Home
914 PC Bots Forum  


<< Start < Prev 1 2 3 4 5 Next > End >>
c6jones720
User

Platinum Boarder
Posts: 350
graphgraph
Karma: 3  
Re:General Status??? - 2008/11/11 16:53 From a personal point of view I believe the different developers started from "ground zero" each time because each one of us had identified something that wasnt available in a previous software version.

Linux users already had player but no equivalent on Windows so Brian came to be.

Brian did everything we ever needed and featured odometry navigation via waypoints etc and could be remotely controlled over a lan.

People didnt like Brian because the setup was complex and it had a DOS like GUI, but still for what it did it was a useful bit of software.

The software that came after this I think was more geared toward making the robot functionality more user friendly. To this end we see programs written in C# and Java.

Unfortunately we also saw that some of the important functionality such as odometry, path planning and navigation by waypoint lost and replaced with other higher level AI features like speech and vision.

Im not saying those program were bad or that it was wrong to add those higher level features. Its like we skipped a step. We should have kept in the lower level features because those are the fundamental building blocks of mobile robotics.

Unfortunately it wasnt always possible to build up from the last persons project because sometimes peolple withheld their source and the others had to start from scratch and refer to the Pilot API and datasheets. Now Ive been an engineer for years so I see datasheets all the time but this one is complicated!

This conversation about the community working together is something I feel strongly about and about 2 years ago thats exactly what we did. A bunch of us, Ernie , Frank, James, Paul and many others bounced ideas off each other.

What came out of these conversations though is a wealth of different ideas and hardware/software, which if we put together we could do something serious with.

As for the cool application - I think there are two different ways of looking at this

1 - You write a cool app and the user uses it but then wants more so a new app has to be written. This is a continuous never ending circle. Examples of this include WhiteboxOS or K9 or BRAIN these are suitable for non-programmers to just pick up and use but may not take you as far as you want

2 - You write the service the cool app needs and make it easy enough for any user to code the extras. Examples of this are Brian, Player and MSRS. These
are designed to be played about with until you are happy.

Incedentally the MSRS software can be broken up and does not have to be used with MSRS.

I have to state this: Those building blocks that will enable a robot to do a useful task are already there in one form or another.

Today and with what we have all done I have seen the robot do nearly all of the following tasks:

1 - Automantically docking with a charger by beaconing and odometry
2 - Navigation and path following + obstacle avoidance /collision detection
3 - Vision processing - the 914 can easily follow a coloured object around a room using a web cam
4 - Line following using duct tape on a carpet
5 - Graphically mapping the environment it is in in 3dimensions
6 - Reading out weather reports and RSS feeds
7 - Performing basic tasks using speech recognition
8 - Display avatars representing robot state
9 - the 914 can vacuum a floor by either driving randomly within a perimeter or just following a path
10 - continually follow a user defined path without deviation for over 10 minutes
11 - Obey motion instructions in custom scripting languages


Why dont you put down some suggestions of what you would like to see? who knows perhaps we could all make it happen.
  | | The administrator has disabled public write access.
croc999
User

Senior Boarder
Posts: 13
graphgraph
Karma: 0  
Re:General Status??? - 2008/11/12 00:35 As I mentioned earlier I'm not a 914 owner, so I don't think its my place to suggest a project, but being robotics fan and an owner of a ER1, I would like to see things in general progress, and in general
Robotics suffers from the "mad scientist" effect, by this I mean the norm is one person toils away on their own robot with custom code written specifically for them and their robot. Don't get me wrong this can be a lot of fun, but as a whole, as a group loose do to this.

I have been trying to get things going on the Er1 and have posted my code and docs at
http://pruim037.proboards107.com/index.cgi?board=nonww&action=display&thread=2443 hoping to get others
Interested in contributing, but the Er1 is pretty much dead, it is however not that different from the 914 and has the same basic HW/SW capabilities, my main reason for not buying a 914 is that it offers nothing over the Er1 other than the skins, would I like to buy one if it offered more?, sure you bet.

But if I did purchase one, then I would be no better off than I am now, no real collaborative effort on which to build upon, think of it this way, if Linus T. had never opened up linux would it be the OS that it is today?, and I think the same goes for robotics, sure things are moving in that direction with MSRS, etc. But with the current tools and capabilities we should be much, much further ahead in what a basic robot can do. And I agree the basic block are all there, but I have to question, if they are there then why hasn't anyone pulled them together to raise the bar?, do you see what I mean?.

I have heard it may times that the robotics industry is like the computer industry back in the 80's, but really it isn't, did manufacturers back then create PC's from scratch?, a few tried and failed, once the x86 was invented/built, others then cloned it, building on what others had already done, thatís when things really took off. I don't see that happening with robotics.

I think I'm starting to ramble even more than before .

Croc4
  | | The administrator has disabled public write access.
jamesbruton
Admin

Admin
Posts: 683
graph
Karma: 8  
Re:General Status??? - 2008/11/12 21:53 Is there any MRDS/MSRS support for the ER1?. If there is then it wouldn't matter whether you had an ER1 or an 914... or a Roomba or a Lego set for that matter - everyone could develop software together.

I guess that's where having a common software platform is more important than the hardware most of the time.

...doesn't have to be MRDS, but it seems like a good start for a Windows based bot.
  | | The administrator has disabled public write access.
c6jones720
User

Platinum Boarder
Posts: 350
graphgraph
Karma: 3  
Re:General Status??? - 2008/11/12 23:31 Im sure I remember Tom Burrick going on about the 914 using ER1 software. Im sure the prototype 914 was supposed to use the same motor controller.
  | | The administrator has disabled public write access.
croc999
User

Senior Boarder
Posts: 13
graphgraph
Karma: 0  
Re:General Status??? - 2008/11/13 06:34 Yes, I believe the 914 controller is the same or very very similar to the Er1 from what I remember hearing, but I have not tried running any 914 SW on my bot to find out either way, as I mentioned before the 914 didnít offer much above and beyond what I have so I decided to wait on buying a 914 and stuck with poking away on the ER1.

And I agree that having the same SW platform is crucial for any real progress, but that goes hand in hand with the HW also, having different HW that may or may not have the same capabilities can serve to de-focus any effort unless very well managed, and given this would be a community effort, then you cannot manage such a beast.
  | | The administrator has disabled public write access.
c6jones720
User

Platinum Boarder
Posts: 350
graphgraph
Karma: 3  
Re:General Status??? - 2008/11/13 16:24 Hi, I just read through your code in Python - interesting. From my point of view it looks similar to the sort of thing we would have been expected to produce if we were using Brian or Player. It looks to me like there is already some sort of server running on the robot providing raw functionality and you program interfaces to that + strabo.

I did consider strabo at one stage but I didnt use it because it looked to me like
1- You had to have apriory knowledge of what the world looks like
2-Im sure it didnt support rotation of the robot in the z-plane.

It is an interesting project though. From what Ive seen in the code there appear to be a lot of similarities in the sensors etc. It probably would be possible to create some sort of program that just about provides similar functionality to both robots.

My only point would be isnt this exactly what MSRS and player are there for?
  | | The administrator has disabled public write access.
<< Start < Prev 1 2 3 4 5 Next > End >>