Robotics Class 2011/Assignment 3 - Revision history

Andrewtron3000 at 00:45, 8 September 2012

2012-09-08T00:45:32Z

ITechGeek at 20:46, 3 April 2012

2012-04-03T20:46:03Z

← Older revision		Revision as of 20:46, 3 April 2012
Line 1:		Line 1:
			[[Category:RobotClass]]

	Write a ROS node that subscribes to the "/face_coords" topic that is generated from the previous face detector homework assignment and use the information provided in that topic to move the robot base by publishing messages on the "/cmd_vel" topic. The goal of the assignment is to build a face tracker that attempts to move the robot to keep faces centered in the camera frame. The "/cmd_vel" topic is of type "geometry_msgs/Twist", and you can learn more about it by typing:		Write a ROS node that subscribes to the "/face_coords" topic that is generated from the previous face detector homework assignment and use the information provided in that topic to move the robot base by publishing messages on the "/cmd_vel" topic. The goal of the assignment is to build a face tracker that attempts to move the robot to keep faces centered in the camera frame. The "/cmd_vel" topic is of type "geometry_msgs/Twist", and you can learn more about it by typing:

Andrewtron3000 at 17:34, 1 April 2012

2012-04-01T17:34:47Z

← Older revision		Revision as of 17:34, 1 April 2012
Line 13:		Line 13:
	It may help to follow the ROS instructions [http://www.ros.org/wiki/simulator_gazebo/Tutorials/StartingGazebo here] in order to install and get gazebo running.		It may help to follow the ROS instructions [http://www.ros.org/wiki/simulator_gazebo/Tutorials/StartingGazebo here] in order to install and get gazebo running.

	The next step will be to make sure you have the ''gazebo_erratic_plugins''. These are extensions to gazebo to support differential drive robots (like the iRobot Create) that have two driven wheels. If you installed the ROS installation using Synaptic, you can search Synaptic for "erratic" and you should see a package named '''ros-~~diamondback~~-erratic-robot'''. You will want to install this package through Synaptic. If you compiled from source, you will want to check out, rosdep install, and rosmake the [http://www.ros.org/wiki/erratic_robot erratic_robot] package.		The next step will be to make sure you have the ''gazebo_erratic_plugins''. These are extensions to gazebo to support differential drive robots (like the iRobot Create) that have two driven wheels. If you installed the ROS installation using Synaptic, you can search Synaptic for "erratic" and you should see a package named '''ros-electric-erratic-robot'''. You will want to install this package through Synaptic. If you compiled from source, you will want to check out, rosdep install, and rosmake the [http://www.ros.org/wiki/erratic_robot erratic_robot] package.

	Once that is complete, you can proceed to checking out the HacDC robot simulation:		Once that is complete, you can proceed to checking out the HacDC robot simulation:

Andrewtron3000 at 20:12, 2 July 2011

2011-07-02T20:12:05Z

← Older revision		Revision as of 20:12, 2 July 2011
Line 3:		Line 3:
	'''rosmsg show geometry_msgs/Twist'''		'''rosmsg show geometry_msgs/Twist'''

	For the simulated robot you will use for this assignment, it is important to know its coordinate system. X is positive going forward, Y is positive going to the left, and Z is positive going up. This coordinate system is a right handed coordinate system. We spent a significant amount of time in class going over the importance of keeping track of the coordinate system of the robot, and always making sure you follow the right-hand rule for dealing with ~~coordinate systems in ROS. (ROS uses~~ the ~~right-handed rule for~~ coordinate ~~frames)~~. Because the robot we are using is a differential drive robot, it has two powered wheels and can only move forward and backward, and it can also rotate in place. Thus, for this assignment, when you publish the "/cmd_vel" topic, you need only worry about populating the ''linear.x'' (forward and backward motion) and ''angular.z'' (turning left and right in place) components of the geometry_msgs/Twist structure. Following the coordinate system above, moving the robot forward would require you to publish a geometry_msgs/Twist message with a positive ''linear.x'' component. Rotating the robot in place to the left would require you to publish a geometry_msgs/Twist message with a positive value in the ''angular.z'' component. No other fields need be populated in the geometry_msgs/Twist message other than ''linear.x'' and ''angular.z''. Remember that you must publish the geometry_msgs/Twist message on the "/cmd_vel" topic. The robot is listening for messages on this topic and will move accordingly.		For the simulated robot you will use for this assignment, it is important to know its coordinate system. X is positive going forward, Y is positive going to the left, and Z is positive going up. This coordinate system is a right handed coordinate system. We spent a significant amount of time in class going over the importance of keeping track of the coordinate system of the robot, and always making sure you follow the right-hand rule for dealing with the robot coordinate frame.

