STM32F4 DC Motor PID Control System

Comprehensive implementation of a digital speed control system for DC motors using the STM32F407VGT6 microcontroller. The system features a robust PID algorithm, non-blocking UART/DMA communication, and a desktop interface for real-time monitoring and control.

---

Table of Contents

• Project Overview
• System Architecture
• Module 1: Hardware Design
• Module 2: Firmware & Resource Allocation
• Module 3: Control Algorithm (PID)
• Module 4: PC Interface & Communication
• Results & Performance
• Authors & Acknowledgments

---

Project Overview

This repository contains the source code and documentation for the EE3067 Embedded Control Systems assignment. The project addresses the challenge of precise motor speed regulation under varying loads by implementing a digital feedback loop.

Key Features:

• Closed-Loop Control: Real-time speed adjustment using an Incremental Encoder.
• High-Performance Communication: Utilizes DMA (Direct Memory Access) to handle data streams without CPU intervention, ensuring the PID loop remains strictly timed.
• Dynamic Tuning: Modify $K_p, K_i, K_d$ parameters on-the-fly via the PC GUI.
• Safety Mechanisms: Integrated "Connection Lost" detection for both TX and RX lines.

---

System Architecture

The system operates on a dual-power supply model (5V USB for logic, 12V Battery for power) and is divided into three main interaction blocks:

Block	Component	Function
Control Unit	STM32F407G-DISC1	Central processing, PID calculation, and PWM generation.
Power Stage	L298N H-Bridge	Modulates 12V power based on PWM signals to drive the motor.
Feedback	Incremental Encoder	Provides quadrature pulse signals (Phase A/B) to measure RPM.

---

Module 1: Hardware Design

The hardware setup ensures signal integrity and component isolation:

• Signal Buffering: Uses the 74HC245 buffer to shift logic levels (3.3V $\to$ 5V) and protect the STM32 GPIOs.
• Noise Filtering: RC Low-pass filters are applied to Encoder lines to eliminate high-frequency glitches caused by motor brushes.
• Power Management: Dedicated AMS1117-5.0 regulator provides a stable reference voltage.

---

Module 2: Firmware & Resource Allocation

The STM32F407 resources are optimized to support multi-tasking (Motor Control + Data Transmission) without latency.

Pin Mapping & Timer Configuration

Resource	Pin	Function	Configuration
TIM1	PE9	PWM Generation	Controls motor speed via Duty Cycle.
TIM3	PC6, PC7	Encoder Mode	Reads Phase A/B to calculate rotation.
TIM2	Internal	PID Loop	Interrupt timer set to $T_s = 10ms$.
USART2	PA2, PA3	Communication	115200 bps, linked with DMA1.

---

Module 3: Control Algorithm (PID)

The core control logic is executed inside the TIM2 Interrupt Service Routine every 10ms.

Discrete PID Formula: $$u(t) = K_p e(t) + K_i \int_{0}^{t} e(\tau)d\tau + K_d \frac{de(t)}{dt}$$

Operational Flow:

1. Read Encoder: Calculate current RPM based on pulse delta.
2. Compute Error: $e(t) = \text{Setpoint} - \text{Actual Speed}$.
3. Compute Output: Apply PID formula and saturate output (Anti-windup).
4. Actuate: Update TIM1 PWM register and L298N direction pins.

---

Module 4: PC Interface & Communication

Protocol Design

The system uses a Fixed-Length Frame (30 Bytes) architecture to maximize DMA efficiency.

• Circular Buffer: DMA automatically transfers incoming UART data to a RAM buffer, eliminating CPU overhead for individual byte reception.
• Command Structure:
  • M_STR: Start Motor with parameters ($K_p, K_i, K_d$, Speed).
  • M_STP: Stop Motor.
  • M_PLT: Plotting Data (Sent from STM32 to PC).

Desktop GUI

A graphical interface allows users to:

• Visualize Setpoint vs. Response in real-time.
• Monitor specific error values and connection status.
• Log debug data (Transmit/Receive packets).

---

Results & Performance

Testing with the DC Servo Motor yielded the following performance metrics:

• Response Time: The system reaches the setpoint in < 0.5 seconds.
• Steady State: Near-zero static error due to the Integral ($K_i$) term.
• Overshoot: Slight overshoot observed (~160 RPM for 150 Setpoint) but quickly dampened by the Derivative ($K_d$) term.

---

Authors & Acknowledgments

Group 6 - Embedded Control Systems (EE3067)

• Instructor: PhD. Nguyen Vinh Hao
• Institution: Ho Chi Minh City University of Technology (HCMUT)

Team Members:

No.	Full Name	Student ID	Assignment
1	Long Vo	2211910	Firmware
2	Uy Le Gia	2212438	Software
3	Vinh Nguyen	2213968	Hardware
4	Nhi Nguyen Ngoc Quynh	2213895	Compilation

---

December 2025