Blog Post 5

Final Product:

Note** Video and pictures were taken before some
final changes were made but for the most part reflect what
the final product looked and functioned like

  

Hardware updates:
Since the last post I have printed two more fingers and added them to the hand. I decided to stick with the make-shift solution to replace the hinges for this prototype as I discussed in previous posts. They ended up working well as they had some flexibility to them so that the fingers could contract different amounts around the curvature of an object. I constructed a housing to hold the hand and the servos in place so that the servos could turn and apply a torque to the strings connected to each finger. All the servos were connected separately to the breadboard and wired up to power and digital pins for signaling. Each servo had to be calibrated to a starting position and then I tested to find the value of the desired contraction for each finger. Finally I printed a new ring to house the button/ control system for the prosthetic hand (shown in one of the videos above).

Software updates:
I implement a timer using millis() in order count the number of button presses during a given time interval. I also set up different cases for the hand which I called modes. I then implemented statements that would toggle the mode if the number of button clicks within the time interval was greater than or equal to the threshold value for switching. All of these vales were not hard coded as to make changes such as adding additional functions and changing the time interval possible. The two functions I coded for the demo were a clench function which simply contracted all of the fingers, and the second function displayed an "okay sign". These were simple as they were just meant to showcase the toggle ability of the hand. Overall, I was very happy with how the toggle function was working. It was pretty seamless and worked really all the time with minimal error It was easy to use and felt natural. I did have some issues integrating the servos on to the hand. The interrupts from the millis() timer seemed to interfere with the signaling to the servos. I solved this problem by adding some delays in the loops and powering the servos off of an external power supply.

Future iterations:
I do developing further versions of this prototype both on the physical quality of the prosthetic and the control side. Eventually, I would want to be able to produce a variety of custom prosthetic devices all with customizable control systems that could be selected by the user. I do plan on communicating with potential consumers in the target market to better determine what they would need in a functional prosthetic like this one. For the more immediate next steps, I would like to continue working on the control software to allow more seamless control to the device. I would want to develop more functions for the user to be able to select and more control options. I would eventually like to develop an interface through which users with or without programming experience could implement their own functions based on their specific needs and even customize the control. Another improvement would be making the controller wireless and housing the electronics inside the prosthetic itself. I hope to make all of the code and files open source and possible provide custom kits to those who would like all of the parts packaged together for them. Overall, I am very happy with the progress I have made and I think this prototype does act as a proof of concept for the idea and of course their is room for improvement. I learned a lot through pursuing this project and am excited to see where it takes me through further iterations and how I will be able to apply this knowledge to future projects.

Blog Post 4

Update:

I have continued to print the various parts for the prosthetic hand. I have solved some of the issues I was having in terms of parts failing mid-print. I have experimented more with trying the hinges including printing a mold and attempting to cast the hinges out of flexible silicone sealant. This did not work too well as the hand had to be printed as 70% of its intended size which resulted in the hinges/joints being too small to the point that they would get stuck in the mold. The silicone also does not have enough restoring force alone to return the fingers to their original position. I have decided to devote my time to finishing the rest of the device first, and, as I said in my previous post, The current make-shift method is sufficient. The mold, however, is pictured below.

I have printed a prototype for the ring controller. I ran into some problems scaling the part between applications, but eventually I overcame this. I am also figuring out how large the tolerance must be to fit all of the parts securely in the ring while not making it too bulky and uncomfortable to wear. I plan on printing the next iteration of the controller soon but the current one is pictured below.

Moulds and Ring Controller                                                          Prosthetic Hand in progress


I have decided that for the purposes of this process it is not necessary to implement wireless control between the controller and the hand. This is something that can definitely be done in later versions of the project; however, for the sake of competition, and taking into account that I would have to spend time learning about how to implement the wireless control, I have decided to hard wire the controller to the hand unless I have time at the end to do otherwise.

I have started to think about how the rest of the electronics including the micro-controller, the servos, and the battery pack ...etc  would be housed in the device. I have thought about placing them in some sort of wearable (possible a bracelet) on the controller hand or on the outside of the prosthetic temporarily because in later versions the technology could be integrated inside the hand design. I figured that since I am still in the prototyping phase, it makes sense to have the electronics easily accessible in case of failure or need for modification.

