Quanser QBot 2 User manual

CAPTIVATE. MOTIVATE. GRADUATE.
USER MANUAL
QBot 2 for QUARC
Set Up and Configuration

© 2015 Quanser Inc., All rights reserved.
Quanser Inc.
119 Spy Court
Markham, Ontario
L3R 5H6
Canada
[e-mail]
Phone: [phone]
Fax: [phone]
Printed in Markham, Ontario.
For more information on the solutions Quanser Inc. offers, please visit the web site at:
http://www.quanser.com
This document and the software described in it are provided subject to a license agreement. Neither the software nor this document may be
used or copied except as specified under the terms of that license agreement. Quanser Inc. grants the following rights: a) The right to reproduce
the work, to incorporate the work into one or more collections, and to reproduce the work as incorporated in the collections, b) to create and
reproduce adaptations provided reasonable steps are taken to clearly identify the changes that were made to the original work, c) to distribute
and publically perform the work including as incorporated in collections, and d) to distribute and publicly perform adaptations. The above rights
may be exercised in all media and formats whether now known or hereafter devised. These rights are granted subject to and limited by the
following restrictions: a) You may not exercise any of the rights granted to You in above in any manner that is primarily intended for or directed
toward commercial advantage or private monetary compensation, and b) You must keep intact all copyright notices for the Work and provide the
name Quanser Inc. for attribution. These restrictions may not be waved without express prior written permission of Quanser Inc.
QBOT 2 - User Manual 2

1 Introduction
The Quanser QBot 2 (Figure 1.1) is an innovative autonomous ground robot system incorporating a robust
educational ground vehicle with the Microsoft Kinect and a Quanserrembedded target. The QBot 2 is comprised
of a Yujin Robot Kobuki platform, a Microsoft Kinect RGB camera and depth sensor, and a QuanserrDAQ with a
wireless embedded computer (also refered to as the target computer). The embedded computer system mounted
on the vehicle uses the Gumstix DuoVero computer [1] to run QUARCr,Quanserr's real-time control software, and
interface with the QBot 2 data acquisition card (DAQ).
Figure 1.1: The Quanser QBot 2
The interface to the target computer is MatlabrSimulinkrwith QUARCr. The QBot 2 is accessible through three
different block sets: the QuanserrHardware-In-the-Loop (HIL) block set to read from sensors and/or write to
outputs, the QuanserrStream API blockset to perform communications over wired and wireless communication
channels, the QuanserrMultimedia blockset to read RGB and depth image data from the Kinect sensor, and The
MathWorksrComputer Vision System toolbox to perform image processing. Controllers are developed in Simulinkr
with QUARCron the host computer, and these models are cross-compiled and downloaded to the target (Gumstix
[1]) seamlessly. A diagram of this configuration is shown in Figure 1.2.
Figure 1.2: Communication Hierarchy
The general system description, component nomenclature, specifications, and model parameters are all given in
Section 2. Section 3 goes into detail on how to setup the QBot 2. Lastly, Section 4 contains a troubleshooting guide.
QBOT 2 - User Manual DRAFT - April 14, 2015

1.1 Prerequisites
To successfully operate the QBot 2, the prerequisites are:
1. To be familiar with the components of the QBot 2.
2. To have QUARCrversion 2.5 or later installed and properly licensed.
3. The MathWorksrComputer Vision System toolbox: http://www.mathworks.com/products/computer-vision/
4. Microsoft Kinect SDK v1.8 installed: http://www.microsoft.com/en-ca/download/details.aspx?id=40278
5. Microsoft Kinect for Windows Developer Toolkit v1.8 installed:
http://www.microsoft.com/en-ca/download/details.aspx?id=40276
6. To be familiar with using QUARCrcontrol and monitor the vehicle in real-time, and in designing a controller
through Simulinkr. See Reference [2] for more details.
1.2 References
[1] Gumstix: http://gumstix.com/
[2] QUARC User Manual (type doc quarc in Matlab to access)
QBOT 2 - User Manual 4

