Noise and vibration can generate inaccurate readings in precision systems such as planetarium astronomy equipment, parabolic antennas and laser based guidance systems that rely on precise pointing to receive or transmit signals. The sources of this problem, commonly called jitter, can range from motors used to position the pointing system to ambient acoustic sources such as simple heating, ventilation and air-conditioning systems or in some cases, noise due to automobiles and aircraft.
Did you know?
The Hubble Space Telescope is equipped with an Observatory Monitoring System that helps astronomers using the telescope determine how much jitter variation occurred during a science observation.
Since ambient sources of noise and vibration cannot simply be eliminated, researchers are developing ways to reduce their impact on sensitive engineering systems. Mechanical engineering Professor Robert Clark is developing real time, adaptive control approaches to compensate for jitter in optical systems. Clark uses optical sensors, accelerometers and microphones to measure jitter and the corresponding sources responsible for this undesired effect. From an initial array of sensors, Clarks research group has developed tools to identify an optimal subset of sensors that can be used to provide a measure of all independent sources of jitter. A weighted-array of these sensors is then used to reduce the complexity of the feedback control system required to minimize jitter in the optical path.
Clark has demonstrated that by using active control, he can identify lesser sources that contribute to jitter, but are masked by the dominant source in the initial measured response. This provides a mechanism for boot-strapping the control system to greater levels of performance and thus finer control of jitter.
Optimizing the performance of precision systems by controlling jitter is a bit like peeling back the layers of an onion. Active feedback control is an effective way to reveal the underlying sources of jitter disturbance and customize the necessary sensors required to further improve performance.
Robert Clark Website: http://www.mems.duke.edu/faculty/clark/index.php