PPT_23A21D5nmskxmkdkdkkdkdxkkkkxj811.pptx

Jkjk

1. DESIGN & DEVELOPMENT OF AI DRIVEN AUTONOMOUS WHEEL CHAIR NAVIGATION FOR PEOPLE WITH DISABILITIES SWARNANDHRA COLLEGE OF ENGINEERING AND TECHNOLOGY DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING A Final PROJECT PRESENTATION ON by G VEERENDRA KUMAR M. Tech, II YEAR - 23A21D5811 Under the guidance of Mr. M Satyanarayana, M. Tech, (Ph.D.) Associate Professor, Dept. of CSE 1

2. Contents Abstract Problem Statement Literature Review Objective Methodology List of Components & Mechanism Circuit Diagram Results & Discussion Conclusion G Veerendra Kumar [23A1D5811] 1

3. Abstract The AI-Driven Autonomous Wheelchair with IoT-based Smart Navigation is designed to enhance mobility for individuals with disabilities. Traditional wheelchairs require physical effort or assistance, limiting user independence, while existing autonomous solutions face challenges in cost, adaptability, and real-time obstacle avoidance. Our Project integrates IoT to enable smart navigation, real-time obstacle detection, and adaptive path planning . In this wheelchair we use ultrasonic sensors for environmental awareness, supports joystick commands for efficient movement. IoT features enable remote monitoring, predictive maintenance, and enhanced safety . G Veerendra Kumar [23A1D5811] 2

4. Problem Statement Physically challenged individuals face mobility, accessibility, and social inclusion challenges globally. According to the WHO , over 1.3 billion people (about 16% of the world’s population) live with some form of disability, with many requiring assistive devices for daily activities . has a disability, with mobility impairments being among the most common. https://www.who.int/news-room/fact-sheets/detail/disability-and-health?utm The RPWD ( Rights of Persons with Disabilities Act-2016) aims to ensure accessibility, education, employment, and social inclusion for disabled individuals. Our AI-Driven Autonomous Wheelchair addresses key mobility challenges by providing smart navigation, obstacle detection, and adaptive control through Joystick input, and real-time monitoring . G Veerendra Kumar [23A1D5811] 3

5. Literature Review G Veerendra Kumar [23A1D5811] 5 SN Title Author Year Methods 1 An Autonomous Wheelchair with Health Monitoring System Based on Internet of Things L.  Hou , et al 2024 Nonlinear Stabilizing Controllers 2 AI-based smart and intelligent wheelchair Rahimunnisa , et al 2020 Gesture Based Control 3 An IoT-Based Smart Wheelchair with EEG Control and Vital Sign Monitoring R. Meligy , et al 2024 Health Tracking 4 Smart Wheelchair Controlled Through a Vision-Based Autonomous System U. Masud , et al 2024 Vision Based Control

6. Objective Challenges Faced by Disabled Individuals: Limited mobility assistance in traditional wheelchairs High cost of existing smart wheelchairs Lack of multi-control accessibility Objective: Develop a cost-effective, multi-mode autonomous wheelchair Enhance safety using ultrasonic sensors and AI navigation G Veerendra Kumar [23A1D5811] 4

7. Methodology Requirement Analysis Design Prototype Development Testing and Validation Deployment Optimization & Feedback Integration G Veerendra Kumar [23A1D5811] 7

8. Flow Chart

9. List of Components Component Specifications Qty MY1016Z2 EBIKE Motor 24V, 250W, 360 RPM 2 Arduino 8-bit microcontroller, 256 KB Flash, 54 digital I/O pins 1 BTS7960 Motor Driver Module 43A high-power motor driver module 2 DC-DC Buck Converter 24V to 5V, 5A, 25W 1 Ultrasonic Sensor (HC-SR04) Range 2cm–400cm, Accuracy ±3mm 4 Joystick Module PS2, Analog Output, X & Y Axes 1 Voice Recognition Module V3, Compatible with Arduino 1 Battery Rechargeable, 24V, 10Ah 1 Emergency Stop Switch LB16A-W1, Push Type, 16mm 1 Power Distribution Board For motor, sensor, and logic circuitry 1 Battery Management System (BMS) For 24V Li-Ion Battery, Overcharge/Discharge Protection 1 Toggle Switch DPDT or SPDT switch, 2-position 1 G Veerendra Kumar [23A1D5811] 9