2 Components
The QBot 2 is made up of four main components: the Kobuki robot platform, the QBot 2 data acquisition card (DAQ),
the Gumstix DuoVero embedded computer, and the Kinect sensor. This section outlines these components in more
detail.
2.1 The Kobuki Robot Platform
The QBot 2 uses an Kobuki mobile robot platform (Figure 2.1). The QBot 2 follows the Quanserrstandard for body
frame axes, where the x-axis is in the forward direction, the y-axis is to the left, and the z-axis is up. The diameter
of the vehicle is 34 cm, and its height (without attachments) is 10 cm. The Kobuki is driven by two differential drive
wheels with built-in encoders. The Kobuki comes with a bumper sensor as well as a built-in gyroscope and cliff
sensors. The embedded computer target can access data from these sensors.
Figure 2.1: The Kobuki mobile robot platform
The QBot 2 is powered by a Lithium ion battery pack (Figure 2.2a) provided by Yujin Robot. The battery fits
underneath the QBot 2, and can last continuously for about 3hours after a full charge. The battery takes less
than 3hours to charge. While charging, the power light pulses slowly with an orange colour. A battery charger is
provided (Figure 2.2b). To recharge the QBot 2, plug in the battery charger and connect it to the charger input port
on the QBot 2 next to the ON/OFF switch (Figure 2.2c).
QBOT 2 - User Manual DRAFT - April 14, 2015

(a) The QBot 2 battery
(b) The QBot 2 charger (c) The QBot 2 charger input
Figure 2.2: The QBot 2 battery and charger input
2.2 System Specifications and Model Parameters
Table 2.1 lists the main parametrs associated with the QBot 2.
Symbol Description Value Unit
DDiameter of the QBot 2 0.35 m
dDistance between the left and right wheels 0.235 m
hHeight of the QBot 2 (with Kinect mounted) 0.27 m
νmax Maximum speed of the QBot 2 0.7m/s
mTotal mass of the QBot 2 3.79 kg
Table 2.1: QBot 2 specifications
2.3 The QBot 2 Data Acquisition Card (DAQ)
The QBot 2 DAQ contains the wiring and circuitry integrating the DuoVero embedded computer, Kobuki robot, Kinect
sensor, and additional input/output (I/O) components connected to the DAQ. Figure 2.3 shows the QBot 2 DAQ with
QBOT 2 - User Manual 6

accessible I/O headers for the user including SPI, I2C, UART, digital I/O, analog input, encoder input, and PWM
output. More details on accessing the QBot 2 I/O is found in 2.6. The QBot 2 DAQ connects to the Kinect sensor
via a USB port (see Figure 2.4). The QBot 2 DAQ connects to the Kobuki via a ribbon cable (see Figure 2.4). The
QBot 2 DAQ is powered via a cable connected to the Kobuki 12 V,5Apower source (see Figure 2.4).
Figure 2.3: The QBot 2 DAQ
Figure 2.4: The QBot 2 DAQ connectors
The QBot 2 DAQ also provides five headers for users to use when integrating additional I/O components. Each
header is a double row where each pin in the first row is electrically connected to the pin in the second row directly
opposite. This allows users to connect a sensor to one row and map the connection from the DAQ I/O to the pins in
the mating row similar to a breadboard. Figure 2.5 shows the location of the headers.
QBOT 2 - User Manual DRAFT - April 14, 2015

Figure 2.5: The QBot 2 DAQ sensor mounting rows
2.3.1 Digital Input/Output Pins (DIO)
The user DIO channels (Figure 2.6) are set as inputs by default. The DIO channels need to be configured as either
inputs (or outputs, but not both) using the HIL Initialize block. Also, if an output needs to be in a known state on
power up, it is recommended that a 10k resistor is put from the I/O to 5Vor GND as needed. The DIO channels are
accessed through the HIL API.
Figure 2.6: The QBot 2 DAQ user digital I/O
2.3.2 Analog inputs
The QBot 2 provides four user analog input channels (Figure 2.7) that are rated for signals between 0−5Vand
uses 12-bit analog to digital converters (ADCs). The analog inputs are accessed through the HIL API.
QBOT 2 - User Manual 8

Figure 2.7: The QBot 2 DAQ user analog inputs
2.3.3 Encoder inputs
The QBot 2 provides two user encoder input channels (Figure 2.8). Each encoder input channel has an A and B
pulse as well as a Z index signal. The encoders are sampled at 100 kHz. The encoder inputs are accessed through
the HIL API.
Figure 2.8: The QBot 2 DAQ user encoder inputs
QBOT 2 - User Manual DRAFT - April 14, 2015

