Anti-Sway Control System for Overhead Mobility Systems

This repository contains the design, implementation, and testing of an Anti-Sway Control System developed as part of the Instrumentation and Control Systems course at Indian Institute of Technology Indore. The project focuses on stabilizing suspended payloads using a real-time PID control mechanism integrated with a custom-built H-belt drive system.

Complete Report for detailed methodology, implementation diagrams, and validation results.

Abstract

The project aims to reduce sway in suspended payloads (like crane systems) by applying corrective actions using a PID controller. The controller reads angular displacement via an MPU-6050 sensor and drives stepper motors to reposition the suspension point using an H-belt drive mechanism. The system is capable of reducing oscillations by up to 90%, improving safety and operational efficiency in suspended load systems.

---

Components Used

• Arduino Uno
• Raspberry Pi (for supervision/interface)
• MPU-6050 (6-axis IMU sensor)
• NEMA 17 Stepper Motors (x2)
• A4988 Stepper Motor Drivers (x2)
• H-Belt drive system (custom-built)
• 2-axis Joystick module

---

Control System

PID Controller Design

A classical PID control algorithm was implemented to minimize the deviation ($\theta$) of the payload from the vertical:

u(t) = Kp * e(t) + Ki * integral(e(t)) + Kd * derivative(e(t))

Each axis (roll and pitch) is controlled using separate PID loops. Parameters were tuned using the Ziegler-Nichols method followed by iterative refinement.

---

Implementation Details

1. Sensor and Hardware Integration

• The MPU-6050 sensor is rigidly mounted on the payload using a custom 3D-printed bracket to reduce measurement errors.
• Sensor data is transmitted to the Arduino Uno via I2C protocol for real-time control.

2. Angle Estimation using Complementary Filter

• The accelerometer provides a noisy but drift-free estimate.
• The gyroscope offers fast dynamic response but accumulates drift.
• A complementary filter is used to combine both, ensuring stable and accurate roll and pitch angles.

angle = alpha * (gyro_angle) + (1 - alpha) * (accel_angle);

3. PID Control Logic

• Independent PID loops are implemented for roll and pitch.
• Each controller adjusts the position of the suspension point to counteract the sway.

4. H-Belt Drive Mechanism

• The motors drive a continuous H-pattern belt that translates PID commands to X-Y movement of the suspension point.
• Stepper motors are operated with 1/16 microstepping, giving a high linear resolution (~6.25µm/microstep).

• Kinematics:

  Motor1 = X + Y
  Motor2 = X - Y
  

5. Joystick-Based Manual Override

• A joystick allows manual control of the suspension point.

• Useful for:

  • Initial positioning
  • Manual testing
  • Inducing controlled disturbances for PID evaluation

---

Experimental Validation

Test Setup

• Adjustable suspension cable (0.5–1.5 m)
• Payloads: 2–4 kg
• Initial disturbances: 5°, 10°, 15°
• Testing both open-loop and closed-loop behavior

Results

• Sway reduction up to 90% compared to uncontrolled motion
• Stable damping and smooth transition between manual and automatic control

---

Future Improvements

Control Algorithms

• Model Predictive Control (MPC)
• Adaptive and Learning-Based Controllers

Hardware Upgrades

• Industrial-grade IMUs
• Load cells for real-time payload feedback
• Larger drive area for heavy-duty operation

Software Enhancements

• ML-based trajectory optimization
• Real-time system identification

---

References

• Anand Control – EOT Crane Anti-Sway
• Ziegler–Nichols Method – ScienceDirect
• Springer Research Article

---

👥 Team Members

• Kaustuv Devmishra (Mechanical Engineering Dept., IIT Indore)
• Kshitij Shetty (Mechanical Engineering Dept., IIT Indore)
• Prachi Patil (Mechanical Engineering Dept., IIT Indore)
• Mihir Hemani (Mechanical Engineering Dept., IIT Indore)
• Jatin Joshi (Mechanical Engineering Dept., IIT Indore)
• Krishan Swami (Mechanical Engineering Dept., IIT Indore)