10. Design Catia-V5 G Veerendra Kumar [23A1D5811] 10 Design of Wheel Chair Catia-V5

11. List of Components G Veerendra Kumar [23A1D5811] 11 250w E-Bike Motor 24v 10ah LIFO4 Arduino Mega BTS7960 Motor Driver DC-DC Buck Converter Voice Recognition

12. Components G Veerendra Kumar [23A1D5811] 12 Rotary Encoder Joystick Module Ultrasonic ( HC-SR04) PDB-XT60 Toggle Switch BMS

13. Work Photos Frame Designing and welding Figure.1: Frame Figure.2: Welding 13

14. Work Photos Figure.3 & 4: Making of Shafts G Veerendra Kumar [23A1D5811] 14

15. Work Photos Figure. 5 & 6: Motors & Shafts Assembly G Veerendra Kumar [23A1D5811] 15

16. Work Photos Figure.7: code Uploading and Connections G Veerendra Kumar [23A1D5811] 16

17. Circuit Diagram

18. Work Photos Figure.8: Assemble of wheelchair G Veerendra Kumar [23A1D5811] 17

19. Software Used Programming Environment: Arduino IDE Control Logic: Reads inputs from Joystick, and Ultrasonic Sensors Processes sensor data for autonomous navigation Switches between Manual, and AI modes AI Integration: Object detection using ultrasonic sensors AI-assisted path planning (future scope) G Veerendra Kumar [23A1D5811] 18

20. Mode Selection Mechanism Toggle Switch for Mode Selection: Position 1: Joystick Mode Position 2: Voice Control Mode Position 3: Autonomes Mode Emergency Stop Button for Safety G Veerendra Kumar [23A1D5811] 19

21. Results & Discussion G Veerendra Kumar [23A1D5811] 20 Motor Synchronization Accuracy PID control ensured smooth motor coordination Indoor Accuracy: 97.94% Outdoor Accuracy: 96.08% Slight drop in outdoor due to uneven terrain

22. Results & Discussion G Veerendra Kumar [23A1D5811] 20 Obstacle Avoidance Performance AI-driven detection tested across varied environments Indoor Success: 92.04% Outdoor Success: 90.22% Lower outdoor success due to dynamic and irregular environments

23. AI Collisions Rate Reduction DQN-based AI Navigation reduced collisions Indoor Reduction: 20.24% Outdoor Reduction: 20.66% Shows effective learning and path optimization by AI Results & Discussion 21

24. Conclusion The AI-Driven Autonomous Wheelchair enhances mobility for individuals with disabilities by integrating AI-based navigation, ultrasonic sensors for obstacle detection, and IoT for remote monitoring. Achieved strong performance in real-world tests: 97.94% Motor Synchronization Accuracy 92.04% Obstacle Avoidance Success ~20% Collision Rate Reduction 18 Km per Charge Power Efficiency It provides a cost-effective, safe, and user-friendly solution compared to traditional and high-cost smart wheelchairs. With multi-mode control (joystick & autonomous), real-time obstacle detection, and IoT connectivity, the wheelchair ensures greater independence and safety. G Veerendra Kumar [23A1D5811] 20

25. Future Scope Smarter Navigation – Improve AI for better path planning and obstacle avoidance. IoT & Remote Monitoring – Enable real-time tracking of wheelchair status and battery. GPS Integration – Add GPS tracking for outdoor navigation and emergency help. Better Battery Management – Optimize power usage and explore solar charging. Mobile App Control – Create an app for remote control and mode switching. Lightweight Design – Make the wheelchair lighter and easier to upgrade . G Veerendra Kumar [23A1D5811] 21

26. Thank You G Veerendra Kumar