			Because the robot we are using is a differential drive robot, it has two powered wheels and can only move forward and backward, and it can also rotate in place. Thus, for this assignment, when you publish the "/cmd_vel" topic, you need only worry about populating the ''linear.x'' (forward and backward motion) and ''angular.z'' (turning left and right in place) components of the geometry_msgs/Twist structure. Following the coordinate system above, moving the robot forward would require you to publish a geometry_msgs/Twist message with a positive ''linear.x'' component. Rotating the robot in place to the left would require you to publish a geometry_msgs/Twist message with a positive value in the ''angular.z'' component. No other fields need be populated in the geometry_msgs/Twist message other than ''linear.x'' and ''angular.z''. Remember that you must publish the geometry_msgs/Twist message on the "/cmd_vel" topic. The robot is listening for messages on this topic and will move accordingly.

	This assignment is more complicated than the first two, and requires several nodes to be running. First, you must install gazebo and be able to run the simple gazebo empty world launch script. The simple command to start gazebo is:		This assignment is more complicated than the first two, and requires several nodes to be running. First, you must install gazebo and be able to run the simple gazebo empty world launch script. The simple command to start gazebo is:

Andrewtron3000 at 20:06, 2 July 2011

2011-07-02T20:06:52Z

← Older revision		Revision as of 20:06, 2 July 2011
Line 3:		Line 3:
	'''rosmsg show geometry_msgs/Twist'''		'''rosmsg show geometry_msgs/Twist'''

	For the simulated robot you will use for this assignment, it is important to know its coordinate system. X is positive going forward, Y is positive going to the left, and Z is positive going up. This coordinate system is a right handed coordinate system. We spent a significant amount of time in class going over the importance of keeping track of the coordinate system of the robot, and always making sure you follow the right-hand rule for dealing with coordinate systems in ROS. (ROS uses the right-handed rule for coordinate frames). Because the robot we are using is a differential drive robot, it has two powered wheels and can only move forward and backward, and it can also rotate in place. Thus, for this assignment, when you publish the "/cmd_vel" topic, you need only worry about populating the ''linear.x'' and ''angular.z'' components of the geometry_msgs/Twist structure. Following the coordinate system above, moving the robot forward would require you to publish a geometry_msgs/Twist message with a positive ''linear.x'' component. Rotating the robot in place to the left would require you to publish a geometry_msgs/Twist message with a positive value in the ''angular.z'' component. No other fields need be populated in the geometry_msgs/Twist message other than ''linear.x'' and ''angular.z''. Remember that you must publish the geometry_msgs/Twist message on the "/cmd_vel" topic. The robot is listening for messages on this topic and will move accordingly.		For the simulated robot you will use for this assignment, it is important to know its coordinate system. X is positive going forward, Y is positive going to the left, and Z is positive going up. This coordinate system is a right handed coordinate system. We spent a significant amount of time in class going over the importance of keeping track of the coordinate system of the robot, and always making sure you follow the right-hand rule for dealing with coordinate systems in ROS. (ROS uses the right-handed rule for coordinate frames). Because the robot we are using is a differential drive robot, it has two powered wheels and can only move forward and backward, and it can also rotate in place. Thus, for this assignment, when you publish the "/cmd_vel" topic, you need only worry about populating the ''linear.x'' (forward and backward motion) and ''angular.z'' (turning left and right in place) components of the geometry_msgs/Twist structure. Following the coordinate system above, moving the robot forward would require you to publish a geometry_msgs/Twist message with a positive ''linear.x'' component. Rotating the robot in place to the left would require you to publish a geometry_msgs/Twist message with a positive value in the ''angular.z'' component. No other fields need be populated in the geometry_msgs/Twist message other than ''linear.x'' and ''angular.z''. Remember that you must publish the geometry_msgs/Twist message on the "/cmd_vel" topic. The robot is listening for messages on this topic and will move accordingly.

