nuerrr's picture
Upload 2 files
060d171 verified
metadata
license: mit
task_categories:
  - robotics
  - time-series-forecasting
language:
  - en
tags:
  - robotics
  - embedded-systems
  - ultrasonic
  - autonomous-robot
  - PIC16F877A
pretty_name: Robot Sensor Telemetry Dataset
size_categories:
  - n<1K

Robot Sensor Telemetry Dataset

Description

This dataset contains real-time sensor telemetry collected from an autonomous robot built for CIE-349/408 Embedded Systems at Zewail City University of Science and Technology (Spring 2026).

The robot uses a PIC16F877A microcontroller reading 4x HC-SR04 ultrasonic sensors at 250ms intervals. Telemetry is streamed over UART to an ESP32 NodeMCU, forwarded via UDP to a Raspberry Pi 3B, and served to a Flutter mobile app over HTTP.

Dataset Structure

Column Type Description
timestamp_ms int Milliseconds since robot boot
front_cm int Front ultrasonic sensor distance (cm)
rear_cm int Rear ultrasonic sensor distance (cm)
left_cm int Left ultrasonic sensor distance (cm)
right_cm int Right ultrasonic sensor distance (cm)
motor_state string Current motor command: FORWARD/BACKWARD/TURN_LEFT/TURN_RIGHT/STOP/IDLE
face_verified bool Whether biometric face recognition lock is verified
tilt_warning bool MPU6050 tilt > 30 degrees detected
fall_detected bool MPU6050 tilt > 60 degrees detected

System Architecture

The robot operates as a 4-tier distributed pipeline:

  1. PIC16F877A — Hard real-time controller, reads sensors via Timer1, drives motors via PWM
  2. ESP32 NodeMCU — WiFi bridge, forwards UART telemetry to Raspberry Pi over UDP
  3. Raspberry Pi 3B — Flask server, face recognition security gate, UDP auto-discovery
  4. Flutter Android App — Mobile dashboard with D-pad control and radar display

Face Recognition Security

Motor commands are blocked unless a verified face is detected by the face_recognition library (dlib-based deep learning model) running on the Raspberry Pi. Face encodings are stored persistently in a .pkl file (biometric database).

Safety Features

  • Software obstacle cutoff: front sensor < 10cm → force STOP
  • Tilt/fall detection via MPU6050 accelerometer
  • Biometric lock: commands blocked until face verified

Team

  • Nouran Essam (202300627)
  • Menna Khaled (202301665)
  • Shrooq Amin (202300112)

Zewail City University of Science and Technology — CIE-349/408 Spring 2026