Ultrasonic Spotlight Tracker by Paul Cheng Po Chen by Jimmy Chin-Hao Chen
Introduction
A spotlight that follows you on its own!
The ultrasonic spotlight tracker is a system that uses a wireless beacon to track a target¡¦s location using both RF signals and ultrasound waves. It then drives a light source to point at the location of the target.
High Level Design
The motivation for an automatic spotlight tracker is to replace expensive manual labor in operating spotlights in theatrical settings. There are also numerous other applications where knowing the real-time location of a remote target can be useful.
In this project, we limit our remote unit¡¦s degree of freedom to two dimensions to prove the concept, although it can be very easily extended to three dimensions. To locate the target, we need to measure the distance of the target from at least two known points. The distances are measured by counting the time it takes for sound to travel from the target to the base station and multiplying it by the speed of sound waves. The speed of sound in air is about 334 meters per second. The 16 MHz microcontroller can count time intervals on the order of 0.0625 microseconds, which gives a theoretical accuracy of 0.02 mm.
There are many tricky parts in this project, particularly the RF ground plane and transducer noise problem. This makes the system performance less stable than the ideal. We are happy to see that the tracker responds to the moving target, albeit slowly. The LCD shows the numeric results that is quite reasonable and this is the first step towards making it a really usable system. There are many things that we can improve upon, particularly the ultrasonic parts. We think that maybe the transducers themselves are of low quality since they are only $1 a piece, compared to some other ones which are easily $20 a piece. The sound frequency may also limit the stability of the system, so moving up to 40 kHz may improve the accuracy of the timing. RF noise is getting coupled everywhere on our circuits when there are long wires. This could be improved by having a low pass filter just before the input to the external interrupt pin on the microcontroller. Also, a more elaborate displaying scheme would have been nice where we can show graphically where the target is relative to the base, like a radar system.
Link: http://instruct1.cit.cornell.edu/courses/e...Jimmy/index.htm
Introduction
A spotlight that follows you on its own!
The ultrasonic spotlight tracker is a system that uses a wireless beacon to track a target¡¦s location using both RF signals and ultrasound waves. It then drives a light source to point at the location of the target.
High Level Design
The motivation for an automatic spotlight tracker is to replace expensive manual labor in operating spotlights in theatrical settings. There are also numerous other applications where knowing the real-time location of a remote target can be useful.
In this project, we limit our remote unit¡¦s degree of freedom to two dimensions to prove the concept, although it can be very easily extended to three dimensions. To locate the target, we need to measure the distance of the target from at least two known points. The distances are measured by counting the time it takes for sound to travel from the target to the base station and multiplying it by the speed of sound waves. The speed of sound in air is about 334 meters per second. The 16 MHz microcontroller can count time intervals on the order of 0.0625 microseconds, which gives a theoretical accuracy of 0.02 mm.
There are many tricky parts in this project, particularly the RF ground plane and transducer noise problem. This makes the system performance less stable than the ideal. We are happy to see that the tracker responds to the moving target, albeit slowly. The LCD shows the numeric results that is quite reasonable and this is the first step towards making it a really usable system. There are many things that we can improve upon, particularly the ultrasonic parts. We think that maybe the transducers themselves are of low quality since they are only $1 a piece, compared to some other ones which are easily $20 a piece. The sound frequency may also limit the stability of the system, so moving up to 40 kHz may improve the accuracy of the timing. RF noise is getting coupled everywhere on our circuits when there are long wires. This could be improved by having a low pass filter just before the input to the external interrupt pin on the microcontroller. Also, a more elaborate displaying scheme would have been nice where we can show graphically where the target is relative to the base, like a radar system.
Link: http://instruct1.cit.cornell.edu/courses/e...Jimmy/index.htm