2.3.4 PWM
The QBot 2 provides four user PWM outputs (Figure 2.9) that users can configure to output pulse-width modulated
digital signals.
Figure 2.9: The QBot 2 DAQ user PWM outputs
2.3.5 SPI
The QBot 2 provides one Serial Peripheral Interface (SPI) bus (Figure 2.10), which is a synchronous serial data bus
for interfacing other sensors or devices with the embedded computer. The SPI channel is accessed through the
Steam API (see the SPI protocol in the QUARCrhelp).
Figure 2.10: The QBot 2 DAQ SPI communication port
QBOT 2 - User Manual 10

2.3.6 UART
The QBot 2 provides one UART serial port (Figure 2.11) that can be used to interface with 3.3V(TTL) serial devices.
The UART is accessed through the Steam API (see the serial protocol in the QUARCrhelp).
Figure 2.11: The QBot 2 DAQ UART communication port
2.3.7 I2C
The QBot 2 provides one inter-integrated circuit (I2C) serial bus (Figure 2.12) for interfacing with external sensors
and devices. The I2C channel is accessed through the Steam API (see the I2C protocol in the QUARCrhelp).
Figure 2.12: The QBot 2 DAQ I2C communication port
QBOT 2 - User Manual DRAFT - April 14, 2015

2.4 Gumstix DuoVero Embedded Computer
The Gumstix DuoVero is a small-scale embedded computer that runs the QUARCrruntime. With QUARCr
installed, code generated from MatlabrSimulinkris cross-compiled, downloaded, and executed directly on the
DuoVero. The DuoVero is connected directly to the QBot 2 DAQ on the bottom side of the PCB. The DuoVero
also comes with integrated 802.11 b/g/nWiFi to allow wireless connection between the target Gumstix and the host
computer and/or other vehicles.
2.5 The Kinect Sensor
The Microsoft Kinect sensor is an integrated RGB camera and depth sensor used in a variety of experiments (Figure
2.13a). The Kinect camera provides RGB image capture and 11-bit depth sensing at a resolution of 640 ×480. The
Kinect's depth sensor utilizes infrared light and has a range of 0.5mto 6m. Due to the type of infrared sensor, the
QBot 2 Kinect should only be used indoors in locations without direct sunlight for best results.
(a) The QBot 2 with Kinect sensor (b) The QBot 2 Kinect thumb screws
(c) The QBot 2 Kinect tilted at 0degrees and 21.5degrees
Figure 2.13: The QBot 2 Kinext sensor
QBOT 2 - User Manual 12

The QBot 2 Kinect is mounted on a structure that allows the Kinect to be oriented horizontally as well as tilted
downward for various viewing positions. The QBot 2 Kinect structure has four thumb screws (two on each side)
as shown in Figure 2.13b. To adjust the Kinect's viewing angle, loosen all of the thumb screws slightly but do not
remove them all the way. Tilt the Kinect to the desired angle and tighten all of the thumb screws. The maximum
and minimum angles are designed to allow the kinect to be oriented facing stright ahead (tilted at 0degrees) as well
as at an angle of 21.5degrees such that the ground in front of the QBot 2 in the viewing angle as shown in Figure
2.13c.
2.6 QBot 2 sensors interface
This section describes the blocks that are used to read the QBot 2 sensors in Simulinkrand write outputs to the
QBot 2 motors. The QUARCrHardware-In-the-Loop (HIL) blockset it used to communicate with Quanserrdata
acquisition cards. For detailed information on the HIL blockset see the QUARCrHIL user guide in the Matlabrhelp
under QUARC Targets |User's Guide |Accessing Hardware.
2.6.1 QBot 2 I/O
The QBot 2 provides an interface to the I/O of the Kobuki robot platform including its wheel encoders, sensors, and
motor outputs as well as several DAQ I/O channels for user expansion, which can be used to interface to a variety
of third-party sensors and actuators. The I/O of the QBot 2 DAQ includes:
• 4 analog inputs
• 8 reconfigurable digital I/O
• 4 PWM outputs
• 2 encoder inputs
• 1 UART
• 1 SPI
• 1 I2C
2.3 shows the location of the I/O listed above on the QBot 2 DAQ.
The Kobuki platform provides the following I/O:
• 2 wheel encoder inputs
• 2 wheel speed outputs
• 2 digital LED outputs
• 4 digital power enable outputs
• 3 analog and digital cliff sensors
• 2 analog motor current inputs
• 3 digital bumper sensors
• 3 digital wheel drop sensors
• 3 digital buttons
• 2 overcurrent sensors
• 2 wheel PWM measurements
QBOT 2 - User Manual DRAFT - April 14, 2015

