Department of Electrical Engineering
The Ohio State University
Winter Quarter, 2002
EE 682P Experimental Biomimicry for Distributed Control
Instructor: Prof. K. Passino, 416 Dreese Laboratory, passino [at] ece.osu.edu
Office Hours: Set up an appointment or stop by
Teaching Assistant: Not yet assigned.
Laboratory: Most likely Rm. 761 Dreese Laboratory (each team leader gets a key)
Prerequisites: EE Core, EE 582, Some with EE 551 (offered in Au and Sp).
General Focus: Biomimicry focuses on understanding and then emulating how biological organisms solve problems for the purpose of developing innovative solutions to challenging technological problems. Biomimicry can proceed according to a theoretical/mathematical agenda, or via experimentation in the laboratory with the construction of circuits, computers, and devices that are designed to emulate biological functionalities. Biomimicry has been used to develop a wide variety of current high technology systems (e.g., vision systems, speech processing, electronic noses, neural/fuzzy systems for control, genetic algorithms for design). The overall focus of this design project is to construct an experiment that emulates a biological system and to then identify potential industrial uses.
Specific Wi'02 Projects:
In this design class as a part of a team you will design, implement, and analyze the performance of one of the following:
- Distributed Sychronization of Electronic Fireflies and Chirping Birds: (a) One species of a "fire fly", upon night fall, will be blinking according to a natural frequency, but a whole group of thousands of them will be desychronized at the start of nightfall. After a few hours the fire flies (thousands of them) manage to sychronize their flashing without the assistance of a leader (or some method of orchestration), but only via local observations and adjustments in timing of flashes by individual flies. It is possible to design a collection of identical electronic fire flies that are able to communicate with their neighbors, and which have an on-board decision-making (control) process for adjusting flash timing. It can be demonstrated that a group of such electronic fire flies can, within a few seconds, sychronize their flashing. It is possible to analyze properties of the flashing dynamics, and support such analysis with simulations of the group of flies. Distributed sychronization is important for distributed computing, the internet, distributed networked groups of autonomous vehicles, and other applications; and (b) Some species of birds, when the are in listening range, will synchronize their chirping. This is similar to (a) above, but the focus is on emulating distributed sychronization via sound. To do this you can use a group of computers, each with an ability to produce a "chrip," listen to all others (or just neighbors?), and then adjust their chirping frequency. This problem is fundamentally different from the fire fly case since here the local decision-making strategy must use all other chirps that it hears, not knowing which "electronic bird" it came from. Again, distributed sychronization is important for distributed computing, the internet, distributed networked groups of autonomous vehicles, and other applications. The design project involves constructing an experiment to emululate biological distributed sychronization, and illustrate and analyze its performance/behavior.
- Distributed Attentional Systems: Many types of animals have an ability to "pay attention" to important objects or organisms (e.g., predators or prey) in their environment. This occurs via a dynamic shifting of focus in order to pay attention to the most important things, and thereby allocate cognitive resources to reacting to the environment. If there is a group of animals (e.g., "swarm") engaged in some activity (e.g., foraging or group predatory activities) then it is sometimes necessary to achieve a type of "distributed attention" where the group of organisms works together in order to maintain an accurate and up-to-date picture of their environment. In this project you will design an implement a circuit/computer that will emulate the shifting of attentional focus across multiple objects. Central to the problem of attentional systems is the problem of resource allocation which is a core problem in the area of distributed computing.
- Distributed Thermoregulation for an Electronic Beehive: A cluster or hive of bees has an ability to perform, without a leader guiding them, distributed thermoregulation of the "brood" (baby bees) to protect them in northern climates during cold weather. They do this by grouping around the brood, utilizing a biological temperature sensor, moving their flight muscles to generate heat (without flying), and by packing the crowd of bees and different densities around the brood (not too tight so as to overheat the brood, but not too sparse so as to allow dangerous temperature drops for the center of the group where the brood is held). In this project you will design a distributed spatial thermoregulation system that utilizes electronic bees that can individually let off heat, and which as a group can choose an appropriate density of packing, in order to maintain proper brood temperature. Distributed thermoregulation is important in a wide variety of industrial processes (e.g., glass manufacturing, chemical process control, etc.).
You will be required to provide a final presentation and demonstration of your project where the EE Faculty will all be invited.