Asad Zia
Published © GPL3+

Pre-Collision Assist with Pedestrian Detection - TensorFlow

Real-time hazard classification and tracking with TensorFlow. Sensor fusion with radar to filter for false positives.

IntermediateFull instructions provided8 hours7,028

Things used in this project

Hardware components

Walabot
Walabot
×1
Raspberry Pi 3 Model B
Raspberry Pi 3 Model B
×1
Android device
Android device
Samsung Galaxy A3 or better.
×1
Intocircuit 26000mAh High Capacity Power Castle
can use any way to power the RPIs on the move.
×1
Raspberry Pi USB Wireless Adapter
×2
Unicorn pHAT
To show alerts.
×1
Adafruit Raspbian Preinstalled Micro SD Card
×2
Raspberry Pi Zero
Raspberry Pi Zero
combined with pHAT to show alerts.
×1
USB-A to Micro-USB Cable
USB-A to Micro-USB Cable
×1
Android Honda In-vehicle infotainment (IVI)
MirrorLink and WiFi Access Point
×1

Software apps and online services

TensorFlow
TensorFlow
Android Studio
Android Studio
unicorn-hat
Mosquitto MQTT
MQTT broker for Raspberry Pi
paho-mqtt
MQTT library for Python on Raspberry Pi
org.eclipse.paho:org.eclipse.paho.client.mqttv3
MQTT library for Android
DarkFlow
Convert YOLO graph to be consumed by TensorFlow

Hand tools and fabrication machines

Soldering iron (generic)
Soldering iron (generic)

Story

Read more

Schematics

Architecture

System components and communication
Slide2 kykemg8mua

Walabot RADAR

Slide5 kwclgnuji1

Code

Walabot TensorFlow

Walabot script TensorFlow on Android., a native .so library and a Java JAR Pimoroni Unicorn pHAT scripts MQTT support

Credits

Asad Zia

Asad Zia

2 projects • 39 followers
Embedded Systems Engineer for Self-Driving Car

Comments

Add projectSign up / Login