Table of contents

• 1. Learning Objectives
• 2. Use Case
• 3. Learning Activities
• 4. Technology
• 5. User Experience
• References

Generation of Robotic Cell Digital Shadow

Digital twins and shadows represent one of the key enablers for the digital transformation in manufacturing systems aligned with the Industry 4.0 principles [1]. Using digital shadows, manufacturing data are collected from real world systems and mirrored in their digital models through unidirectional data flow between real-world systems and their digital representation [1, 2, 3]. They are based on several enabling technologies [2, 4], among which Internet of Things (IoT) and Extended Reality (XR) can be singled out. In this respect, IoT represents a basis for data acquisition, whereas XR, through digital replicas of real-world objects, provides immersive experience to users.

Building digital shadows of manufacturing equipment requires the education of experts in this area. Considering the necessity of domain specific knowledge, it would be beneficial that future mechanical/manufacturing engineers, simultaneously with designing the manufacturing equipment, are capable to generate their digital shadows. This requires competencies in several fields, including modeling of equipment, creation and understanding of its control systems, i.e., underlying operational technologies, acquisition of data from the equipment using different communication protocols within IoT framework, and XR technologies. Whereas mechanical/manufacturing engineering higher education curricula readily cover topics related to generation of various digital models of manufacturing equipment and systems, as well as their control, the topics related to XR and IoT technologies, and generation of digital shadows are not frequently met.

This workflow represents support for teaching and learning of digital shadow creation and can be integrated into courses covering this topic. Through the presented workflow students develop a digital shadow of a robotic assembly cell. It is assumed that students are proficient in CAD modelling and programming of robots by teaching, whereas through this workflow the skills and competencies in XR and IoT based data acquisition are obtained. In particular, through this workflow students generate Virtual Reality (VR) representation of the robotic cell and the interface between robot controller and robotic-cell digital replica.

1. Learning Objectives

The objectives of workflow are to develop in students the following knowledge:

1. Factual Knowledge

   1. UDP-based communication in Client-Server configuration (connectionless datagram type socket);

   2. Recalling robot programming by teaching;

   3. Development of VR-based digital shadow of manufacturing systems.

2. Conceptual Knowledge

   1. Communication between digital shadows and controllers;

   2. Relation between robot and gripper motion and changes in robot controller registries containing encoder and digital output data;

   3. Relation between robot programmed motion and changes in robot controller registries;

3. Procedural Knowledge

1. Acquisition of data from controllers using UDP socket;

2. Procedure for creation manufacturing system digital shadow in VR;

3. Applying VR programming tools for unidirectional communication with a robot controller.

1. Metacognitive Knowledge

   1. Understanding the complexity of manufacturing systems digital shadows generation;

   2. Self-evaluate the understanding of methods for generation of robotic cell VR-based digital shadow.

The listed knowledge is related to the specific Intended Learning Outcomes (ILOs) that are listed in Table 1.

Table 1: ILOs with associated knowledge

2. Use Case

Using the workflow students develop a digital shadow of robotic cell (Figure 1) that carries out the last but one operation of assembly of a simple product, i.e. joining the Front Endcap to the remaining parts (Figure 2). The robotic cell (Figure 1) consists of an industrial robot Yaskawa Motoman SIA10F with seven degrees of freedom (1) equipped with pneumatic gripper Norgren M/160336/M/12 (2), assembly work holder (3), Front Endcaps magazine (4) and worktable (5). Gripper is controlled by electrically driven monostable 5/2 directional control valve that is activated through robot controller digital output.

