THANTHAI PERIYAR GOVERNMENT INSTITUTE
OF TECHNOLOGY VELLORE.
TEAM 14
SMART DRIVER ASSISTANCE SYSTEM
FOR TRANSPORTATION SECTOR
TEAM MEMBERS
EAM GUIDE: BALAJI P - 513121106013
. S. SATHISHBABU, M.E., M.B.A ., DHANUSH P - 513121106021
HARIHARAPRASAD R - 513121106
PROBLEM STATEMENT REFERENCE: SIH1535
• In the transportation sector, driver fatigue and
blind spots are significant causes of road
accidents, leading to severe injuries and fatalities.
• Traditional monitoring systems lack real-time
drowsiness detection and blind spot awareness,
making drivers vulnerable to potential collisions.
• The lack of an integrated system combining
drowsiness detection and blind spot monitoring
increases the risk of accidents, especially in high-
traffic or long-distance driving scenarios.
2
LITERATURE SURVEY
• The literature review reveals significant limitations in
current vehicle safety systems, including delayed
drowsiness detection, limited-range blind spot
monitoring, and lack of centralized fleet monitoring
capabilities.
• While advanced technologies like MediaPipe, YOLO,
IoT, GPS, and Spring Boot offer promising individual
solutions, existing research shows these are rarely
integrated .
• Our Smart Driver Assistance System (SDAS) directly
addresses these identified gaps by developing a real-
time, vision-based drowsiness detection system,
implementing YOLO-based blind spot monitoring, and
creating an IoT-enabled analytics platform for fleet
management.
3
EXISTING SYSTEM
• Existing solutions often rely on manual monitoring or basic alert systems, which
are not efficient in preventing accidents caused by driver inattention and fatigue.
• Conventional blind spot monitoring solutions rely heavily on basic proximity
sensors that have limited range and can miss fast-moving vehicles or objects in
adjacent lanes.
• Steering-based detection may not be reliable in scenarios like high-speed cruising
or adaptive cruise control, where minimal steering adjustments are needed, leading
to missed detections.
• Many existing solutions lack cloud-based data analytics and historical trend
analysis, preventing fleet operators or individuals from tracking long-term driver
behavior patterns and making data-driven safety improvements.
4
PROPOSED SYSTEM
INTEGRA
REAL- USER- MOBILE
BLIND TION FRIENDLY
TIME APPLICAT
DROWSI SPOT WITH INTERFA
MONITOR ION
NESS CE &
ING RASPBER ALERTS SUPPORT
DETECTI
ON RY PI
Utilizes The system GPS-Based Cross-platform
Deep learning
camera-based processes Vehicle compatibility
models to
object data using Tracking ensuring the
detect driver
detection to Raspberry Alerts through system works
fatigue based
identify Pi, ensuring buzzer, LED on various
on eye closure
vehicles real-time indicators, and Android
and facial
detection mobile devices
expressions.
notifications.
5
BLOCK DIAGRAM
6
CIRCUIT DIAGRAM
3
4
2 1. Raspberry Pi
2. BN-880 GPS
3. 16X2 LCD DISPLAY
4. BUZZER
5. CAMERA
7
HARDWARE SETUP
8
RESULTS
DROWSY STATUS ACTIVE STATUS
9
BLIND SPOT DETECTION OUTPUT SMS ALERT
10
10
DRIVERMONITOR WEB DRIVER BEHAVIOR
DASHBOARD ANALYTICS DASHBOARD
11
• https://sdas2025new.vercel.app/
11
MOBILE APPLICATION LIVE LOCATION
12
REFERENCES
• Deng, Wanghua, and Ruoxue Wu. “Real-Time Driver-
Drowsiness Detection System Using Facial Features.” IEEE
Access, vol. 7, pp. 118727–38. IEEE Xplore,
https://doi.org/10.1109/ACCESS.2019.2936663.
• Muzammel, Muhammad, et al. “Blind-Spot Collision
Detection System for Commercial Vehicles Using Multi Deep
CNN Architecture.” Sensors, vol. 22, no. 16, Jan. 2022, p.
6088. www.mdpi.com, https://doi.org/10.3390/s22166088.
• SIH1535 – Problem Statement
13
CONFERENCE ATTENDED
14
CONFERENCE ATTENDED
15
CONFERENCE ATTENDED
16
THANK YOU !
17