• 3-axis gyroscope
• 1 Z-axis angle measurement (heading)
• 1 battery voltage sensor
The analog, digital, and robot and DAQ I/O listed above is accessed using the QUARCrHIL blockset. The
UART, SPI, and I2Ccommunication channels are accessed through the QUARCrStream blockset. For more
information on accessing communication stream data see the QUARCrhelp under QUARC Targets |User's Guide
|Communications. Table 2.2 lists the HIL blocks used to communicate with the QBot 2's data acquisition hardware.
Block Description
The HIL Initialize block selects the DAQ board and configures the board
parameters. The HIL Initialize block is named via the Board name parameter,
and all other HIL blocks reference the corresponding HIL Initialize through its
name. The HIL blocks will interface to the DAQ specified in the HIL Initialize
Board type parameter qbot2.
The HIL Read Write block is used to read sensor measurements from the DAQ
and robot and write motor commands to the QBot 2 motors. The inputs and
outputs are specified with numeric channel numbers given in Table 2.4 and
Table 2.3, respectively.
The HIL Watchdog block is used to set the timeout limit for the watchdog timer.
For the QBot 2 DAQ board, if there is no motor output command received for
a consecutive period of time exceeding the watchdog timeout value then the
watchdog will trigger, forcing the motor outputs to 0. The default timeout value
for the watchdog is 50 ms unless specified otherwise with this block. This block
can be used to change the timeout value if 50 ms is not suitable.
Table 2.2: HIL blocks
QBOT 2 - User Manual 14

Channel type Write channel numbers Description Units
Analog none -
PWM 0 - 3 User PWM outputs %
Digital 0 - 7 Reconfigurable digital I/O
8 User LED on QBot 2 DAQ
9 LED1 red colour
10 LED1 green colour
11 LED2 red colour
12 LED2 green colour
13 Enable 3.3V power
14 Enable 5V power
15 Enable 12V/5A power
16 Enable 12V/1.5A power
Other 2000 Right wheel speed m/s
2001 Left wheel speed m/s
16000 Predefined sound (0=on sound, 1=off sound,
2=recharge, 3=button, 4=error, 5=task start,
6=task end)
-
Table 2.3: QBot 2 output channels
QBOT 2 - User Manual DRAFT - April 14, 2015

Channel type Read channel numbers Description Units
Analog 0 - 3 User analog inputs V
4 Supply voltage (battery) V
5 Right cliff sensor V
6 Central cliff sensor V
7 Left cliff sensor V
8 Right motor current A
9 Left motor current A
Encoder 0 - 1 User encoder inputs counts
2 Right wheel (2578 counts per revolution) counts
3 Left wheel (2578 counts per revolution) counts
Digital 0 - 7 User reconfigurable digital I/O
8 Right bumper
9 Central bumper
10 Left bumper
11 Right wheel drop
12 Left wheel drop
13 Right cliff detected
14 Central cliff detected
15 Left cliff detected
16 Button B0
17 Button B1
18 Button B2
19 Right overcurrent
20 Left overcurrent
Other 13000-13001 User encoder 0-1 index position counts
12000 Timestamp s
16000 Charge state (0=discharging, 2=docking
charged, 6=docking charging, 18=adapter
charged, 22=adapter charging)
-
11000 Right wheel PWM %
11001 Left wheel PWM %
1002 Z-axis angle rad
3000 X-axis angular velocity rad/s
3001 Y-axis angular velocity rad/s
3002 Z-axis angular velocity rad/s
Table 2.4: QBot 2 input channels
QBOT 2 - User Manual 16

