstacle avoidance task generating unit (701), an obstacle avoidance task self-checking unit (702) and an alarm unit (703); the obstacle avoidance task generating unit (701) is used for analyzing and processing relevant information and parameters in a database (8) in the system and generating an obstacle avoidance task of the robot body (1); the obstacle avoidance task self-checking unit (702) is used for self-checking the execution progress of the obstacle avoidance task generated by the obstacle avoidance task generating unit (701); the alarm unit (703) is used for carrying out alarm reminding on the unexecuted obstacle avoidance task after the obstacle avoidance task self-inspection by the obstacle avoidance task self-inspection unit (702).

3. The visual inspection and obstacle avoidance system for an industrial robot of claim 2, wherein: the alarm unit (703) comprises an audible and visual alarm and a remote alarm which are installed on the robot body (1), and the remote alarm is remotely connected with the obstacle avoidance task self-checking unit (702) through a communication module (6).

4. The visual inspection and obstacle avoidance system for an industrial robot of claim 1, wherein: the walking module (3) comprises a mobile execution unit (301), the mobile execution unit (301) is used for generating and executing walking tasks of the robot body (1), and a control end of the mobile execution unit (301) is connected with the control module (2).

5. The visual inspection and obstacle avoidance system for an industrial robot of claim 1, wherein: the mobile execution unit (301) comprises a mobile wheel arranged at the bottom of the robot body (1), and a servo driving assembly and an angle adjusting assembly are arranged on the mobile wheel.

6. The visual inspection and obstacle avoidance system for an industrial robot of claim 1, wherein: the binocular image acquisition unit (503) comprises a front behavior image library (5031) and an obstacle image library (5032), wherein the front behavior image library (5031) is used for acquiring obstacle avoidance behavior images of pedestrians in front of the robot body (1), so that the later learning module (4) can conveniently perform simulated obstacle avoidance deep learning on the robot body (1); the obstacle image library (5032) is used for acquiring and processing images of the surrounding environment in the walking task of the robot body (1), storing the images into the database (8) and providing data reference for the generation of the obstacle avoidance task.

Background

With the rapid development of electronic information products, industrial robots are not only the post industry of a country but also represent the innovation capability and modernization level of a country as a product of the era. As a high-end automatic product, the automatic production line integrates a plurality of disciplines such as mechanics, electronic informatics, automation, computer discipline, bionics and the like, and has the advantages of high production efficiency, good product quality, capability of continuously working in a severe environment and the like. The characteristics lead the industrial robot to not only free human from heavy physical labor, but also greatly improve the living conditions of people in various aspects of production and life, and the industrial robot is inevitably developed in the direction of intellectualization and densification in order to better serve the human. So far, the number of robots in China is 40 thousands, except for some occasions where the robots cannot replace human beings, the multi-robot and man-machine cooperation will be the mainstream trend of industrial robot development and application. However, whether multiple robots cooperate or man-machine cooperate, the precision of obstacle avoidance of the robots makes the robot field face a great challenge.

Disclosure of Invention

The invention aims to provide a visual detection and obstacle avoidance system for an industrial robot, which aims to solve the problems in the background technology.

In order to achieve the purpose, the invention provides the following technical scheme: a visual detection and obstacle avoidance system for an industrial robot comprises a robot body, a control module, a walking module, a learning module, an image acquisition module, a communication module, an obstacle avoidance module and a database;

the control module is used for intelligently controlling the robot body and each component in the system;

the walking module is used for driving the moving direction of the robot body during working;

the image acquisition module is used for acquiring, processing and transmitting images in the system; the image acquisition module comprises an image pre-importing unit, a monitoring image acquisition unit and a binocular image acquisition unit;

the image pre-importing unit is used for importing various related images of a preset working environment into the robot body in advance, analyzing obstacles in the images and importing the analyzed data into a database in the robot body; the monitoring image acquisition unit is used for connecting a database of the robot body with monitoring camera equipment in an actual working environment of the robot body through a communication module, analyzing and processing monitoring camera images in the actual working environment of the robot body in real time and transmitting the analyzed and processed monitoring camera images to the database; the binocular image acquisition unit is used for acquiring related images through a binocular camera arranged on the eyes of the robot body, and storing and updating the related images into the database in real time;

the learning module is used for autonomous deep learning and practical application of the robot body; the learning module comprises a pre-obstacle avoidance behavior unit, a monitoring image learning unit and a binocular shooting learning unit;

the obstacle pre-avoidance behavior unit is used for generating a simulated obstacle avoidance behavior learning mode for the quantitative obstacle according to the data information imported by the image pre-importing unit in the database and generating simulated obstacle avoidance behavior training for the robot body aiming at the quantitative obstacle; the monitoring image learning unit is used for processing according to data information imported by the monitoring image acquisition unit in the database, and updating and upgrading the simulated obstacle avoidance behavior learning mode in real time; the binocular camera shooting learning unit is used for analyzing and learning image data acquired by binocular cameras in the database, updating and upgrading a simulated obstacle avoidance behavior learning mode in real time;

the communication module is used for transmitting, receiving and processing data signals among all components in the system;

the obstacle avoidance module is used for analyzing and processing the information of the obstacles and the surrounding environment thereof collected by the sensors and the data collector in the system, generating an obstacle avoidance behavior task, sending the obstacle avoidance behavior task to the control module through the communication module, and controlling the robot body to perform obstacle avoidance;

the database is used for storing relevant data and configuration parameters in the system.

