WEEK 13 – Wow.
02/28/2014
These past few weeks have been crazy. I have learned more over the past few weeks about the robot than ever. I have successfully combined voice navigation and the follower application. I decided to incorporate an android device with my project for ease of startup and access. The android device serves as my kill switch for testing, and also allows me to monitor the follower node individually. So far everything is going smoothly. I only wish I had more time so I could expand my work even further.
WEEK 12 – I FOUND A BOOK!!!!!
04/18/2014
I found a book called “ROS-By-Example” just recently. How this did not come up in my initial research is beyond me. I had to follow three links just to get to the book’s website. This book not only gives more in depth explanations of ROS functionality and how things work, but it also has its own package repository!!!! These packages contain applications that allow the turtlebot to color track, receive voice commands, talk, line follow, etc. This is HUGE. I have immediately begun installing and testing these applications. Everything is going extremely smoothly right now. My plan is to combine the voice navigation functionality with the follower functionality in order to have a form of autonomous navigation and have a good test subject for cross application communication.
WEEK 10 – THINGS FINALLY COMING TOGETHER
04/04/2014
I have not written a journal entry in a while. This is likely due to my large amount of progress with the ROS software. I have been extremely wrapped up in testing and altering various applications. I think I finally have a solid understanding of why all these applications are separated instead of combined. I have discovered the documentation on “ROS Plumbing” which has helped me figure out how various pieces of the Robot send and receive messages. With this new knowledge I have been experimenting with running applications side by side to see what kind of communication interrupts occur. Once I have a good understanding of what forms of communication take precedence I think I can try combining applications.
WEEK 7 – PROGRESS!
03/14/2014
I finally have all the individual applications working with ROS (so it seems). I have been playing around with their functionalities to see how they work and react in an environment. I am starting to focus more on smaller applications like the Follower Demo so that I can get a good feel for how the Turtlebot handles visual data. I am currently capable of running all the applications I want on the Turtlebot, but am still having issues with visuals and monitors on the Worstation side. The weird thing is that sometimes they work and sometimes they don’t. I am hoping to figure out a more efficient way to run and test applications. Currently start up time is a bit ridiculous (roughly 5-10 minutes). Now that I finally feel confident that my ROS installation is working I am doing a lot of playing with various applications. I want to feel confident that I know how they are all functioning.
WEEK 5 – LETS TRY AGAIN
02/28/2014
I had no more success with the groovy build than I did with the hydro build. I decided to start again from scratch last week. I am using the hydro build because it is the most up to date. I have finally had good success with this install. I am getting a better feel for how ROS works and how the applications are supposed to function. I need to keep digging around so I can get a better understanding of how ROS is working.
WEEK 3 - STILL LEARNING
02/14/2014
I have made a good amount of progress getting to know the software behind the turtlebot. There is still a lot I don't know yet, but I think I am to the point where I need to decide which aspects I don't need to worry about in order to make progress on development. I did get the robot moving around with a game pad. I can watch live feed on my computer and drive the robot around without being in the same room as it. I have moved on from my original ROS distro. I have decided to give the groovy build a shot as that is what was used on this project last year. I believe I have a working development environment set up, but I think I need to come up with something a little more efficient that what I have. If I screw something up I will have to start almost from scratch at this point. I have a feeling I can work some sort of repository system into the Turtlebot's netbook. I am attempting to get everything Joe set up last year set up for myself. I am hoping to spend time getting to know how everything works so I can perhaps add to the system. At this point I am still a little unsure on how the development aspect is going to work with this project.
I am getting close to the point where I can start trying to do a little development. I need to decide on a place to start. I think I will probably focus mainly on getting the robot to "learn" paths. At this point I feel comfortable with the hardware and software environments for the most part. I feel like I do not quite know exactly how everything is operating behind the scenes yet, but I think I am close. I am going to finish getting Joe's set up installed so I can start forming ideas for whats next.
WEEK 2 - GETTING TO KNOW TURTLEBOT
02/07/2014
I spent the majority of this week getting a feel for how Turtlebot works. I have found a lot of good open source software to start working with. I have installed a form of ROS and have gotten the turtlebot talking to my workstation machine. My goal is to get the robot fully functional via the third-party software that comes with the ROS build. There is a piece of software for just about every individual function of the robot which is really nice. I have live video feed and some movement from the Turtlebot via these applications. I have been working on teleoperating the robot via some sort of game pad. So far it has proven to be a bit tough, but I think I am close.
