Measuring speed, angle and height

It all sounds like a bit of a stretch for an arduino in real time. You could use an array of ultrasound receivers and measure time delays. The ball will be long gone by the time you've done the calculations. You stand a pretty good chance of missing the  ball altogether.

Or perhaps an array of light sensors detecting a laser scan rather than a full image scheme. It would have to be in 2 orthogonal planes.

You might be able to use capacitive sensing to detect your object, but you probably need upgrade to a Cypress Semiconductor device to get the sensitivity and speed you need.

DAB

Thank you for your input DAB, much appreciated! Salvador

Could be tricky - the tennis ball is an insulator and at least (according to initial spec) 5cm away from the sensor plane. Now if the tennis ball could be charged up to a few kV it might be possible .......

MK

For an extreme case, if the measuring device is external to the tennis ball, then you could use the low cost camera on a Raspberry Pi as an image sensor, to measure distance traveled over time between frames.  Need to add in features for angle to camera and distance from camera to the equation. Difficult but could be done.

Can you interface a camera capable of 2000 frames per second to an RPI and could it be cheap ?

MK

Thanks all for you input and feedback. I chatted with the person doing this project and passed on all your concerns and recommendations and as a result he will using the RPi and a camera.  Again, thanks for all your input. Salvador

How about using a webcam and some tracking software?

Put it against a gridded background.

QED.

Are you repeating Galileo's gravity experiments by any chance?

Here is a technique that might work - the accuracy depends on how well it is implemented.

• start with a narrow beam LED (preferably collimated) or a laser
• shine it through a section of a cylindrical lens to create a light beam that looks like a vertical line or "plane" of light
• oscillate the light at some frequency above 100 KHz - the higher the frequency the more accurate the speed measurement can be
• use a light sensor to detect reflections of the light off the ball as it goes through the beam
• a high-pass filter or bandpass filter on the sensor signal will eliminate ambient light
• count the number of reflected pulses to know how long the ball was in the beam - this is a measure of ball speed
• use the same type of system except make the light plane horizontal instead of vertical, and again count pulses as the ball passes through the second beam
• knowing the speed from the first beam, you can calculate the angle of trajectory by how long it takes to pass through the second beam

Of course it only works if the ball passes all the way through both beams. If the second beam is not horizontal, say it is at 45 degrees, it is still possible to calculate trajectory angle, but it is harder to get an accurate measurement.

With a PSoC4 you could make a rudimentary system for under $10 in parts, ($15 if you include a display)

An arduino should also be fast enough to count these pulses.

You could do even this without even needing a microcontroller - just some counters and logic.

Doug

Thanks COMPACT. While perhaps not a Webcam, i does seem as tough he liked the idea of a camera.

We did not go into too much detail of the project, other than what I posted, but from what he mentioned, it did not give me the idea that he was going to reproduce Galileo's gravity experiment. Regards,Salvador