AutoRedL - Autonomous Human Tracking Laser System

Project Overview

AutoRedL is an embedded AI system designed to automatically detect and track human targets using computer vision, with a laser pointer that continuously follows the detected person. The system is built around the Orange Pi Zero 3 single-board computer and features a two-degree-of-freedom gimbal for precise laser positioning.

System Architecture

Hardware Components

Category	Component	Quantity	Status
Core Controller	Orange Pi Zero 3 (1GB)	1	Active
Vision System	2-DOF Gimbal Kit with 100W CSI Camera	1	Active
Actuators	MG90S Metal Gear Servo Motors	2	Integrated
Targeting	650nm Red Laser Module (30mW)	1	Active
Circuit Base	Breadboard + Dupont Wires (Female-to-Female)	1 Set	Active
Power Supply	18650 Battery Holder (2-Cell Series, with Switch)	1	Active
Voltage Regulation	AMS1117-5V Voltage Regulator Module	1	Active
Energy Storage	18650 Rechargeable Lithium Batteries	2	Active
Main Power	5V/2A DC Power Adapter	1	Active
Storage	16GB+ TF (MicroSD) Card	1	Active
Debug Interface	USB-to-TTL Serial Module (CH340/CP2102)	1	Active

System Block Diagram

[18650 Batteries] → [AMS1117-5V] → [Power Distribution]
                                              ↓
[CSI Camera] → [Orange Pi Zero 3] → [Servo Control] → [2-DOF Gimbal]
     ↓                    ↓
[Vision Processing] → [Target Detection] → [Laser Control] → [Laser Module]

Technical Specifications

Hardware Specifications

• Processor: Orange Pi Zero 3 (H618 Quad-core ARM Cortex-A53)
• Memory: 1GB LPDDR3 RAM
• Camera: 100W CSI interface camera module
• Servo Motors: MG90S (4.8V-6V, 1.8kg·cm torque)
• Laser: 650nm red laser, 30mW output power
• Power: 7.4V (2×18650) with 5V regulation
• Operating System: Orange Pi OS (Linux-based)

Software Stack

• Operating System: Orange Pi OS
• Programming Language: Python 3.x
• Computer Vision: OpenCV 4.x
• Hardware Control: RPi.GPIO
• AI Framework: TensorFlow Lite / ONNX Runtime
• Target Detection: YOLOv5/v8 or MobileNet-SSD

Key Features

Real-time Human Detection

• Advanced computer vision algorithms for human body detection
• Optimized for embedded systems with limited computational resources
• Configurable detection sensitivity and tracking parameters

Precise Target Tracking

• Two-degree-of-freedom gimbal system for smooth tracking
• PID control algorithm for accurate positioning
• Real-time coordinate transformation and servo control

Laser Targeting System

• Synchronized laser pointer that follows detected targets
• Safety features including power control and beam management
• Adjustable laser intensity and targeting precision

Embedded AI Processing

• On-device inference for real-time performance
• Optimized neural network models for ARM architecture
• Low-latency processing pipeline

System Requirements

Hardware Requirements

• Orange Pi Zero 3 development board
• Compatible CSI camera module
• Two MG90S servo motors
• 650nm laser module with appropriate power rating
• Power management system (18650 batteries + voltage regulator)
• Breadboard and connecting wires

Software Requirements

• Orange Pi OS or compatible Linux distribution
• Python 3.7 or higher
• OpenCV 4.x
• Required Python packages (see requirements.txt)

Performance Metrics

Detection Performance

• Detection range: 2-10 meters (depending on lighting conditions)
• Detection accuracy: >90% in optimal conditions
• Processing latency: <100ms per frame
• Tracking smoothness: 30 FPS target

System Performance

• Power consumption: <5W total system power
• Operating temperature: -10°C to +60°C
• Battery life: 2-4 hours continuous operation
• Servo response time: <50ms for 90-degree movement

License

This project is open source. Please refer to the license file for detailed terms and conditions.