2.6.2 QBot 2 Kinect
The QBot 2 is integrated with a Microsoft Kinect sensor, which is capable of capturing RGB image data as well as
12-bit depth data at several resolutions and framerates. The Kinect sensor data is available through the QUARCr
Multimedia blockset under QUARC Targets/Multimedia in the Simulink Library Browser. Table 2.5 shows the various
blocks used to interface with the Kinect and describes their purpose. Note: to use the QUARCrKinect blocks,
you will need to have the Microsoft Kinect SDK v1.8 and the Microsoft Kinect for Windows Developer Toolkit v1.8
installed on the host PC.
Block Description
The Kinect Initialize block is required to initialize the Kinect sensor and provide
a name for the Kinect that other blocks use to associate with that sensor.
The Kinect Video Capture block is used to capture RGB or greyscale images
from the Kinect sensor. You can specify the desired resolution as well as the
output format (RGB or greyscale). The Video Capture block outputs a boolean
new signal to indicate when a new frame is available so that it can be used to
trigger image processing only once per frame.
The Kinect Get Depth block is used to output a depth-mapped image at the
specified resolution. The Kinect can measure depth from 0.5mto 6m.
The Kinect Get Image block is used to acquire RGB or greyscale images from
the Kinect. The image sensor type can be specified as the color or infrared
image sensor. The image resolution can be selected as either 640 ×480 or
1280×960. The image frame number is output from the block and can be used
to trigger image processing when a new image is available.
The Display Image block is used to display image data on the host computer.
The Display Image block accepts greyscale and RGB image matrices as its
input. The Display Image block acts like a scope and is not synchronized
with the model; in other words, if the Display Image block cannot keep up with
the performance of the model it will not block the model thread(s). Since the
Display Image block requires the images to be transmitted between the target
and the host it is recommended that the images used for display purposes be
downsampled or sent at a slower rate.
Table 2.5: Kinect blocks
QBOT 2 - User Manual DRAFT - April 14, 2015

3 Setup
3.1 Setting up the QBot 2
Follow these steps to setting up the QBot 2:
1. Connect the battery inside the Kobuki robot.
2. Press the power button shown in Figure 2.2c. This should turn on both vehicle and QBot 2 DAQ and Gumstix
computer. After you make sure that this works, turn off the robot and follow the steps below to establish network
connection.
3.2 Establishing Network Connection
The QBot 2 package comes with a pre-configured wireless router and automatically connects to the WiFi network
Quanser_UVS. It uses TCP/IP connection for communicating with the host computer and/or other Quanserr
unmanned vehicles. The Host PC and each of the vehicles must have unique IP addresses and the range of
these addresses are defined below (suggested):
Host PC(s) 192.168.2.10 to 192.168.2.19
Quanser vehicles (Gumstix) 192.168.2.20 to 192.168.2.254
These Steps outlined below for setting up the host computer wireless connection only need to be performed once.
1. Power up and turn on the wireless router.
2. After turning on the router that is provided, wait for about 60 seconds for the wireless network to establish and
then turn on the QBot 2 power and wait for it to boot up (approximately 60 seconds).
3. Connect your PC network card to any of the ports of the router (e.g. port number 1 to 4) using the network
cable provided (you can also connect to the Quanser_UVS or Quanser_UVS-5G wireless network if your
computer has wireless adapters, however, wired connection between your PC and the router is preferred for
better performance). If you choose wireless connectivity between your PC and the router, you should use the
password UVS_wifi to connect to the wifi network.
4. Using the Windows Network Connection utility in the Windows Task Bar, open the Network and Sharing Center.
QBOT 2 - User Manual 18

Figure 3.1: Open network settings. The Unidentified network is related to the wired connection to the router. If you are
using a wired connection to the router and you don't have wireless adapter, you will not see Quanser_UVS network.
5. Click on Change adapter settings.
6. Right-click on the Local Area Connection x, Unidentified network connection and click Properties. If you chose
wireless connectivity between your PC and the router, choose the Wireless Network Connection instead, right
click on it and click Properties.
7. Under This connection uses the following items:, select Internet Protocol Version 4 (TCP/IPv4) and click
Properties.
Figure 3.2: Network properties
8. Instead of obtaining an IP address of the computer automatically, select Use the following IP address and enter
the following:
QBOT 2 - User Manual DRAFT - April 14, 2015

IP address: 192.168.2.10 (For multiple host PCs, use different IP addresses within the valid range)
Subnet mask: 255.255.255.0
Figure 3.3: IP settings
9. Make sure you can ping the router by typing ping 192.168.2.1 in the Run box in Windows (go to the Start menu
and search for Run and click Run, Figure 3.4). If the connection to the router is successful you will see the ping
replies in the command window. If you cannot ping the router, check network connectivity and your IP address
before going to the next steps.
10. If the QBot 2 is powered on, the QBot 2 can be pinged by typing ping {IP of the QBot 2}in the Run box in
Windows (go to the Start menu and click Run, Figure 3.4). The QBot 2 IP address is labeled on the DAQ PCB
as shown in Figure 3.5. If the connection is successful you will see the ping replies in the command window.
If you cannot ping the robot, power cycle the QBot 2 and wait for about 60 seconds to reboot.
Note: You may need to disable Windows firewall to establish a connection.
Figure 3.4: Pinging the QBot 2
QBOT 2 - User Manual 20