🚦 Red Light Jump Detection System 🚦

Top 6 – Bosch BOROSA Hackathon 2025

An onboard intelligent system designed to detect Red Signal Jump violations using vehicle-mounted sensors only, without relying on external traffic infrastructure, using XIAO ESP32.

---

📌 Problem Statement

Develop an intelligent system capable of identifying red signal jump violations through an onboard embedded module while minimizing false positives and handling real-world traffic edge cases.

---

🎯 Objectives

• Detect red signal jump events accurately
• Operate using onboard hardware components only
• Reduce false detections caused by partial crossings
• Handle complex traffic scenarios
• Generate structured violation and driver-behavior reports

---

⚙️ System Architecture

The system follows a multi-sensor fusion and state-based decision model.

Core Modules

1. Sensor Acquisition Layer

   • Collects motion, orientation, and proximity data

2. Signal State Evaluation

   • Determines current signal condition (Red / Yellow / Green)

3. Vehicle Motion Analysis

   • Tracks speed, acceleration, direction, and stop-line behavior

4. Violation Detection Engine

   • Applies rule-based and threshold logic
   • Eliminates false positives

5. Communication & Reporting

   • Sends violation and behavior data wirelessly
   • Generates long-term driver behavior flags

---

🧠 Detection Logic Overview

A violation is confirmed only when all conditions are satisfied:

• Signal state = RED
• Vehicle velocity > minimum movement threshold
• Vehicle crosses stop-line region
• Direction and orientation indicate forward motion
• No emergency override condition detected

This multi-condition validation ensures high precision and reliability.

---

🚨 Edge Cases Handled

• Low visibility signal scenarios
• Minor or partial stop-line crossings
• Yellow → Red transition ambiguity
• Vehicles halted beyond stop-line without forward motion
• Illegal turns at 4-way intersections
• Emergency vehicle movement filtering

---

⭐ Unique Feature (USP)

Agentic Driver Behavior Reporting System

Beyond real-time violation detection, the system continuously evaluates:

• Rash driving tendencies
• Driver responsiveness
• Reaction time patterns
• Alertness consistency
• Suspicion indicators of impaired driving

Based on cumulative sensor data, a Boolean risk flag is generated to support authority-level monitoring and intervention.

---

🔧 Hardware Components

• Arduino (Controller Unit)
• ESP32-S3 by Seeed Studio Extras:
• MPU6050 (Gyroscope + Accelerometer)
• IR Line Sensors
• LiDAR Modules
• LoRa RA-02 (SX1278)

---

💻 Software Stack

• Arduino Framework
• Embedded C/C++
• Edge Impulse
• State Machine Logic
• Sensor Fusion Algorithms
• Threshold-Based Decision System
• LoRa Communication Protocol

---

📡 Communication

• Long-range wireless transmission using ESP32-S3 and LoRa SX1278
• Supports low-power data transfer
• Enables centralized logging and analytics

---

📂 Repository Structure

/ ├── CameraWebServer ├── CameraWebServer_copy_20250410133304 ├── Final Backups/ │ └── Movement-Code-for-DUAL-MOTOR ├── Final_error_Code ├── Movement-Code-for-DUAL-MOTOR ├── debug-code-by-kaalu ├── libraries └── README.md

🧪 Testing Parameters

• Static signal testing
• Slow-roll partial crossings
• Sudden acceleration scenarios
• Stop-and-go traffic simulation
• Multi-direction turn validation

---

🚀 Future Enhancements

• Vision-based signal recognition integration
• Cloud-based violation dashboard
• Driver scoring model
• V2I (Vehicle-to-Infrastructure) integration
• Real-time authority alerting system

---

📜 License

This project is intended for educational, research, and hackathon demonstration purposes only.

---

👥 Team

• Rahul Raut (Team Lead)
• Siddhant Patil
• Ranjeet Wadkar

---