Current and Potential Projects
Energy-related projects:
- Sunpower 4000 watt inverter,
wireless monitor, perhaps with web interface.
- Solar cell shade tracking and dynamic panel isolation.
Dynamic reconfiguration of panels for I-V control.
- MPPT using data telemetered from solar panel.
-
Passive solar panel cooling for higher efficiency
in hot climates: computation/simulation/testing
-
Solar cell I-V, full power simulator.
- Solar hot water energy monitor, rs232 to RESOL controller
+ custom interface to flow meter.
- Solar hot water/secondary heater control and circulator
- Remote solar water pump controller -- tanks 100 meters from pump
- Development of tracker/locator modules for attachment to animals.
The requirements of low weight, long battery life, sparse and transient connectivity make this a challanging wireless project. The project is sponsored by the Lab of Ornitholgy. Three documents are available which show examples of trackers, (Microwave Week, ISBE, RF Migration talk). Contact is David Winkler.
- Bird Nest Box instrumentation. (contact is David Winkler or Jim Moore)
Projects for Prof Winkler's research at the Ponds. From an educational point of view we are providing opportunities for collaboration between engineers and biologists. From the engineering prospective we need expertise in the following areas to support the biology students:
- Network infrastructure
- Data management
We need a great deal more work in data management and curation, integrating inputs from a local weather station together with biological data on the birds in the boxes. From the biology side, we are moving to doing everything from data manipulation to statistical analysis and visualization with R. And it would be great to have a student interested in building the applications that gather together the appropriate data and run the analyses, some of them of a standard sort and others of a more exploratory nature.
- Image processing
wrote scripts that manage the inflow of files from the cameras, putting them in smaller day- and hour-specific folders very soon after they arrive. But if we go with timer-triggered images, we will definitely need to do some post-processing on the images to identify those images in which the state of the nest image changed.
- Electro-mechanical design
In the area of electro-mechanical we would like a remotely controlled "wig-wag" which will close off the a nest box entrance so the birds can be captured for various purposes. The emphasis here would be a simple, yet elegant design that would not be offensive to the birds! Swallows are sensitive to changes around the nest box, especially if it looks like a snake!
- RFID technology
Robert Johnson, is working on an RFID system that will be connected to our intranet and sense the passage of the swallows through the nest box opening. A related project to this would be a design that would use the same technology to detect swallows (with implanted tag) if they landed on a perch.
- Peltier Heating/cooling system on the nest boxes
- Electroporator
Circuit and controller to inject materials into single cells. Contact is Joe Fetcho.
- Soccer ACL acceleration/twist/bending recording for Bob Nafis:
Bob Nafis is a member of Prof Yingxin Gao's team to study "Why female Soccer players suffer 3 to 8 times the rate of exposure to knee ACL injuries compared to male soccer players.. We have a somewhatunique system for measuring the ACL strength and shape using
ultrasound. We are also attempting to instrument a member(s) of theCornell womans soccer team and record and create histograms of the shock across the knee during daily practice to compare with weeklymeasurements of the cross section/position of the ACL. It is our
intention to insert a 3 axis accelerometer in the heel of thesoccer boot,with wireless transmission to a fanny pack where a "to
be designed histogram algorithym" will record the frequency ofintervals of the multiplexed accelerometer signals. . We intend to
install similar accelerometers above and below the knee joint, and in a shin guard. We also intend to measure the knee bend angle with
a magnetic sensor that Philips has created with a digital output. We have yet to figure out how to measure the twist between femur
and tibia. Packaging and antenna is a critical factor. Konix has donated the accelerometers and demo boards, and we have purchased 3
Philips chips. The Freescale processor looks interesting and depending on whether to use Bluetooth or Zigbee as the wireless media. The fannypack receiver will need up to 10 channels and histogram storage for a week.(12 hours of field practice) .
Microcontroller Projects
- VGA generation code running on Mega644 (and another AVGA)
- Atmel UDP -- use microchip ENC28J60 -- see ECE4760 project
- Combine Mega and FPAA (e.g.www.anadigm.com)
- USB Peripheral Controller with SPI Interface -- maxim-ic.com MAX3421
- Near field communication and using a quad inverter as a tranceiver at 13 MHz.
- LED touch sensor and more
- Floating Point Coprocessor for Mega32 (micromegacorp.com) This would extend Kenny Lo's project with faster hardware.
