3D motion tracker
kelvin james
Posts: 531
Scrapped a bunch of other ideas to work on this, to work within my budget.
Game Plan
I am looking to measure movement within the 3D plane. The device would be attached to a moving object and would be be able to track the exact movement in units against a fixed 3 dimensional area. The sample rate does not have to be fast, probably in the 10th /sec range, so it leaves quite a bit of execution time. Readings will be measured in probably 2 inch increments, which should be reasonable to monitor accurately. Looking at using accelerometer / gyro(yaw) combo X 2, that will give 6 axis of movement, the 2 boards would be at 90 degrees to one another to accomplish this. Why yaw? To be able to rotate within the 3D plane on the the 2 axis. I will probably have to put the input data into a matrice calculation, to determine the combined values against the set 3D area.
It will be wired at first to get it working, and will be made wireless afterwards, if it can be done.
Questions:
Anyone here ever deal with matrices? I have done it before on an old PC, it took some time to calculate, but it was a slow processor.
Would you know the speed required to do this? It is a pretty complex calculation.
Accererometer / gyro offset. The real world is not level, can an offset be programmed to the sensors to calibrate a zero plane?
More questions to follow.
kelvin
Guess i should of said 'matrix' , instead of matrice
Post Edited (kelvin james) : 3/7/2005 3:59:12 AM GMT
Game Plan
I am looking to measure movement within the 3D plane. The device would be attached to a moving object and would be be able to track the exact movement in units against a fixed 3 dimensional area. The sample rate does not have to be fast, probably in the 10th /sec range, so it leaves quite a bit of execution time. Readings will be measured in probably 2 inch increments, which should be reasonable to monitor accurately. Looking at using accelerometer / gyro(yaw) combo X 2, that will give 6 axis of movement, the 2 boards would be at 90 degrees to one another to accomplish this. Why yaw? To be able to rotate within the 3D plane on the the 2 axis. I will probably have to put the input data into a matrice calculation, to determine the combined values against the set 3D area.
It will be wired at first to get it working, and will be made wireless afterwards, if it can be done.
Questions:
Anyone here ever deal with matrices? I have done it before on an old PC, it took some time to calculate, but it was a slow processor.
Would you know the speed required to do this? It is a pretty complex calculation.
Accererometer / gyro offset. The real world is not level, can an offset be programmed to the sensors to calibrate a zero plane?
More questions to follow.
kelvin
Guess i should of said 'matrix' , instead of matrice
Post Edited (kelvin james) : 3/7/2005 3:59:12 AM GMT