The robotic cell is controlled using industrial robot controller FS100 [5] which is open and enables access to robot interfaces and motions using different communication protocols. Open functionality of the controller is achieved through special Yaskawa software solution - MotoPlus (Motoman Professional Programming Language for Superior Use), which represents a programming IDE (Integrated Development Environment) enabling applications developed on PC in C language to run as tasks in FS100 controller [6]. MotoPlus TCP/IP socket library, in particular “motoPlus.h” [7] is utilized to develop Ethernet-based communication capability of FS100. Using this library communication between the controller and other machines/devices is established in client/server model in which the client, through functions of TCP/IP socket library, requests processes from FS100 acting as server. Different processes are available, including retrieving position variable (mpGetPosVarData), retrieving the current position of robot axes in terms of encoder pulse count (mpGetPulsePos), retrieving the current servo speed in pulse counts (mpGetServoSpeed), retrieving digital I/O values (MpReadIO) [7].

Figure 1: The robotic cell whose digital shadow is created in the workflow

Figure 2: Product assembled within robotic cell in the workflow

Robotic cell operation is programmed by teaching using the programming pendant [8]. Within the workflow, the interface between VR-based digital shadow and robot controller is established using Ethernet-based UDP (User Datagram Protocol) communication. Although it is connectionless and does not assure reliability of communication, UDP protocol is used due to lower overhead than TCP (Transmission Control Protocol) and, thus, better suitability for real-time operation. During communication VR application acts as a client which obtains information from FS100 using unidirectional communication and requesting the following process through MotoPlus:

1. MpReadIO - retrieves several I/O values at a time. This process has the following arguments [7]:

   • sData - I/O address range that defines I/O type,

   • rData - I/O address in the selected range,

   • num - Number of I/O values that are retrieved;

2. MpGetFBPulsePos - retrieves the feedback position in pulse counts. This process has the following arguments [7]:

   • sData - the pointer to the data structure which specifies the control group (input),

   • rData - the pointer to the data structure which receives the feedback position coordinates in pulses (output).

These processes are utilized within the following functions implemented in robot controller:

void ReadIO(char* cmd, char* ret, BOOL bTCP)

{

int nType;

int nStartIndex;

int nQty;

int i;

LONG nRet;

MP_IO_INFO dxIOInfo[128];

USHORT retData[128];

nType = parseParameter(cmd, “a1”);

nStartIndex = parseParameter(cmd, “a2”);

nQty = parseParameter(cmd, “a3”);

for (i=0; i < nQty; i++)

{

dxIOInfo[i].ulAddr = GetIOAddress(nStartIndex + i, nType);

}

nRet = mpReadIO(dxIOInfo, retData, nQty);

if (nRet == OK)

{

outputAndLog(“OK”, bTCP);

for (i=0; i < nQty; i++)

sprintf( ret, “%s\r\n%d= %d”, ret, dxIOInfo[i].ulAddr, retData[i]);

}

else

sprintf(ret, “Failed to read I/O data”);

}

and

void GetFBPulsePos(char* cmd, char* ret, BOOL bTCP)

{

LONG nRet;

MP_CTRL_GRP_SEND_DATA dxSendData;

MP_FB_PULSE_POS_RSP_DATA dxRet;

dxSendData.sCtrlGrp = parseParameter(cmd, “a1”);

nRet = mpGetFBPulsePos(&dxSendData, &dxRet);

if (nRet == OK)

{

sprintf(ret, “%d;%d;%d;%d;%d;%d;%d;%d;\r”,

dxRet.lPos[0],

dxRet.lPos[1],

dxRet.lPos[2],

dxRet.lPos[3],

dxRet.lPos[4],

dxRet.lPos[5],

dxRet.lPos[6],

dxRet.lPos[7]);

}

else

sprintf(ret, “Failed to get feedback pulse position”);

}

These functions are invoked by the client using TCP/IP socket library. The code for calling function ReadIO is 2, and for calling function GetFBPulsePos is 17.

3. Learning Activities

Workflow consists of six learning tasks through which students acquire intended knowledge and learning outcomes. The sequence of tasks along with their inputs and outputs is presented in Figure 3, whereas Table 2 contains the description of tasks and ILOs they contribute to. Task T2.0 is performed in CAD modelling software, task T3.0 in VR only, task T4.0 on the real-world robotic cell, whereas tasks T5.0 and T6.0 require both - the real-world robotic cell and its VR representation.

