Frontiers in Computing and Intelligent Systems 
ISSN: 2832-6024 | Vol. 14, No. 1, 2025 

 

18 

Design	and	Implementation	of	an	Intelligent	Control	
System	for	3D	Printers	Based	on	Unity3D	
Yichen Zhou, Yang Xiang, Can Jian, Jiawei Jiang, Mingchang Zhang, Kunpeng Li, Tianhao Zhao, 
Yanze Jin  

Shanghai University of Engineering Science, Shanghai 201620, China 

 

Abstract: With the rapid development of additive manufacturing technology, the demand for high-precision control and real-
time monitoring of 3D printing systems has become increasingly urgent. To address issues such as step loss errors, insufficient 
real-time performance, and weak remote monitoring capabilities in traditional 3D printers, this paper proposes an intelligent 3D 
printing system that integrates digital twin technology with incremental PID closed-loop control. By constructing a virtual-
physical synchronized digital twin model and combining it with Unity3D for high-precision virtual simulation, the system 
achieves real-time mapping of the motion trajectory and state parameters of the physical printer. A Raspberry Pi is employed as 
the core controller, integrating the MKS SERVO42C closed-loop driver board and an incremental PID algorithm to resolve step 
loss issues during high-speed operation of stepper motors, thereby improving printing accuracy. A network communication 
architecture based on the HTTP protocol is designed to enable real-time data interaction among the Raspberry Pi, the main 
control board, and the virtual twin. Additionally, a MySQL database is utilized for storing and analyzing printing process data. 
The system significantly optimizes printing efficiency and quality. 

Keywords: Unity3D; Digital Twin; PID Closed-loop Control; 3D Printing System; Real-time Monitoring. 
 

1. Introduction 
With continuous advancements in additive manufacturing, 

3D printing has been increasingly adopted across multiple 
industries. Compared with conventional machining methods, 
this technology offers significant advantages, including mold-
free fabrication, design flexibility, and the capability to 
manufacture complex structures. In sectors such as 
manufacturing and healthcare, 3D printing can substantially 
reduce production cycles, lower costs, and enable product 
innovation, thereby broadening its application in design and 
development.[1][2] 

Although mature control system solutions for 3D printing 
have been developed abroad, domestic technological progress 
remains relatively lagging, with weaker adaptability and a 
lack of reliable control platforms. Consequently, enterprises 
often encounter challenges such as low efficiency and 
unstable print quality in practical operations. Enhancing the 
performance and reliability of domestic 3D printing control 
systems has thus become an urgent task. 

To this end, this paper proposes a Raspberry Pi–based 
digital factory control system, which leverages digital twin 
technology to enable real-time monitoring and simulation of 
the 3D printing process. The system aims to overcome spatial 
limitations, while simultaneously improving both printing 
efficiency and output quality.[4][7] 

2. Overall Design of Digital Twin 
3DPrinter 

2.1. System Architecture Design 
This 3D printer is developed based on the open-source 

Klipper system, using the MKS Monster8 V2.0 as the main 
control board, in combination with the MKS SERVO42C 
closed-loop stepper motor driver board to achieve closed-loop 
control and prevent print failures caused by step loss. A 

Raspberry Pi 4B running the Fluidd system integrates both the 
operating system and the web-based Klipper interface, 
enabling remote task management and real-time monitoring 
of printer status.[9] 

 
Figure 1. Architecture Diagram of the Digital Twin Control 

System 
 

To accurately reproduce the behavior of the physical printer, 
a digital twin was developed in Unity3D. The twin simulates 
key parameters such as nozzle position, movement speed, 
extrusion volume, nozzle temperature, bed temperature, and 
current print file status. Communication between the physical 
printer and the digital twin is achieved via the HTTP protocol, 
ensuring real-time synchronization. The Unity application is 



 

19 

developed in C#, which provides efficient development tools 
and robust debugging capabilities for simulation, ensuring 
both accurate reproduction of the printer’s behavior and a 
user-friendly interaction interface.[5] 

To manage and analyze printer state information, a MySQL 
database is employed. It stores process data, enables large-
scale data analysis, and contributes to mathematical modeling 
of the printer’s performance. SQL is used for backend 
operations. This integrated system architecture allows 
comprehensive monitoring and fault analysis of the printer, 
thereby supporting improvements in performance and 
reliability.[6] 

2.2. Hardware Selection and Configuration 
2.2.1. MKS Monster8 

The MKS Monster8, developed by Makerbase (MKS), is a 
high-performance mainboard designed for 3D printers. It 
executes print instructions, controls stepper motors, heaters, 
and sensors. Equipped with a 32-bit processor, it provides 
high computational and control capability, enabling faster and 
more precise motion control. The board also features 
extensive expansion interfaces, allowing connections with a 
wide range of peripherals, and supports multiple 3D printer 
firmware systems such as Klipper, Marlin, and Repetier. In 
addition, its flexible parameter configuration accommodates 
diverse printing requirements. 