As a preferred scheme of the present invention, the obstacle avoidance module includes an obstacle avoidance task generating unit, an obstacle avoidance task self-checking unit, and an alarm unit; the obstacle avoidance task generating unit is used for analyzing and processing relevant information and parameters in a database in the system and generating an obstacle avoidance task of the robot body; the obstacle avoidance task self-checking unit is used for self-checking the execution progress of the obstacle avoidance task generated by the obstacle avoidance task generating unit; and the alarm unit is used for carrying out alarm reminding on the unexecuted obstacle avoidance task after the obstacle avoidance task self-checking unit carries out self-checking.

As a preferable scheme of the invention, the alarm unit comprises an audible and visual alarm and a remote alarm which are installed on the robot body, and the remote alarm is remotely connected with the obstacle avoidance task self-checking unit through a communication module.

As a preferable scheme of the present invention, the walking module includes a mobile execution unit, the mobile execution unit is used for generating and executing a walking task of the robot body, and a control end of the mobile execution unit is connected to the control module.

As a preferable scheme of the present invention, the movement performing unit includes a movement wheel installed at the bottom of the robot body, and the movement wheel is provided with a servo driving assembly and an angle adjusting assembly.

As a preferred scheme of the invention, the binocular image acquisition unit comprises a front behavior image library and an obstacle image library, wherein the front behavior image library is used for acquiring obstacle avoidance behavior images of pedestrians in front of the robot body, so that a later learning module can conveniently perform simulated obstacle avoidance deep learning on the robot body; the obstacle image library is used for collecting and processing images of the surrounding environment in the walking task of the robot body, storing the images into the database and providing data reference for the generation of the obstacle avoidance task.

Compared with the prior art, the invention has the beneficial effects that:

the visual detection and obstacle avoidance system for the industrial robot, provided by the invention, can not only provide an optimal obstacle avoidance behavior task for a robot body in work, but also carry out all-around learning and training on an obstacle avoidance behavior mode of the robot; before the robot is put into use, conventional obstacle avoidance behaviors can be trained for various working environments, and when the robot is put into use, the learning mode of the simulated obstacle avoidance behaviors of the robot can be timely upgraded and adjusted according to the actual specific working environment of the robot, so that the robot is advanced with time, and the sensitivity and the response speed of the obstacle avoidance behaviors are improved.

Drawings

FIG. 1 is a schematic view of the overall structure of the present invention;

FIG. 2 is a schematic diagram of an embodiment of an image capture module according to the present invention;

FIG. 3 is a schematic diagram of a learning module according to the present invention;

fig. 4 is a schematic structural diagram of an obstacle avoidance module according to the present invention.

In the figure: 1. a robot body; 2. a control module; 3. a walking module; 301. a mobile execution unit; 4. a learning module; 401. an obstacle avoidance behavior unit; 402. a monitoring image learning unit; 403. a binocular shooting learning unit; 5. an image acquisition module; 501. an image pre-import unit; 502. a monitoring image acquisition unit; 503. a binocular image acquisition unit; 5031. a front behavior image library; 5032. a bank of obstacle images; 6. a communication module; 7. an obstacle avoidance module; 701. an obstacle avoidance task generating unit; 702. an obstacle avoidance task self-checking unit; 703. an alarm unit; 8. a database.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In the description of the present invention, it should be noted that the terms "vertical", "upper", "lower", "horizontal", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.

