KR20220089828A - Cooperation robot system - Google Patents

Abstract

The present invention provides a user interface unit and a recognition unit communicating with the user interface unit to recognize an action taken by the user when the user manipulates the user interface unit, a control unit communicating with the recognition unit, and a control unit communicating with the control unit. It relates to a cooperative robot system for performing a multi-step task including an operation unit of a robot, wherein the recognition unit recognizes an action taken by the user while the user manipulates the user interface unit based on a work model of the multi-step task, the The recognition unit recognizes the start and completion of manual and automatic operations, and the recognition unit instructs the control unit to perform the automatic operation by the operation unit of the robot based on the work model of the multi-step operation when the manual operation is completed. do.

Description

Cooperative robot system. {COOPERATION ROBOT SYSTEM}

The present invention relates to a cooperative robot system, and to a cooperative robot system that enables operation and management of an advanced smart factory using a human-machine collaboration package.

In the smart factory production site, efficient operation and management is possible when using a human-machine collaboration package. To this end, big data is generated through 4M1E (Man, Machine, Material, Method, Environment) analysis, and through this, automation and optimization through FOM (Factory Operation Management) data-based informatization and CPS (Cyber Physical System)-based process analysis It is necessary to achieve energy efficiency through FEM (Factory Energy Management)-based process analysis. Through such comprehensive management, it is possible to manage the convergence of each module included in the smart factory, thereby enabling the operation of an advanced MI-NPS (Meta Intelligent-New Production System) type smart factory.

For example, in Republic of Korea Patent Publication No. 10-1550740, productivity is improved through individual manpower management and production performance management at the production site, and production performance is managed based on the achievement rate and share to manage 4M (Man, Machine, Material, Method). ), a management system that can easily perform management is disclosed, and a digital factory is operated based on 4M data in this system.

However, in order to apply to the MI-NPS type smart factory advanced by human-machine collaboration with human-machine collaboration process and manufacturing process simulation analysis technology in the smart factory, these data need to be applied to the process in real time.

The present invention has been devised in view of the problems of the prior art as described above, and an object of the present invention is to provide a cooperative robot system capable of efficiently operating a cooperative robot for a human-machine cooperative process.

The cooperative robot system of the present invention for achieving the above object includes a user interface unit and a recognition unit communicating with the user interface unit to recognize an action taken by the user when the user manipulates the user interface unit, the recognition unit and A cooperative robot system for performing a multi-step operation including a control unit communicating with the control unit, and an operation unit of the robot controlled by communication with the control unit, wherein the recognition unit is based on a work model of the multi-step operation while the user manipulates the user interface unit Recognizes the action taken by the user, the recognition unit recognizes the start and completion of manual and automatic actions, and when the manual operation is completed, the recognition unit automatically operates the robot's operation unit based on the multi-step work model based on the work model. It is characterized in that it instructs the control unit to perform.

At this time, the work model of the multi-step operation is 4M1E (Man, Machine, Material, Method, Environment) analysis, FOM (Factory Operation Management) data based analysis, CPS (Cyber Physical System) based process analysis, and FEM (Factory Energy Management) It can be derived from the results of the based process analysis.

In addition, the recognition unit stores a work model of a multi-step operation through CPS (Cyber Physical System)-based process analysis, and the CPS-based process analysis receives data from a sensor provided in the operation unit of the robot, and the recognition unit It is possible to create analysis data by analyzing the data received from the sensor, and to update the working model of the multi-step task by simulating the analysis data.

When the cooperative robot system according to the present invention is applied, the efficient operation of the cooperative robot for the human-machine collaboration process is possible in the smart factory, thereby enabling the efficient operation of the smart factory.

1 is a conceptual diagram schematically showing a cooperative robot system of the present invention.
2 is a result of analyzing the equipment operation performance before (a) and after (b) applying the CPS-based process analysis.
3 is a result of analyzing the hourly production volume before (a) and after (b) applying the CPS-based process analysis.

Hereinafter, the present invention will be described in detail.

Smart factory is the application of digital technology to the production system of the manufacturing process, and includes the possibility of utilizing various advanced technologies in information utilization, automation, monitoring, sensing, modeling, and networking fields. In the process, connectivity that enables information exchange by connecting all objects and participants to each other, and collaboration between interconnected components sharing information and resources, and performing independent decision-making and collaboration based on this are important factors.

In the smart factory, the process model technology field can be divided into element technologies such as application application technology, platform technology, device/network technology, and manufacturing security technology.

The application technology is the top-level software system of the smart factory ICT solution, and it performs various manufacturing executions on platforms such as MES, ERP, PLM, and SCM. It is an intermediate software system that plays a role in delivering, analyzing the data collected by the device, and providing optimization information through modeling and virtual physics simulation. In addition, the device/network technology is the lowest hardware system of the smart factory ICT solution. It detects the location, environment and energy through smart sensors and recognizes the positions of workers and workpieces through robots. It consists of a system that can transmit data to the platform. The manufacturing security technology relates to information protection and industrial confidentiality protection technology and countermeasures that can safely protect various data, systems, manufacturing facilities, etc. for all fields from sensors to applications.

