| Looking for your input. |
|
|
1 2
|
| 1 2
|
|
10/13/2009 13:43
mahoon 
|
10/13/2009 13:43
mahoon
|
Looking for your input.
Hello All.
I have been a Dupuytren's sufferer here in the UK for a few years now. I had surgery on my right hand back in 2006 which managed to straighten my third finger but still both my little fingers are bent almost to the palm. I am not seeking further surgery at the moment (it wasn't a barrel of laughs back in 2006 as I'm sure others will understand...).
I am in the final year of my BSc (Hons) degree in Computing and for my final year-long project I am investigating the use of non-traditional Human-Computer Interface devices (traditional being mice, joysticks and keyboards) to aid people with hand problems (Dupuytren's, RSI, etc) interface more effectively with both the computer and the real-world.
Currently I am looking at the use of a soft, cloth glove with accelerometer attached (enables the glove to report its position in X,Y and Z planes - so it knows when the hand wearing the glove moves forwards, backwards, left, right and up and down) with Flexometers to measure the bend of each finger.
(cont...)
|
|
|
|
|
10/13/2009 13:43
mahoon
|
10/13/2009 13:43
mahoon
|
Re: Looking for your input.
(...cont)
My initial thoughts are to devise a self-calibration system whereby the user puts on the glove and the system detects the amount of movement available to the user (up, down etc) and which, if any, fingers can be bent. After calibration, the system then recognises where the hand is lacking 'normal' functionality and compensates for it. By that I mean if someone can only move their index finger, the system recognises this and maps the movement of the index finger to all other fingers. The idea is to then create a full-working 3D model of the compensated hand which the user can use to interact with in a 3D environment.
The thinking behind this approach is that the glove wearer can then use the 3D hand to interact with objects in the 3D environment that they might not be able to do in the real world. Although this might not have any 'real world' value - it is an academic project, after all - I was wondering if anyone had any thoughts they'd like to share with me on the project? Suggestions, views and opinions are all welcome.
I am, however, constrained by certain things: the glove must be used and there must be a 3D environment to interact with (so things like on-screen keyboards and the like are out).
Best Regards and looking forward to your input.
|
|
|
|
|
10/13/2009 14:56
jimh 
|
10/13/2009 14:56
jimh
|
Re: Looking for your input.
So a 'virtual' finger is being created as a substitute for a real, but non-functional, finger? How would the user control a virtual finger independent of the others, if he can't move the corresponding real finger?
What did the surgeon say about the possuibilty of straightening your 5th fingers? I wouldn't give up on those yet.
Edited 10/13/09 17:58
|
|
|
|
|
10/14/2009 02:33
Issleib2not registered
|
10/14/2009 02:33
Issleib2not registered
|
Re: Looking for your input.
I/m away from home and tired but I am so excited about what you are doing!!!! Before I was retired I worked with children with physical handicaps. What you are workng on could be a breakthrough for them. I won't have computer accese again until sat or sun. I will post again then. Please keep checkng the web site untie I can get home and post when I'm awake
Colleen MD
|
|
|
|
|
10/14/2009 02:48
flojo 
|
10/14/2009 02:48
flojo
|
Re: Looking for your input.
I taught students with special needs and now work with teachers getting their special education credentials. I will be interested in hearing about how your project develops.
Flora
|
|
|
|
|
10/14/2009 07:59
mahoon
|
10/14/2009 07:59
mahoon
|
Re: Looking for your input.
Quote:
So a 'virtual' finger is being created as a substitute for a real, but non-functional, finger? How would the user control a virtual finger independent of the others, if he can't move the corresponding real finger?
If I understand what you're saying, this is the idea of the 'compensation' system. If a real finger can't be moved, another input method (still using the glove) would be devised to compensate for it - a push forward and quick flick up of the hand could be read as a 'grasp' command, for example. Similarly, perhaps a 'modifier' can be applied - whereby the actions of the glove become context sensitive, for example by selecting an on-screen button the actions of the glove can be changed depending on what's required - forward and backwards movements could be translated into first finger bend and flex actions, up and down to second finger, pitch and yaw to third and so on. These are only possibilites however, I will have to see what's achievable and what makes sense as a realistic interface.
Thanks for your input though - this is the just the kind of stuff I'm looking for.
Quote:
What did the surgeon say about the possuibilty of straightening your 5th fingers? I wouldn't give up on those yet.
The surgeon came to see me in recovery and said (I remember this well...) "Well, I'm afraid that didn't go as well as I'd hoped...". I haven't scheduled any further appointments to see him yet...
|
|
|
|
|
11/22/2009 21:07
mahoon
|
11/22/2009 21:07
mahoon
|
Re: Looking for your input.