 
Figure 2. MKS Monster8 Mainboard 

2.3. Software Dev & Implementation 
Proportional–Integral–Derivative (PID) control is a widely 

used closed-loop control algorithm in industrial automation, 
valued for its simplicity, stability, and efficiency. PID control 
calculates the deviation between the desired value and actual 
value, and dynamically adjusts outputs using proportional, 
integral, and derivative operations to achieve rapid and stable 
system responses. 

 
Where: 

k — sampling index, k=0,1,2,⋯;  
r(k) — the k-th setpoint; 
c(k) — the k-th actual output value; 
u(k) — the k-th output control quantity; 
e(k) — the k-th deviation;  
e(k−1) — the (k−1)-th deviation; 
KP — proportional coefficient;  
TI — integral time constant; 
TD — derivative time constant;  
T — sampling period. 

2.3.1. Position-type PID 
Principle 
Position-type PID directly calculates the absolute value of 

the control quantity, i.e., the control output u(k)calculated 
each time represents the control quantity at the current 
moment. 

Formula 
The basic expression of PID control is as follows: 

 
Where: 

u(k): Control output at the current moment; 
e(k): Error at the current moment (the difference between 

the setpoint r(k) and the actual value y(k)); 
KP: Proportional coefficient; 
KI: Integral coefficient; 
KD: Derivative coefficient; 
∑ ek

i=0 (i): Accumulation of errors (integral action); 
e(k)−e(k−1): Rate of error change (derivative action). 
1.The control quantity is an absolute value: Directly 

calculate the target control output (rather than the variation). 
2.Greatly affected by the integral term: Long-term 

operation may cause integral saturation due to error 
accumulation, leading to system overshoot or even oscillation. 

3.Suitable for systems without accumulated errors (e.g., 
systems requiring stable control of liquid level, temperature, 
etc.). 

2.3.2. Incremental PID 
Principle 
Incremental PID calculates the increment of the control 

quantity, i.e., the value Δu(k) calculated each time represents 
the variation relative to the previous control quantity. 

Formula 
Incremental PID is derived from the position-type PID 

formula to calculate the increment: 
 

 
 
Control Quantity Update: 

 
Where: 

Δu(k): Control increment; 
e(k): Current error; 
e(k−1), e(k−2): Errors of the previous two moments; 

Other parameters are the same as those of position-type 
PID. 

Characteristics 
The control quantity is an increment: Avoid sudden 

changes in control output and reduce system impact. 
Strong anti-interference ability: Since the output is an 

increment based on error change, error fluctuations do not 
cause cumulative effects — suitable for systems with large 



 

20 

external disturbances. 
Suitable for digital control systems (e.g., single-chip 

microcomputers or PLC control) to avoid excessive control 
output caused by calculation overflow. 

3. Digital Twin Construction in 
Unity3D 

3.1. Virtual Simulation of the Digital Printer 
The printer model was first created in SolidWorks and 

imported into Unity for simulation. The designed printer 
adopts a UM2 Core X-Y motion structure, whose behavior is 
reproduced in Unity using interpolation functions. 
Interpolation methods, including linear interpolation and 
cubic Bezier curves, are used to generate smooth motion paths 
between keyframes. Unity’s built-in Lerp and SmoothStep 
functions allow precise computation of nozzle trajectories. 

Printing paths originate from G-code generated by slicing 
software. G-code specifies motion trajectories, extrusion 
volumes, and thermal parameters. By parsing G-code, motion 
control points are extracted and interpolated to reconstruct 
nozzle trajectories. Additional simulation features include 
mesh filling and layer merging to reproduce deposition 
thickness and layer transitions, enabling a highly realistic 
virtual print process. 

3.2. User Interaction with the Digital Printer 
A comprehensive human–machine interaction framework 

was built using Unity’s C# scripts, event system, UI 
components, and I/O modules. 

3.2.1. Interaction Function Modules  
View Control:  
 

 MoveInput 
Vector3 move = Vector3.zero; 
 
        
        move.x = Input.GetAxis("Horizontal") * 

moveSpeedWithKeyBoard * Time.deltaTime; 
        move.y = Input.GetAxis("Vertical") * 

moveSpeedWithKeyBoard * Time.deltaTime; 
        move.z = Input.GetAxis("Sagittal") * 

moveSpeedWithKeyBoard * Time.deltaTime; 
 
 
         
        if (!MouseStatus.isPointerOnUI(canves)) 
        { 
             
            if (Math.Abs(Input.GetAxis("Mouse 

ScrollWheel")) >= 0.1) 
                move.z = Input.GetAxis("Mouse 

ScrollWheel") * moveSpeedWithScroll * Time.deltaTime; 
 
            if (Input.GetMouseButton(2)) 
            { 
                move.x = -Input.GetAxis("Mouse X") * 

moveSpeedWithMouse * Time.deltaTime; 
                move.y = -Input.GetAxis("Mouse Y") * 

moveSpeedWithMouse * Time.deltaTime; 
            } 
        } 
        return move; 
 

 
Keyboard inputs are read via Unity's Input.GetAxis() 

method (e.g., WASD keys control forward, backward, left, 
and right movement; mouse scroll wheel adjusts view zoom).      
Mouse drag rotation uses Transform.Rotate(), enabling 360 
degree observation of printer status. 