The present invention relates to a collaborative robot system capable of improving factory efficiency by improving the platform and device/network of the smart factory, and is schematically configured as shown in FIG. 1 .

That is, the cooperative robot system of the present invention includes a user interface unit, a recognition unit communicating with the user interface unit to recognize an action taken by the user when the user manipulates the user interface unit, a control unit communicating with the recognition unit, and the It is configured to be able to perform a multi-step operation by including an operation unit of the robot that is controlled by communication with the control unit.

In particular, the recognition unit recognizes an action taken by the user while the user manipulates the user interface unit based on the work model of the multi-step task, and the recognition unit recognizes start and completion of manual and automatic operations, the recognition When the manual operation is completed, the unit instructs the control unit to perform an automatic operation by the operation unit of the robot based on the work model of the multi-step operation.

The user interface unit may include a control device capable of controlling the robot, and the user may control each part of the robot to perform a specific operation. In addition, the user interface unit may be configured to include a display unit so that a user can observe the operating state of the robot and apply a change therethrough.

In addition, by configuring the display unit using a system using augmented reality (AR) or virtual reality (VR), the user may improve proficiency through the interface unit and work efficiency may be increased through collaboration with the robot. In this case, the display unit may select and set various images, and may overlap or adjust images.

In addition, the recognition unit may use a computer, a smart device, or other data processing and storage device. The recognition unit may include a program stored in a computer that performs recognition and other procedures, and may be operated through an application installed in a smartphone or tablet PC. In addition, a device located at a remote location connected through a network may be used, and in this case, the recognition unit may be driven by utilizing an Internet of Things (IoT) system. That is, since the user interface unit and the recognition unit are connected through communication, there is no need to be physically coupled.

In addition, the recognition unit may include a device using robot kinematics to recognize specific movements of the user. In this case, it is possible to recognize the action taken by the user while the user manipulates the user interface unit based on the multi-step task model. The operation of this recognition unit enables cooperative work of the entire robot system through mutual information exchange in the process of performing multi-function tasks.

In addition, the robot system including the user interface unit and the sensor may be configured to perform image processing by observing or recording the robot's movement and work environment in real time based on the user's motion and a signal received from the sensor. . Even in this case, an AR or VR system may be used to improve the user's information acquisition and response efficiency.

The recognition unit recognizes the start and completion of manual and automatic operations, and the recognition unit is configured to instruct the control unit to perform an automatic operation by the operation unit of the robot based on the work model of the multi-step operation when the manual operation is completed. .

That is, the recognition unit determines an operation including a manual operation, an automatic operation, and, if necessary, a semi-automatic operation by the robot system including a sensor under the control of the controller.

The manual operation is performed similarly or identically to the operation performed in a conventional remote control robot system. In addition, the automatic operation is performed entirely under the control of the robot, and there is no user's involvement in the operation.

That is, the user can move a part of the system including the sensor to a position where one step and subsequent steps are to be performed in each step of the multi-step operation, and in this case, the robot system is driven by manual operation.

In this case, the recognition unit may recognize that the step is a manual operation state performed by the user, and may instruct the automatic operation to be performed at the time when the manual operation is terminated based on the work model. For example, the recognition unit may transmit, through communication, an instruction for moving the robot arm back to a position where it can pick up and return a part in a state where the robot arm is moved by a manual operation.

In addition, the recognition unit may recognize that the user has moved a part of the robot system to a position for performing a specific step during a multi-step operation. In this case, the recognition unit may instruct the control unit to perform another step to assist the user based on the work model. For example, there is a case in which the user has to operate one robot arm to instruct the operation of the robot arm. In this case, the object may be maintained in a fixed state for the user by automatic operation.

The work model is a work model for performing the multi-step task of the present invention, which is 4M1E (Man, Machine, Material, Method, Environment) analysis, ABL (Actual task Based Learning) based analysis, FOM (Factory Operation Management) data based analysis , CPS (Cyber Physical System)-based process analysis, and FEM (Factory Energy Management)-based process analysis results are derived and constructed.

The 4M1E is a resource related to the production activity of the smart factory and includes the environment in addition to the 4M resources, such as a worker (Man), a machine facility (Machine), a material (Material), and a product production method (Method). This 4M1E can be used as a database, and it can be used to establish a production plan and predict production performance through pre-simulation according to work scenarios, resource management such as personnel, accounting, material and order management, and digital factory work. It can include both on-site management, which grasps the situation in real time and collects and manages work performance by resource, and the process of manufacturing a production site database using various data collected from the field. Therefore, the data collected through the 4M1E analysis enables modeling of various states predicted in the process using a collaborative robot, and through this, it is possible to establish a work model.

In addition, it is possible to revise and manage the work model by analyzing it based on FOM (Factory Operation Management) data. To this end, the cooperative robot system may separately include a FOM analysis unit. The FOM analysis unit generates a smart factory operation management analysis result including key performance indicator (KPI) or LOB analysis based on the big data module, which may be performed based on the work data of the cooperative robot system. In addition, the big data analysis data and the smart factory operation management analysis result may be provided to the customer company, and the work model may be modified by reflecting the feedback on the provided result.