- XBee development
- Toys for human development study
Object 1 --Target Toy Broad Description:
A toy that when shaken from side to side, like a rattle, produces either a sound or lights. When rolled, the toy produces the opposite, either a sound or lights. It will be used in experiments concerning function and social understanding with young toddlers, aged 17 months-3 years old. Dimensions: About 9-10" long, about 1" circumference in the middle. The circumference enables young children to handle the toy. The object needs to be able to roll. So when propelled on a table or floor, it rolls. The exact shape, for instance whether the ends are of wider circumference than the middle, is up to you. Color: Any color or number of colors is fine, just notably distinguishable from the color of Object 2. I will be adding decorative things to it, but the base is not of particular significance to me. Texture: The object will be handled by 1-2 year olds, who do attempt to eat essentially anything, so it does need to be resistant to pressure of that sort. Also, it should be solid, ie. not bend or droop. While it shouldn't have any ridges that would break skin, it does not need to be perfectly smooth or made of any particular material. Lights: Lights on different sides of the toy so that when the lights are activated, they can be seen from more than one side or perspective. While having at least four lights spread apart across the surface of the toy would enable them to be perceived well from different perspectives, the exact number and type are up to you. When activated, the lights should be easily perceivable in a well lit room from 3-4' away from the toy. The particular size and shape of the lights, or whether they descend from the surface or are right at the surface, is free for innovation. Also, whether they are of different color shades or all of the same color, is up to you. Whether the lights blink or not, and, if so, at what rate, is free for innovation. The lights could activate one after the other, or all at once. Sound: The sound produced by the toy should be "pleasant." Less vaguely, it should be loud enough to be heard well within five feet, so perhaps in the 60-70 dB range is most suitable. Voices or music would be fine, as long as there isn't any bad language. Other noises that are pleasant, like animals or what not, would be great. Activations: Young kids have a tough time with motor skills, so I was thinking of being able to roll it five inches and have it activated within that distance; this would clearly allow for having the tube revolve more than once. Soon after it stops rolling, within 5 s for instance, the sound or light should stop. In terms of shaking, I was thinking of holding it in the middle and shaking it by twisting the wrist, like a rattle. Within the toy, something could be forced by that movement to trigger the light or sound. The kids will easily be able to perform such an action repeatedly and quickly, moving the ends up and down 1 1/2"; whether it takes each side moving up and down once or twice or three times to activate it is not a huge concern to me, but within three times is a good boundary point. Additional Notes: Each time the toy is activated, the pattern of lights and sounds should be the same, so that the child notices that when they do x, y results. The lights and sounds activated should deactivate soon after the rolling or shaking ends, for instance within 5 s or so. Standard batteries are not at all objectionable, but other sources of power are fine as well.
Object 2 -- Control Toy Broad Description:
A rectangular toy that when bent, produces lights or sounds. When tapped twice on one end, it produces the opposite, either sound or lights. It will be used in experiments concerning function and social understanding with young toddlers, aged 17 months-3 years old. Dimensions: 9-10" in length, 2" wide, 1" thick. The object needs to bend easily at around half way. Color: Any color or number of colors is fine, just notably distinguishable from the color of Object 2. I will be adding decorative things to it, but the base is not of particular significance to me. Texture: The object will be handled by 1-2 year olds, who do attempt to eat essentially anything, so it does need to be resistant to pressure of that sort. While it shouldn't have any ridges that would break skin, it does not need to be perfectly smooth or made of any particular material. One of the ends should activate either lights or sounds when tapped twice; this could be in the form of a button. If it is a button, then it just needs to be Lights: The lights should be of a different color and, if blinking, do so at a notably different rate than the first toy. Lights on different sides of the toy so that when the lights are activated, they can be seen from more than one side or perspective. While having at least four lights spread apart across the surface of the toy would enable them to be perceived well from different perspectives, the exact number and type are up to you. When activated, the lights should be easily perceivable in a well lit room from 3-4' away from the toy. The particular size and shape of the lights, or whether they descend from the surface or are right at the surface, is free for innovation. Also, whether they are of different color shades or all of the same color, is up to you. Whether the lights blink or not, and, if so, at what rate, is free for innovation. The lights could activate one after the other, or all at once. Sound: The sound should be notably different than that of the 1st toy. The sound produced by the toy should be "pleasant." Less vaguely, it should be loud enough to be heard well within five feet, so perhaps in the 60-70 dB range is most suitable. Voices or music would be fine, as long as there isn't any bad language. Other noises that are pleasant, like animals or what not, would be great. Activations: The tapping activation could be in the form of a button, though I would prefer that the button not extend beyond the surface of the end. Young kids have a tough time with motor skills, so I was thinking of being able to tap the entire end; it could just be a large proportion of the end, though. It should be about as sensitive as a computer key. The toy does not need to bend completely into itself; to about a 90 degree angle would be great. Whether it does so to one side or two is up to you. In terms of sensitivity, when held by one end in the air, the toy should not bend of itself and hence activate itself, rather, it should take human pressure to bend it. But again, given that its for small children, it should not be more difficult than bending a laptop screen. I would prefer that it activate just on being bended, and not on being straightened back. Additional Notes: Each time the toy is activated, the pattern of lights and sounds should be the same, so that the child notices that when they do x, y results. The lights and sounds activated should deactivate soon after the bending or tapping ends, for instance within 5 s or so. Standard batteries are not at all objectionable, but other sources of power are fine as well.