hector-slam-wreck

A robotic system designed to take advantage of already existing technologies to create a 2d/3d map of a room.

Modules

1. An obstacle-avoidance system written in Python that uses ultrasonic sensors to navigate and 4 DC motors through PWM for movement.
2. A lidar-laser module that collects 360 deg information regarding it's surrounding to later create a 2d/3d map using ROS.

Requirements

• HC-SR04 Ultrasonic Range Finder (3 of them)
• DFRobot 4WD Arduino-Compatible Platform with Encoders
• Pololu DRV8833
• https://www.thingiverse.com/thing:781918 (3D printed enclosures for ultrasonic sensors)
• LIDAR-Lite Laser Rangefinder v1
• Slip Ring with 22mm rim
• 12V, 350mA, 28oz-in NEMA-17 Bipolar Stepper Motor
• EasyDriver - Stepper Motor Driver
• http://www.thingiverse.com/thing:662646 (3d printed case for microcontroller and stepper motor for rotating the lidar laser)
• Raspberry Pi 2 Model B ARM Cortex-A7 Quad Core CPU 900MHz 1GB RAM
• Arduino Uno

General instruction to run the project

• Upload /utils/rotate-stepper.cpp to the Arduino Uno connected to the stepper motor in the 360 case
• Install Ubuntu ARM on a Raspberry Pi 2 Model B (or newer, similar hardware configuration)
• Install ROS Indigo distribution for Ubuntu ARM with desktop tools
• sudo apt-get install github
• git clone https://github.com/anthonymonori/hector-slam-wreck.git
• cp ./utils/gpioThing.sh /home/ubuntu/gpioThing.sh
• install lidarlite_scan package
• setup nodes
• run roscore
• start necessary nodes, see main.launch file
• rosrun lidarlite_laser laser_scan
• python /utils/obstacle-avoidance.py
• Using rviz, visualize the generated map.

For further description of the project and how to replicate it, please consult the paper.pdf report.

Contributors:

• utommo
• emilioamores
• 42f87d89
• anthonymonori