	This assignment is more complicated than the first two, and requires several nodes to be running. First, you must install gazebo and be able to run the simple gazebo empty world launch script. The simple command to start gazebo is:		This assignment is more complicated than the first two, and requires several nodes to be running. First, you must install gazebo and be able to run the simple gazebo empty world launch script. The simple command to start gazebo is:

Andrewtron3000 at 20:04, 2 July 2011

2011-07-02T20:04:32Z

← Older revision		Revision as of 20:04, 2 July 2011
Line 44:		Line 44:
	Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.		Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.

	In class we also went over PID control. There is an excellent article on PID [http://www.eetimes.com/ContentEETimes/Documents/Embedded.com/2000/f-wescot.pdf here] that is recommended reading. Note that for this particular problem, the error term should be thought of as the number of pixels between the center of the detected face and a vertical line running down the center of the image frame. Since the images from the camera are 352 pixels wide, the center of the image is at 176. Then the x coordinate of the "/face_coords" messages can be subtracted from 176 to find the error term that is fed into the controller of your choice to output a commanded velocity that should be fed into the ''angular.z'' field of the geometry_msgs/Twist message. Note for this homework, it is recommended to start by only rotating the robot in place to keep an image in the center of the frame. Once you get that working you can think about moving the robot forward and backwards to keep the face at a constant scale, but this will also require modifications to the "/face_coords" message since "/face_coords" currently only provides the center of the located face, not the size of the bounding rectangle. Extra credit is assigned for getting these modifications into your solution.		In class we also went over PID control. There is an excellent article on PID [http://www.eetimes.com/ContentEETimes/Documents/Embedded.com/2000/f-wescot.pdf here] that is recommended reading. Note that for this particular problem, the error term should be thought of as the number of pixels between the center of the detected face and a vertical line running down the center of the image frame. Since the images from the camera are 352 pixels wide, the center of the image is at 176. Then the x coordinate of the "/face_coords" messages can be subtracted from 176 to find the error term that is fed into the controller of your choice to output a commanded velocity that should be fed into the ''angular.z'' field of the geometry_msgs/Twist message. Note for this homework, it is recommended to start by only rotating the robot in place to keep an image in the center of the frame (i.e. only populate the ''angular.z'' of the geometry_msgs/Twist message). Once you get that working you can think about moving the robot forward and backwards to keep the face at a constant scale, but this will also require modifications to the "/face_coords" message since "/face_coords" currently only provides the center of the located face, not the size of the bounding rectangle. Extra credit is assigned for getting these modifications into your solution.

	There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.		There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.

Andrewtron3000 at 20:03, 2 July 2011

2011-07-02T20:03:43Z

← Older revision		Revision as of 20:03, 2 July 2011
Line 44:		Line 44:
	Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.		Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.

	In class we also went over PID control. There is an excellent article on PID [http://www.eetimes.com/ContentEETimes/Documents/Embedded.com/2000/f-wescot.pdf here] that is recommended reading. Note that for this particular problem, the error term should be thought of as the number of pixels between the center of the detected face and a vertical line running down the frame. Since the images from the camera are 352 pixels wide, the center of the ~~images~~ is at 176. Then the x coordinate of the "/face_coords" messages can be subtracted from 176 to find the error term.		In class we also went over PID control. There is an excellent article on PID [http://www.eetimes.com/ContentEETimes/Documents/Embedded.com/2000/f-wescot.pdf here] that is recommended reading. Note that for this particular problem, the error term should be thought of as the number of pixels between the center of the detected face and a vertical line running down the center of the image frame. Since the images from the camera are 352 pixels wide, the center of the image is at 176. Then the x coordinate of the "/face_coords" messages can be subtracted from 176 to find the error term that is fed into the controller of your choice to output a commanded velocity that should be fed into the ''angular.z'' field of the geometry_msgs/Twist message. Note for this homework, it is recommended to start by only rotating the robot in place to keep an image in the center of the frame. Once you get that working you can think about moving the robot forward and backwards to keep the face at a constant scale, but this will also require modifications to the "/face_coords" message since "/face_coords" currently only provides the center of the located face, not the size of the bounding rectangle. Extra credit is assigned for getting these modifications into your solution.

	There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.		There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.