To Do:

I have decided to use a button as the primary sensor due to its low level of difficulty for the user and fairly reliable ability to not be activated by accident. I am currently designing the next ring controller to house the specific micro button that I have selected and plan on integrating them soon.
Once the controller is completed and the rest of the parts for the hand are printed, I still have to integrate the control system with the hand and servos into one system. Finally, I am still figuring out an interface that would allow the user to change the function of the hand easily. I have decided that the primary function will be a simple clasp and release of all the fingers simultaneously as this seems the most intuitively useful function. I have also thought about the idea of having a built in hardware toggle capability for the controller to switch functions without an external interface. This would allow for a faster switch for the user, even if it was just toggling between a few preselected functions that are more useful for the user. I am still confident that the fundamentals of this project will be completed on time.

 

Blog Post 3

Update:

I have begun 3d printing the prosthetic hang out of basic PLA and attaching the joints together. I ran into trouble creating the hinges as they are meant to be printed out of a flexible filament that would allow the fingers to clasp when the force is applied to the string, but snap back into place when the force is released. I have temporarily attached the fingers to stabilize them with a somewhat flexible material and have secured rubber bands on the back of each joint to restore the fingers to the open position. For the purposes of this project, I think this solution is more than satisfactory as the hand will still work. If I have more time in the end I may design hinges. The current progress is shown in the picture below.



I have prototyped a basic ring controller device using a rubber band and a button in order to test the practicality of the idea. It is comfortable enough to wear, and it does not inhibit mobility of the controlling hand.  The basic prototype for the controller is also shown in the image to the left.

I have setup some basic code to turn the servos that will pull on the tendons/strings that will contract the fingers. The amount of rotation for each servo will need to be calibrated precisely later after they are mounted and permanently connected.

I have been running into some trouble with some of the parts failing during printing and am trying to generate better support structures and increase contact with the plate in order to compensate.


Still to do:

Along with printing the remaining parts for the hand, I also have to design and print the ring controller in a more finalized fashion. I am still thinking about which sensor would work best in this application to control the hand. Some limitations are that the sensor must be very small such that it can fit in a ring/controller between the fingers. Another is that the sensor must be reliable and easy to operate for any user. For example, I have thought about using an FSR (Force Sensitive Resistor); however, I am worried that it the threshold setting for force applied may be too specific and that this value could also differ depending on the user. Once the controller is completed, I have to integrate it with the hand and servos into one system. Finally, I will work on an interface that would allow the user to change the function of the hand easily. I am still confident that the fundamentals of this project will be completed on time.

Blog Post 2

Problem:

Prosthetic technology is an essential component in many people's lives, and while the technology continues to improve,  it is still not available to many. Most basic prosthetics can cost thousands of dollars and cannot perform much function for the user. I plan on developing a basic prosthetic that can perform a few basic functions and that will be available to everyone.

Solution:

My idea is that while expensive, more advanced prosthetics may work better, the technology is still primitive and limited by monetary constraints. For many, having a less advanced, but still functional prosthetic is much better than no prosthetic at all. In addition, many high end prosthetics still require the user to spend time "learning" how to use the technology. Even prosthetics that sense the movements of tendons or electric impulses in the body have a learning curve before the user can use the limb functionally. While this approach seems intuitive, it seems clear to me that if the user must "re-learn" how to use the prosthetic, they can also "learn" how to operate a prosthetic that does not rely on these advanced technology.

For the purposes of this project, I have decided to focus on a prosthetic hand/arm; however, I do believe this idea can be applied to many other prosthetics. I plan on using 3d printable materials to build the major components. This would allow the prosthetic to be fully customizable for each user and, in theory, the user would be able to print out and assemble the parts at home. Once I have assembled the robotic hand, I plan on working mostly on the control aspect and developing new ways for the hand to be controlled that do not limit the users motion in any other sense. Since the hand is wirelessly controlled, the select functions it can perform can be updated wirelessly depending on what activity the user is doing; moreover, different users might have different needs for what the prosthetic needs to be able to do, and they will be able to select the corresponding functions. By designing the prosthetic to be more customizable in the software, the hardware can be more universal and therefore less expensive. Overall, this project aims toward designing a cost effective control system for a prosthetic hand that would allow the technology to be available to everyone.