To give this topic a bump and because I thought someone might be interested.
I am required (in order to pass my exams) to keep a log of the development of my application.
This is the current state of the project. The log is below. WARNING - it's intented to be read by a programming-literate audience - but I don't think it's overly technical and if people say 'Stop posting this trash, we're not interested', I will. Until then:
Project Log
Project Title: Virtual Hand Assistant
2.10.09
Decided on Virtual Hand Assistant project after some discussion with course leader
9.10.09
Downloaded appropriate software from DGTech (dataglove manufacturers- the college have provided the glove, I just need the latest software). Attempt to install Vhand but missing both USB connector and Bluetooth Dongle.
16.10.09
Got bluetooth dongle from college. Installed bluetooth dongle, installed software and managed to get computer to see dataglove. Can open the calibration screen that is supplied with the glove software and all areas seem to be working accurately. Attempted to explore the example given with the glove's Software Development Kit, (SDK) but as it's all commented in Italian so I am struggling to comprehend what's going on.
17.10.09
Installed Microsoft's XNA GameStudio and Visual C# Express (required to run XNA GameStudio) as this might be the software I use to create the 3D environment for the glove to operate in. First impressions are not good however, as although the software seems to support a variety of input devices – guitars, mice, joysticks, gamepads and so on, none of them seem to support an input device with a z axis – which is crucial for proper glove input. Will have to further search the documentation to find if this is definitely the case. Suspect I just don't really know what I'm looking at or for...
27.10.09
Spent a thankless few hours attempting to compile/run some XNA demos. Had to reinstall Visual Studio 8 and XNA again before I could get them to open. They proved to be unhelpful.
|
|
|
|
|
11/22/2009 21:09
mahoon
|
11/22/2009 21:09
mahoon
|
Re: Looking for your input.
30.10.09
More time spent searching for HID/Bluetooth controllers. Nothing found. Seem I may have to investigate the Windows HID API – looks terribly complicated though.
5.11.09
Have gone back to using Dark Basic as I've located a program that uses the Windows Kernel32.dll to read serial ports. No luck in getting the program to work however.
15.11.09
Had an epiphany and realised I needed to send an instruction to the hand to begin transmitting data! This was simply the character 's' ('e' ends the sampling).
17.11.09
Can now see data streaming in from the glove. Currently looking at the values of bytes 3 and 4 –which are used in the glove's SDK example to read the acceleration axis x (data is sent constantly in a 20 byte package from the glove to the pc, describing the amount of bend from the flexometers and its movements in 3D space thanks to the accelerometers) . Next step is to ensure I'm using the right bytes and that they're being combined in the right way to get a meaningful figure I can use for the glove's x axis acceleration
20.11.09
Great progress.
Couldn't think of anything that proved effective. I decided to create the 3D hand model in Dark Basic. Went extremely well and can now bend all fingers with keypresses. Seems to only take 4 at a time though – have tried pressing five keys simultaneously and it won't have it. Next task is to translate the keypresses into finger-bend readings.
21.11.09
Still goes well, can translate the bends to movements on screen. Need to double-check the finger bending routines though – they aren't working as well as I'd like.
Was indeed bytes 3 and 4 that determine the amount of bend for finger 1 (thumb). Have played around with the numbers to reduce the readings from 125 to just + or – 5 when at rest. Just makes it easier to see what's going on really.
22.11.09
Decided to put what I could into functions within the main body of the code. This is because I intend to reuse the code during set up and calibration etc. Took a long while to do – learned something about global and local variables I think!
Managed to get the majority functioned(??sp) and now can call the main reading of the bytes and moving of the hand in reaction to those bytes separately if required – so can use the hand in different situations now – like calibration for example.
Decided to have a look at the accelerometers. Reduced their readings by 1000 to make them readable (could probly affect same result with different bitwise operations). Tried something to read the values and make a + number roll the hand right, and – number roll it left. It worked first time and I've even managed to build a 'deadzone' in so that the hand rolling left and right is not so sensitive.
(Note: I intend to continue updating this progress file as and when I think I have made enough progress).
|
|
|
|
|
11/23/2009 15:48
jimh
|
11/23/2009 15:48
jimh
|
Re: Looking for your input.
I'm a software guy too and this is interesting to me - I'm leaving town for a week and will read this in detail when I get back.
|
|
|
|
|
|
|
|
1 2
|
| 1 2
|
403 users online
Login