Andrewtron3000 at 20:00, 2 July 2011

2011-07-02T20:00:29Z

← Older revision		Revision as of 20:00, 2 July 2011
Line 40:		Line 40:
	'''rosrun teleop_twist_keyboard teleop_twist_keyboard.py'''		'''rosrun teleop_twist_keyboard teleop_twist_keyboard.py'''

	You can following the onscreen instructions on how to use it, but it is advisable to slow the robot commands down by pushing the "z" key a few times, so that the "speed" is around 0.2. This has been found to be a reasonable maximum speed for linear motion with this particular robot. You can move forward by pressing "i", and turn left by pressing "j", but these are all on the on-screen instructions when you run the teleop_twist_keyboard node.		You can following the onscreen instructions on how to use it, but it is advisable to slow the robot commands down by pushing the "z" key a few times, so that the "speed" is around 0.2. This has been found to be a reasonable maximum speed for linear motion with this particular robot. You can move forward by pressing "i", and turn left by pressing "j", but these are all on the on-screen instructions when you run the teleop_twist_keyboard node. At this point, if you are still subscribed to the ''/stereo/left/image_rect'' image stream, you should be able to drive around and see what the robot sees, and you should be able to see the faces on the cubes.

	At this point, if you are still subscribed to the ''/stereo/left/image_rect'' image stream, you should be able to drive around and see what the robot sees, and you should be able to see the faces on the cubes.

	Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.		Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.

Andrewtron3000 at 20:00, 2 July 2011

2011-07-02T20:00:09Z

← Older revision		Revision as of 20:00, 2 July 2011
Line 45:		Line 45:

	Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.		Now, you could start up your face detection node from the previous example. Since the simulated robot is publishing the image topic named "/stereo/left/image_rect", your face detection system should work on the faces in the simulated world.

			In class we also went over PID control. There is an excellent article on PID [http://www.eetimes.com/ContentEETimes/Documents/Embedded.com/2000/f-wescot.pdf here] that is recommended reading. Note that for this particular problem, the error term should be thought of as the number of pixels between the center of the detected face and a vertical line running down the frame. Since the images from the camera are 352 pixels wide, the center of the images is at 176. Then the x coordinate of the "/face_coords" messages can be subtracted from 176 to find the error term.

	There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.		There is a complete solution to the homework assignment if you are interested in studying it. It can be checked out from the ROS HacDC repository. There are two packages. The first is a generic PID control library and the second is the controller itself.

Andrewtron3000 at 19:53, 2 July 2011

2011-07-02T19:53:04Z

← Older revision		Revision as of 19:53, 2 July 2011
Line 60:		Line 60:

	Don't forget for it to work you need to start a face detector first, either your own from the last homework, or the example solution from the last assignment.		Don't forget for it to work you need to start a face detector first, either your own from the last homework, or the example solution from the last assignment.

			[http://wiki.hacdc.org/index.php/Robotics_Class_2011 Back to Robotics Class 2011]

← Older revision		Revision as of 00:45, 8 September 2012
Line 20:		Line 20:

	'''cd''' to where you store your downloaded ROS packages		'''cd''' to where you store your downloaded ROS packages

	'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/irobotron_description'''		'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/irobotron_description'''

	'''rosmake irobotron_description'''		'''rosmake irobotron_description'''

Line 34:		Line 36:

	'''cd''' to where you store your downloaded ROS packages		'''cd''' to where you store your downloaded ROS packages

	'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/floating_faces'''		'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/floating_faces'''

	'''rosmake floating_faces'''		'''rosmake floating_faces'''

Line 53:		Line 57:

	'''cd''' to where you store your downloaded ROS packages		'''cd''' to where you store your downloaded ROS packages

	'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/pid_control'''		'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/pid_control'''

	'''rosmake pid_control'''		'''rosmake pid_control'''

	'''cd''' to where you store your downloaded ROS packages		'''cd''' to where you store your downloaded ROS packages

	'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/face_follow'''		'''svn co http://hacdc-ros-pkg.googlecode.com/svn/trunk/face_follow'''

	'''rosmake face_follow'''		'''rosmake face_follow'''

	To run the face tracker, you can launch it as follows:		To run the face tracker, you can launch it as follows:

	'''roslaunch face_follow follow.launch'''		'''roslaunch face_follow follow.launch'''