Difference between revisions of "ProjectSpeedDating2"

Revision as of 19:27, 26 February 2014

This is the dump that came out of my text editor. Please feel free to clean it up! ~~

Simprints

Alexandra

- Wireless fingerprint reader. Lots of mobile health applications are being built and health workers in Africa have mobile phones typically. However, identifying people is very difficult - they often don’t know DOB, the names are al the same, etc.

The idea is to build a fingerprint scanner that can work over bluetooth. Nothing on the market is designed to work with mobiles, or if they are they are very expensive. It should be

- Rugged
- Waterproof
- Shock proof
- Cheap

There’s a prototype already but they are looking for people to fill various roles.

- Hardware
- Embedded software developers
- Android developers (need to port a matching algorithm to mobile)
- Cloud systems developers (network, server - initially they want to setup the matching algorithm on the server)
- Solidworks developers

alexandra@simprints.com

64skies

Pamela Wesson (pam@fantasia.uk.com)

Based on the experience of an installation by an artist where the sky was seen through a white square in the roof (my writeup doesn’t do this justice)

Art website of 64 skies from around the world. 64 webcams showing skies from all over the world. ability to zoom in and out.

This is strictly an art piece :)

Suggestions - make a kit that requires a webcam, rPi, not much else! Auction the squares?

Cellular AutomaPONG

Danny (danny.garden@gmail.com)

An evolution of a 1D pong (simple example) game based around a generic cellular automaton unit.

The unit (equivalent of each LED in the simple example above:

• A bright RGB led to illuminate it. The unit should have a diffuse plastic case to allow it the glow nicely.
• A simple state machine that can take an incoming "ball" and relay it on, or perhaps change the direction using game modes.
• Needs some kind of self aligning magnet setup to allow multiple units to be chained to make large networks.
• An input method to change the "mode" the unit is in e.g. relay, splitter, wall, multiplier etc.
• Needs to be cheap, we need like 100 of them to make an interesting game. If the prototype units aren't cheap we should at least have a route to cost down.

As well as the generic unit, there may need to be a Marshalling unit that keeps an eye on the game (watching for resonance behaviours and perturbing them to make sure the game flows freely, keeping score, initiating "special" game events etc). I'm thinking a RPi with a web interface or something along those lines. Also the Marshalling unit needs to have an accurate model of the active network of units, for this it might be worth implementing some kind of network search algorithm to "map" the network topology.

The aim is to build this system for the STEM Ambassador program in Cambridge(link) (or anyone else who might be interested) so that they can deploy it as an interactive game during events, for example the upcoming Cambridge Science Fair.

Live monitoring system (LMS)

Stuart Wilson

Non-profits trying to slow down the rate of illegal logging.

Corruption is rife, so human-based systems are difficult to make work.

Can we use vehicle tracking systems to track the large vehicles that are essential for logging. Feed the data from the vehicle tracking system into a web-based frrontent. Overlay the vehicle activity over legal areas. Do the analysis to generate suspicious or illegal activity.

Allows you to better allocate your scarce human resources to the right place!

Lots of development has happened already and there’s a small amount of funding to make this work and to demonstrate something to the Congolese government.

Required: - Software engineers for the web-side - People interested in the overall objective

Sub-Division

John - jon.aas@ctinnovations.co.uk

Engineer by background who has worked for many years on subdividing a shape into cubes.

This is a practical use of 3D printing to demonstrate how subdivision can be done. There is an algorithm to determine how the subdivision should happen.

Currently the input/analysis stage is very time consuming. The aim of the project is to do the simulation and the input to the simulation very fast. Software required to allow a user to manipulate the properties of the shape in 3D - the software would need to direct/control some very powerful machines (eg multiCPU and GPGPU).

What’s required? - People interested in the problem and into software - People to work with them to make it happen

Storage Robot for Makespace

Mark Tillotson

A machine for storing the members’ boxes in the 4ft high spaces under the kitchen and cafe. The rooms are about 2x5m but not practical for humans to go in.

The idea is to make a machine that can collect boxes for people!

What’s done: Mark is working on the mechanical solution and can pickup boxes. Is planning on the gantry side What needs to be done: Software, help with mounting etc Other skills sought: Someone to think about whether this is commercializable!

Softprinter

Tom Myers & Alex McFadden

the dream of 3D printing has always been to move manufacturing back into our own homes so we can print something as and when we want it. Despite the years of development in 3d printing we’re still stuck with yoda statues and shot glasses

So, let’s make something that makes things more useful! Let’s make clothes. Download a file offline, customise it to your body shape, print it in the machine. Things will fit far better than things off the oeg.

Clear technological idea about what should be done. Old knitting machines needed a human operator, but we can mechanise this and actuate all parts of the machine that need it.

Currently doing hardware prototyping. They’re in the process of building the carriage. Not yet at the electronics.

Potential to disrupt the fashion industry. Short supply chain, custom made clothes.

What’s required - Hardware people - physical prototyping comes first. Some software people to help build things that are controllable.

tm476@cam.cam.ac.uk