 
Print Control: 
Monitors mouse click events 

(Input.GetMouseButtonDown()) to trigger print job start, 
pause, or termination. Via UnityWebRequest, it sends HTTP 
commands to the Raspberry Pi to control the physical printer's 
behavior. 

 
    public Reset resetX; 

    public Reset resetY; 

    public Reset resetZ; 

    public GCodeIdentify gCodeIdentify; 

    public PrinterManager printerManager; 

 

 

    /// <summary> 

    /// </summary> 

    public void doReset() 

    { 

        Console.createMessage(3, " Start resetting "); 

 

        resetX.stop = resetY.stop = resetZ.stop = false; 

        Reset.isReset = 0; 

         

        StartCoroutine(resetX.axisReturn()); 

        StartCoroutine(resetY.axisReturn()); 

        StartCoroutine(resetZ.axisReturn()); 

    } 

 
 
Information Feedback:  
Key parameters such as printing progress, nozzle 

temperature, and layer height are displayed in real-time on the 
UI panel. Status information is dynamically updated via the 
TextMeshPro component, with support for multi-language 
switching. 

 
 
public TextMeshProUGUI statusText; 
    void UpdateStatus(string message) 
    { 
        statusText.text = " Current status:" + message; 
    } 
 

4. System Data Mgmt & Anal. 

4.1. Design & Impl. of Communication Module 
The system uses Fluidd and Moonraker as control and 

communication interfaces. Fluidd provides a lightweight web 
UI for managing print tasks, uploading G-code, and 
monitoring progress, while Moonraker is a Python-based web 
server that exposes APIs via HTTP, WebSocket, or MQTT. 

The Raspberry Pi communicates with the MKS Monster8 
V2.0 board via Unix domain sockets, ensuring efficient inter-
process communication. Between Unity and the Raspberry Pi, 
HTTP requests are used for state queries and control 
commands. 



 

21 

4.2. Data Storage and Analysis 
The backend is implemented using Go, while the frontend 

employs Vue, with MySQL as the database. Key parameters 
(time, height, speed, temperature, etc.) are stored in structured 
form. Analysis algorithms are applied to detect anomalies 
such as overheating, extrusion inconsistencies, or nozzle 
misalignments. 

5. Conclusion 
This study presents an intelligent 3D printing monitoring 

and control system built upon digital twin technology, 
achieving real-time synchronization and remote control of the 
virtual and physical printers. Techniques including 
incremental PID control, G-code parsing, and process data 
analysis enhanced printing precision, stability, and 
intelligence. 

The contributions of this work lie in digital twin modeling, 
intelligent control algorithms, and data-driven analysis, 
offering new perspectives for smart manufacturing and 
additive manufacturing. Future improvements should focus 
on adaptive PID control, intelligent data analytics, and 
integration with cloud computing and AI algorithms to further 
enhance system intelligence. 

Competing Interests 
The authors declare that they have no competing of interest. 

References 
[1] Zhang Jian. Research and Implementation of a Control System 

for Desktop FDM 3D Printers [D]. Jiangsu University of 
Science and Technology, 2019. 

[2] Liu Lugang, Wang Kun, Guo Haidong. Design and Fabrication 
of 3D Printers [M]. 2023. 

[3] Bai Yuhui. Mechanical Structure Design of 3D Printers. 
Electronics Technology, 2023. 

[4] Liu Yonggang, Yao Liquan. Application of Digital Twin–
Based Virtual Simulation in Remote Training for Intelligent 
Manufacturing. Journal of Hubei Open Vocational College, 
2023. 

[5] Liu Dawei, Zhang Guoquan, Li Zeshan, et al. Design and 
Application of a Unity3D-Based Virtual Simulation Platform 
for Engineering Graphics. Electromechanical Engineering 
Technology, 2022. 

[6] Tan Renming, Zhang Renjie, Jiang Tao. Design of a Stepper 
Motor Position Closed-Loop Control System Based on STM32. 
Journal of Heilongjiang Institute of Technology, 2023. 

[7] Lu Jianhua, Long Chaoping. Exploration of Raspberry Pi–
Based Embedded Course Teaching. Education Forum, 2019. 

[8] Lin Rui. Research and Design of an IoT-Based Industrial 
Monitoring Platform [D]. Yangtze University, 2021. 

[9] Wang Yichen. Comparison and Study of Incremental and 
Positional PID Algorithms. Industrial Control Computer, 2018. 

[10] Jin Tao, Ma Jie, Yang Lichuang. Design of a WiFi-Based 
Remote Control System for 3D Printers. Sensors and 
Microsystems, 2018.