MOVR's Arduino Driver

Arduino driver for MOVR - a small-size autonomous vehicle

This Arduino sketch subscribes to a "drive command" topic. The messages received are used to control the vehicle's (front) steering motor and (rear) drive-motor. The sketch also creates a ROS publisher to publish odometry data generated from the wheel encoders (and IMU).

Table of Contents

• Dependencies
• How to use
• Development
• References

Dependencies

• rosserial_arduino
  • Arduino ROS extensions to run rosserial_clients.
• ackermann_msgs
  • ROS messages for vehicles using ackermann steering.
• PID Library
  • An easy to use Arduino PID library.

Installation

# If the ROS 'ackermann_msgs/AckermannDrive' has not been installed
$ sudo apt install ros-kinetic-ackermann-msgs

# Install ROS rosserial, rosserial-arduino, rosserial-python
$sudo apt install ros-kinetic-rosserial ros-kinetic-rosserial-arduino ros-kinetic-rosserial-python

# Generate Arduino ros_lib
http://wiki.ros.org/rosserial_arduino/Tutorials/Arduino%20IDE%20Setup

How to use

# STEP 1: Run roscore
$ roscore

# STEP 2: Run rosserial node
$ rosrun rosserial_python serial_node.py /dev/ttyACM1

# STEP 3: Run the node that will publish the drive commands
# For example the movr_teleop hello_movr.py
# Note that you must 'source' the movr_ws for the movr_teleop package to be visible
$ rosrun movr_teleop hello_mvr.py

Development

Robot odometry

Encoder

LPD3806 600BM G5 24C Rotary Encoder

Wire Mappings:

Wire Color     Connections   
RED            5 - 24V DC   
BLACK          Ground   
GREEN          A Phase   
WHITE          B Phase   

Spefications:

• 600 pulses/revolution for a single phase.
• Therefore, two-phase output leads to 24000
• Maximum mechanical speed: 5000 Revolutions / minute
• Response Frequence: 0 - 20KHz

Basic math

CALIBRATION_FACTOR (Distance Traveled Per Click)
= PI * Wheel Diameter / Number of Clicks per Revolution

DISTANCE TRAVELED BY A WHEEL
= Calibration Factor * Number of Clicks Traveled by Wheel

AVERAGE DISPLACEMENT BY BOTH WHEELS
= (Distance Traveled by Left Wheel + Distance Traveled by Right Wheel) / Wheel Baseline

# Wheel Baseline = The Distance Between the Left Wheel and Right Wheel Contact Point

References

• Brett Bearegard's PID Introduction
• Henry's Bench Digital to Analog Tutorial
• Tori's Electronics Lab Rotary Encoder Tutorial
• ROS Serial
• ROS Serial Arduino
• TTU Advanced Robotics Odometry Tutorial
• Correl Lab's Forward Kinematics of Car-Like Mechanisms
• Steven LaValle's Kinematics of a Simple Car
• ROS Q&A on Twist to Ackermann-type Conversion
• Ackermann Messages Defined by the ROS Ackermann Interest Group