In addition, CPS (Cyber Physical System)-based process analysis is a system for performing smart factory processes by interconnecting physical and virtual systems. In an embodiment, the CPS modifies sensor data such as temperature, humidity, pressure, and voltage from an external sensor, and receives and stores equipment data from various external equipment such as a robot, an actuator, a lathe, a grinder, and a cutting machine. Through this, it is possible to manage management commands including controls, settings, inquiries, etc. entered by the administrator and a work model based on them. For example, the priority of the management command may be determined, and the work model may be set to perform a role corresponding to the management command according to the priority.

To this end, the cooperative robot system may separately include a CPS analysis unit. The CPS analysis unit registers, manages, and modifies the process and attribute information of the smart factory, stores information for simulation based on data, and verifies the validity of the obtained data to change the priority of previously entered management commands or You can change the working model. In addition, for verification of the validity, deep learning through machine learning or an internal/external artificial intelligence module may be used based on a reference algorithm. In addition, when the CPS analysis unit performs the simulation, a simulation including a change in the production quantity of a product, a change in a product model, a bypass process in case of a specific process failure, etc. may be performed. Data processed by the CPS analyzer may be virtual data.

For the CPS analysis, the process of producing an air intake duct_ capable of reducing automobile exhaust is as follows as an example. First, 3D modeling is produced using Inventor for parts necessary for CPS analysis. Change the 3D modeled parts into a file in a format applicable to the CPS program, and use it to establish the CPS manufacturing process Next, arrange equipment and logistics according to the above process and create a layout. Simulation through these operations According to the above simulation, the task of learning the points necessary for the operation of the robot is performed.When the robot is operated based on this, the operation performance of the vibration welding robot and the operator compared to before the application of CPS-based process analysis as shown in Fig. 2. It can be seen that this average is significantly improved from 58.8% to 89.8% In addition, as a result of analyzing the output per hour (UPH), it was confirmed that the productivity improved by 43.3% compared to before the application of the CPS-based process analysis as shown in FIG. .

A result of performing the CPS-based process analysis is transmitted to the recognition unit, and the recognition unit may modify the work model of the multi-step job through the CPS-based process analysis and store it. In addition, the CPS-based process analysis is performed by receiving data from a sensor provided in the operating unit of the robot in the robot system. Thus, it is possible to perform process analysis by collecting them. In addition, in the process analysis, the recognition unit analyzes all data received from the sensor, equipment data, and data obtained from an external sensor, thereby creating analysis data. The analysis data may be simulated by the recognition unit, and if the size of the data is too large, the analysis data may be transmitted to an external computer to process and simulate the analysis data in the external computer. Such a simulation can be performed using deep learning through machine learning or an internal or external artificial intelligence module, and the work model of the multi-step task can be automatically updated through the use of such machine learning or artificial intelligence module. In addition, the updated working model may be additionally stored in the recognition unit.

In addition, it is possible to modify or manage the work model through FEM (Factory Energy Management)-based process analysis. The FEM is a factory energy management system that can manage the efficient use of energy, and makes it possible to efficiently manage factors affecting the production efficiency of the smart factory. To this end, the cooperative robot system may separately include an FEM analysis unit. The FEM analysis unit generates an energy management analysis result of a smart factory based on a big data module like the FOM, which may be generated based on the work data of the cooperative robot system. In addition, it is possible to modify the work model based on the big data analysis data and energy management analysis results.

In addition, the control unit is operated in communication with the recognition unit, and the control unit receives a signal from the recognition unit and controls the operation unit of the robot to perform a multi-step operation.

The present invention described above is not limited by the above-described embodiments, and various substitutions, modifications, and changes are possible within the scope without departing from the technical spirit of the present invention. it is clear to one

Claims (3)

A user interface unit and a recognition unit communicating with the user interface unit to recognize an action taken by the user when the user manipulates the user interface unit, a control unit communicating with the recognition unit, and operation of a robot controlled by communicating with the control unit A collaborative robot system for performing multi-step tasks including a unit,
The recognition unit recognizes an action taken by the user while the user manipulates the user interface unit based on the task model of the multi-step task,
The recognition unit recognizes the start and completion of manual and automatic operations,
The cooperative robot system, characterized in that the recognition unit instructs the control unit to perform an automatic operation of the operation unit of the robot based on the work model of the multi-step task when the manual operation is completed.
The method according to claim 1,
The work model of the multi-step operation is 4M1E (Man, Machine, Material, Method, Environment) analysis, FOM (Factory Operation Management) data based analysis, CPS (Cyber Physical System) based process analysis, and FEM (Factory Energy Management) based process A cooperative robot system, characterized in that it is derived from the results of the analysis.
The method according to claim 1,
The recognition unit stores the work model of multi-step work through CPS (Cyber Physical System)-based process analysis,
The CPS-based process analysis receives data from a sensor provided in the operating unit of the robot,
The recognition unit generates analysis data by analyzing the data received from the sensor, and by simulating the analysis data to update the work model of the multi-step task.

Images