Cheap Alternatives to Flex Sensors for Gesture Glove Project

Hey Aniket_kumar_raj ,

I've done a little edit to your message that might get a bit more input for you, and moved it to a forum more suited for your question - hopefully you'll receive some suggestions 

I think velostat probably can work, maybe with copper tape or aluminium foil strips and use it in a voltage divider circuit with resistors, and read the analogue values with an arduino, it won't be as accurate as flex sensors, but it could work out. The trick would be to compensate in code. Or maybe you can get some potentiometers at the finger joints, or maybe a sliding mechanism with string and a potentiometer. A few options there.

I can't see how you can realistically do anything if the budget is so tight, and you've got 19 days to order, prototype, and prepare a demo. This is called leaving it too late (you and your team have only yourselves to blame; don't think you should entirely blame the team, since you're involved too).

Instead of five sensors, you could reduce effort and cost by fivefold by converting the project to a gesture mitt that senses only one movement.

You didn't state the aim, but if it were me on such a timeframe, I'd move the goal and detect different objects rather than grips, or detect grips by picking up different objects. For instance, conductors on the forefinger and thumb, each connected to one end of an ohmmeter, or to a microcontroller ADC. Then, have cubes with two conductive surfaces on opposite sides, wired internally with a known-value resistor. When you pick up an object, the circuit completes, and the ohmmeter will register a measurement. Cubes with different resistors will all be distinguishable from one another.

You could also experiment with trying to make cubes that vary resistance the harder they are pressed, but that could be a phase two.

shabaz said:

I'd move the goal and detect different objects rather than grips, or detect grips by picking up different objects. For instance, conductors on the forefinger and thumb, each connected to one end of an ohmmeter, or to a microcontroller ADC. Then, have cubes with two conductive surfaces on opposite sides,

I'm thinking "what would be the utmost minimal viable product?" in a project like this, and to me that would be "switches stuck to a glove with wires" and then identifying which switches they are, then you change to it being 'gesture' based at the touch points (index finger to thumb, for example, or little finger to thumb) and if two switches are pressed together at the same time, you have a proof of concept that requires very little to work, maybe some wires and a light bulb to say "this represents a circuit, with budget this could give a signal" and then you're talking about what you could do.

Difficult to determine without knowing what parts are available for sure.

That's a good idea; simpler and provides something easy to demo, and unlikely to go wrong on the day.

Bro, school exhibition announcement was done on Monday, that's why we are actually on time for the school but in real scenario we are not, by the way this is extremely good idea, but cost is not that tight, I remember once tesla coil was made and it was actually mind-blowing for judges, I want to tell that judges are too dumb to understand these smart tech things (don't mind) ,they only make things win which look like magic , even if low effort has been given. And its easy to make

You need a project plan, and at some stage you need to have completed some research, then come up with a potential design that won't stretch your entire plan beyond its end date (i.e. keep perspective in your goal/objectives), place your parts orders, build your prototype, experiment with it, and iterate if needed, perhaps in a phase 2, but demo what was successful.
You can't go from zero to hero without some plan and experimentation.

Case in point: I experimented with making my gloves touch-capable over winter. No point if the project stretched beyond winter, so my project deadline was clear. I knew I had to order quickly and try to experiment fast. My first prototype didn't work. The second one did, to an extent, it was usable although not great. Still, I met the self-imposed timeline. If I'd shot for 'killer' and 'next-level' it would probably be next winter by the time I'd have completed, and sometimes having something to use or demo fast that you can iterate later, is better than having nothing to demonstrate for a year.

If using Velostat, them perhaps switch from gloves to socks.

Aim for sensor pad areas below the heel, the ball, and the toe where you typically have control of pressure over. Typically there will be two feet, so six analogue sensor inputs.

Then come up with multiple interesting use cases for a foot-based controller. 

Add sound and light to attract attention to your project.

Bro u got the winning point I was trying to say, add something to make it attract attention. Love the idea , but adding it under foot could be paining,I mean I can't move my foot too much up and down continuously, Any Ideas that you have that's very unrelated but attracts attention

Perhaps try standing - easier to change pressure under different parts of your foot when standing. 

Connect it to lengths of pixel LEDs and/or a synth as a wearable and start walking/dancing.

Perhaps think 'Big' with pressure pads and visual and audible feedback.

https://www.youtube.com/watch?v=CF7-rz9nIn4

Could create a two-player game paddle controller and connect to RetroPie. People like interacting with stuff.

Turn it into a Wii Fit type balance board controller and use the XY data to create generative art.

Recreate a science type experiment showing how the properties of Velostat could be used in STEM subjects.