Once I get to know how a lot of this software works and when I know the turtlebot is functioning correctly I plan to try and implement some of the work from Joe's project last year. I want to see if I can get to where he was at with it and start refining things in order to have a solid working ...something... I look forward to getting into some of the code.
WEEK 1 - LET THE RESEARCH BEGIN
01/30/2014
Today I got my project assignment, and it was not what I expected. I am going to be doing some robotics and engineering. The goal of the project is ultimately to develop an Artificial Intelligence(AI) system so that this robot can guide people around a nursing home. My robot is of a build called the "Turtlebot" which looks like a bookshelf on wheels. This robot's base is made from one of those smart vacuums that has an infrared sensor to avoid walls and such. Off the base there is a bookshelf like rack for placing a laptop which will connect to the smart vacuum. Then to top it all off there is a Microsoft Kinect hard wired to the top of the robot so that its features can be incorporated.
To start out I am going to have to do a ton of research on this robot and how it works. I am told that there is a lot of open source software out there that might help me reach my end goal. The problem here is that "open source" and "functioning" are far from the same thing. So it may be difficult to come up with a working API to do what I want. I think I should start out by trying to get the robot to move around the room with perhaps a joystick or game pad so that I can see the movement style and how the robot interacts with the computer. I want to then move on to information storing. For example, can I get the robot to store directions to a certain location? Then I want to perhaps get the robot to follow me places (I have no clue how I will do that recognition) and store the directions to where ever we go, because robot minions that follow you places are cool. From there I should have all the functionality I need to work on getting the damn thing to move on its own.
This project is way outside my normal realm of development. As of now I have little to no resources to work off. I was given the blog site of the graduate who tried to do this last year. Though robotics is not my thing, I am hoping to reach the point where this project can be taken forward in the future by other students who are interested in robotics.
Next Steps: RESEARCH THE ROBOT! find open source software to look over, choose a platform for development.
02/28/2014
These past few weeks have been crazy. I have learned more over the past few weeks about the robot than ever. I have successfully combined voice navigation and the follower application. I decided to incorporate an android device with my project for ease of startup and access. The android device serves as my kill switch for testing, and also allows me to monitor the follower node individually. So far everything is going smoothly. I only wish I had more time so I could expand my work even further.
WEEK 12 – I FOUND A BOOK!!!!!
04/18/2014
I found a book called “ROS-By-Example” just recently. How this did not come up in my initial research is beyond me. I had to follow three links just to get to the book’s website. This book not only gives more in depth explanations of ROS functionality and how things work, but it also has its own package repository!!!! These packages contain applications that allow the turtlebot to color track, receive voice commands, talk, line follow, etc. This is HUGE. I have immediately begun installing and testing these applications. Everything is going extremely smoothly right now. My plan is to combine the voice navigation functionality with the follower functionality in order to have a form of autonomous navigation and have a good test subject for cross application communication.
WEEK 10 – THINGS FINALLY COMING TOGETHER
04/04/2014
I have not written a journal entry in a while. This is likely due to my large amount of progress with the ROS software. I have been extremely wrapped up in testing and altering various applications. I think I finally have a solid understanding of why all these applications are separated instead of combined. I have discovered the documentation on “ROS Plumbing” which has helped me figure out how various pieces of the Robot send and receive messages. With this new knowledge I have been experimenting with running applications side by side to see what kind of communication interrupts occur. Once I have a good understanding of what forms of communication take precedence I think I can try combining applications.
WEEK 7 – PROGRESS!
03/14/2014
I finally have all the individual applications working with ROS (so it seems). I have been playing around with their functionalities to see how they work and react in an environment. I am starting to focus more on smaller applications like the Follower Demo so that I can get a good feel for how the Turtlebot handles visual data. I am currently capable of running all the applications I want on the Turtlebot, but am still having issues with visuals and monitors on the Worstation side. The weird thing is that sometimes they work and sometimes they don’t. I am hoping to figure out a more efficient way to run and test applications. Currently start up time is a bit ridiculous (roughly 5-10 minutes). Now that I finally feel confident that my ROS installation is working I am doing a lot of playing with various applications. I want to feel confident that I know how they are all functioning.