In the description of the present invention, it should also be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly and may, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

Referring to fig. 1-4, the present invention provides a technical solution: a visual detection and obstacle avoidance system for an industrial robot comprises a robot body 1, a control module 2, a walking module 3, a learning module 4, an image acquisition module 5, a communication module 6, an obstacle avoidance module 7 and a database 8;

the control module 2 is used for intelligently controlling the robot body 1 and each component in the system;

the walking module 3 is used for driving the moving direction of the robot body 1 during working;

the image acquisition module 5 is used for image acquisition, processing and transmission in the system; the image acquisition module 5 comprises an image pre-importing unit 501, a monitoring image acquisition unit 502 and a binocular image acquisition unit 503;

the image pre-importing unit 501 is configured to pre-import various images related to a preset working environment into the robot body 1, analyze obstacles in the images, and import data after the analysis into the database 8 in the robot body 1; the monitoring image acquisition unit 502 is used for connecting the database 8 of the robot body 1 with monitoring camera equipment in the actual working environment thereof through the communication module 6, analyzing and processing the monitoring camera images in the actual working environment of the robot body 1 in real time and transmitting the images to the database 8; the binocular image acquisition unit 503 is used for acquiring related images through a binocular camera installed at the eyes of the robot body 1, and storing and updating the related images into the database 8 in real time;

the learning module 4 is used for autonomous deep learning and practical application of the robot body 1; the learning module 4 comprises a pre-obstacle avoidance behavior unit 401, a monitoring image learning unit 402 and a binocular shooting learning unit 403;

the obstacle pre-avoidance behavior unit 401 is configured to generate a simulated obstacle avoidance behavior learning mode for the quantitative obstacle according to the data information imported by the image pre-import unit 501 in the database 8, and generate simulated obstacle avoidance behavior training for the robot body 1 for the quantitative obstacle; the monitoring image learning unit 402 is configured to process data information imported by the monitoring image acquisition unit 502 in the database 8, and update and upgrade a simulated obstacle avoidance behavior learning mode in real time; the binocular camera learning unit 403 is used for analyzing and learning image data acquired by the binocular cameras in the database 8, updating and upgrading the simulated obstacle avoidance behavior learning mode in real time;

the communication module 6 is used for transmitting, receiving and processing data signals among all components in the system;

the obstacle avoidance module 7 is used for analyzing and processing the information of the obstacles and the surrounding environment thereof collected by the sensors and the data collector in the system, generating an obstacle avoidance behavior task, sending the obstacle avoidance behavior task to the control module 2 through the communication module 6, and controlling the robot body 1 to perform obstacle avoidance;

the database 8 is used to store relevant data and configuration parameters within the system.

Further, the obstacle avoidance module 7 includes an obstacle avoidance task generating unit 701, an obstacle avoidance task self-checking unit 702, and an alarm unit 703; the obstacle avoidance task generating unit 701 is used for analyzing and processing relevant information and parameters in the database 8 in the system and generating an obstacle avoidance task of the robot body 1; the obstacle avoidance task self-checking unit 702 is configured to perform self-checking on the execution progress of the obstacle avoidance task generated by the obstacle avoidance task generating unit 701; the alarm unit 703 is configured to perform alarm reminding on the unexecuted obstacle avoidance task after the obstacle avoidance task self-inspection by the obstacle avoidance task self-inspection unit 702.

Further, the alarm unit 703 comprises an audible and visual alarm and a remote alarm installed on the robot body 1, and the remote alarm is remotely connected with the obstacle avoidance task self-inspection unit 702 through a communication module 6.

Further, the walking module 3 includes a movement executing unit 301, the movement executing unit 301 is used for generating and executing a walking task of the robot body 1, and a control end of the movement executing unit 301 is connected with the control module 2.

Further, the mobile execution unit 301 includes a mobile wheel installed at the bottom of the robot body 1, and a servo driving assembly and an angle adjusting assembly are installed on the mobile wheel.

Further, the binocular image collecting unit 503 comprises a front behavior image library 5031 and an obstacle image library 5032, wherein the front behavior image library 5031 is used for collecting obstacle avoidance behavior images of pedestrians in front of the robot body 1, so that the later learning module 4 can perform simulated obstacle avoidance deep learning on the robot body 1 conveniently; the obstacle image library 5032 is configured to collect and process an image of the surrounding environment in the walking task of the robot body 1, store the image in the database 8, and provide a data reference for generating an obstacle avoidance task.

The working principle is as follows: the visual detection and obstacle avoidance system of the industrial robot intelligently controls a robot body 1 and each component in the system through a control module 2, drives the robot body 1 to move through a walking module 3, generates changes of direction and angle, guides, collects, processes and transmits images required by obstacle avoidance and learning training of the robot through an image acquisition module 5, performs autonomous deep learning, training and practical application of the robot body 1 according to related data acquired by the image acquisition module 5 through a learning module 4, and realizes data signal transmission, reception and processing among each component in the system through a communication module 6; the obstacle avoidance module 7 analyzes and processes the relevant information and configuration parameters transmitted to the database 8 by the sensors and the data acquisition units in the system, generates an obstacle avoidance behavior task, and transmits the obstacle avoidance behavior task to the control module 2 through the communication module 6, so that the robot body 1 is controlled to avoid obstacles.

It is worth noting that: the whole device realizes control to the device through the controller, and the controller is common equipment and belongs to the existing mature technology, and the electrical connection relation and the specific circuit structure of the controller are not repeated herein.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.