Figure 3: Learning Activities

Table 2: Workflow for the generation of Robotic cell Digital Shadow

4. Technology

During the development of the workflow the following technologies are utilized:

• Robotic assembly cell based on SIA10F robot controlled by FS100 robot controller;

• MotoPlus Yaskawa software solution;

• Unity3D v6000.0.3f1 as a Virtual Reality (VR) development platform;

• SolidWorks for the generation of work cell 3D CAD model – the model was exported to neutral glTF format, which is lightweight, suitable for real-time rendering and keeps information regarding textures and reflective properties of surfaces assigned in 3D model;

• C# for the development of the following functionalities:

  • Acquisition of pose and digital output data from robot controller using UDP;

  • Movement of robotic work cell elements;

• Oculus Rift S Virtual Reality device including headset with cameras and audio devices, as well as two hand controllers for providing students with immersive and interactive experience (Figure 4).

Figure 4: VR device – Oculus Rift S

5. User Experience

Using the presented workflow students can develop competencies in digital shadow generation in an attractive way. In addition to mastering digital shadow generation, they simultaneously develop skills in IoT and VR as cutting-edge technologies. Furthermore, they recall the programming of robots by teaching and this workflow represents an excellent addition to learning robot programming.

The workflow enhances students’ understanding of the subject and contributes to the following learning outcomes:

1. Generation of manufacturing systems VR representation

Using the presented workflow students master the creation of manufacturing systems VR representation. The simultaneous motion of real-world robotic cell and its VR-based replica significantly improves the students’ understanding of relation between motion of elements in VR and their real-world counterparts. Namely, through the development of digital shadow students can directly correlate the commanded motion in robot controller with the commanded motion in VR.

2. Deeper understanding of IoT in industrial applications

Within the workflow students utilize industrial internet of things to connect the PC running the VR-based digital shadow and real-world robot controller. They gain indispensable hands-on experience in establishing communication through UDP/IP protocol. In addition to mastering this protocol, students get experience with service-based client/server architecture that is frequently met in industrial applications and understand better the underling theoretical concepts. Finaly, they obtain skills related to preparing and parsing data streams.

3. Generation of manufacturing systems digital shadow

Through combination of VR and IoT tools, students get skilled in generation of manufacturing systems’ digital shadow. Although the workflow includes communication with only one controller, students can extrapolate this knowledge to multiple controllers through connection to different devices using the network layer of TCP/IP model.

4. Easier Self-Evaluation

Critical and self-critical thinking represent significant outcomes of engineering education. An important part of the workflow in this context represents the self-evaluation of the developed digital shadow - through immersion into VR students can test the correctness of the VR-based digital shadow and spot errors if they exist.

References

1. Kritzinger W, Karner M, Traar G, Henjes J, Sihn W, Digital Twin in manufacturing: A categorical literature review and classification, IFAC-PapersOnLine, Vol. 51, No. 11, pp. 1016-1022, 2018

2. Attaran M, Celik B. G, Digital Twin: Benefits, use cases, challenges, and opportunities, Decision Analytics Journal, Vol. 6, art. no. 100165, 2023

3. Brecher C, Dalibor M, Rumpe B, Schilling K, Wortmann A, An Ecosystem for Digital Shadows in Manufacturing, Procedia CIRP, Vol. 104, pp. 833-838, 2021

4. Javaid M, Haleem A, Suman R, Digital Twin applications toward Industry 4.0: A Review, Cognitive Robotics, Vol. 3, pp. 71-92, 2023

5. Yaskawa, MOTOMAN FS100 Industrial Robot Controller, YR-FS100, A-03-2017, A-No. 157077, 2017

6. Yaskawa, FS100 Options Instructions - User’s Manual for New Language Environment MotoPlus, Manual No. HW1480837, 2012

7. Yaskawa, FS100 Options Instructions - Programmer’s Manual for New Language Environment MotoPlus, Manual No. HW1480839, 2012

8. Yaskawa, FS100 Operator’s Manual, Manual No. RE-CSO-A043, 2011