WEEK 5 – LETS TRY AGAIN
02/28/2014
I had no more success with the groovy build than I did with the hydro build. I decided to start again from scratch last week. I am using the hydro build because it is the most up to date. I have finally had good success with this install. I am getting a better feel for how ROS works and how the applications are supposed to function. I need to keep digging around so I can get a better understanding of how ROS is working.
WEEK 3 - STILL LEARNING
02/14/2014
I have made a good amount of progress getting to know the software behind the turtlebot. There is still a lot I don't know yet, but I think I am to the point where I need to decide which aspects I don't need to worry about in order to make progress on development. I did get the robot moving around with a game pad. I can watch live feed on my computer and drive the robot around without being in the same room as it. I have moved on from my original ROS distro. I have decided to give the groovy build a shot as that is what was used on this project last year. I believe I have a working development environment set up, but I think I need to come up with something a little more efficient that what I have. If I screw something up I will have to start almost from scratch at this point. I have a feeling I can work some sort of repository system into the Turtlebot's netbook. I am attempting to get everything Joe set up last year set up for myself. I am hoping to spend time getting to know how everything works so I can perhaps add to the system. At this point I am still a little unsure on how the development aspect is going to work with this project.
I am getting close to the point where I can start trying to do a little development. I need to decide on a place to start. I think I will probably focus mainly on getting the robot to "learn" paths. At this point I feel comfortable with the hardware and software environments for the most part. I feel like I do not quite know exactly how everything is operating behind the scenes yet, but I think I am close. I am going to finish getting Joe's set up installed so I can start forming ideas for whats next.
WEEK 2 - GETTING TO KNOW TURTLEBOT
02/07/2014
I spent the majority of this week getting a feel for how Turtlebot works. I have found a lot of good open source software to start working with. I have installed a form of ROS and have gotten the turtlebot talking to my workstation machine. My goal is to get the robot fully functional via the third-party software that comes with the ROS build. There is a piece of software for just about every individual function of the robot which is really nice. I have live video feed and some movement from the Turtlebot via these applications. I have been working on teleoperating the robot via some sort of game pad. So far it has proven to be a bit tough, but I think I am close.
Once I get to know how a lot of this software works and when I know the turtlebot is functioning correctly I plan to try and implement some of the work from Joe's project last year. I want to see if I can get to where he was at with it and start refining things in order to have a solid working ...something... I look forward to getting into some of the code.
WEEK 1 - LET THE RESEARCH BEGIN
01/30/2014
Today I got my project assignment, and it was not what I expected. I am going to be doing some robotics and engineering. The goal of the project is ultimately to develop an Artificial Intelligence(AI) system so that this robot can guide people around a nursing home. My robot is of a build called the "Turtlebot" which looks like a bookshelf on wheels. This robot's base is made from one of those smart vacuums that has an infrared sensor to avoid walls and such. Off the base there is a bookshelf like rack for placing a laptop which will connect to the smart vacuum. Then to top it all off there is a Microsoft Kinect hard wired to the top of the robot so that its features can be incorporated.
To start out I am going to have to do a ton of research on this robot and how it works. I am told that there is a lot of open source software out there that might help me reach my end goal. The problem here is that "open source" and "functioning" are far from the same thing. So it may be difficult to come up with a working API to do what I want. I think I should start out by trying to get the robot to move around the room with perhaps a joystick or game pad so that I can see the movement style and how the robot interacts with the computer. I want to then move on to information storing. For example, can I get the robot to store directions to a certain location? Then I want to perhaps get the robot to follow me places (I have no clue how I will do that recognition) and store the directions to where ever we go, because robot minions that follow you places are cool. From there I should have all the functionality I need to work on getting the damn thing to move on its own.
This project is way outside my normal realm of development. As of now I have little to no resources to work off. I was given the blog site of the graduate who tried to do this last year. Though robotics is not my thing, I am hoping to reach the point where this project can be taken forward in the future by other students who are interested in robotics.
Next Steps: RESEARCH THE ROBOT! find open source software to look over, choose a platform for development.