Over-ground walk/run velocity measurement methods

[Angela M Berg Robertson]

Good afternoon -
I need to measure the velocity of subjects walking and running along a walkway of 10+ meters. Our lab brainstorming has produced several ideas, and we would like to hear about your experiences with the methods below, and how well they meet the criteria below. Any other ideas are also welcome!
Thank you,
Dr. Angie Berg Robertson

Ideal criteria:
- simple
- wireless (i.e., subject not tethered)
- can cover 10 meters
- measurements can be recorded and saved/exported
- precision of 0.1 m/s or better at walk/run speeds
- can provide instantaneous velocity data (though averages would suffice)
- not affected by subject posture or sway
- can synchronize with other equipment

Ideas that meet some/most of the above criteria:
- accelerometer
- cameras
- ultrasonic positioning system (e.g., http://en.souvr.com/product/200712/274.html#facts)
- indoor GPS (e.g., http://www.polhemus.com/?page=Motion_Liberty_Latus)
- laser range finder
- tow-line
- speed gun
- light gates
- your suggestion!

[Jonathan Williams]

Re: Over-ground walk/run velocity measurement methods

The simplest and cheapest way would be to employ some walkway timing gates. We have some excellent light gates which do not require anything attached to the individual and can be made bespoke to cover any distance. Data is automatically transfered to a PC avoiding common light gate problems of having to read a stopwatch. Furthermore the system can be made to be synchronised to other systems, is fully portable and you can easily add additional gates for minimal extra cost. If you are interested in this low cost, simple solution contact Dr Robinson at hrobinson@h-scientific.co.uk Hope that helps

[Gaspar Morey Klapsing]

Re: Over-ground walk/run velocity measurement methods

Dear Angie,

of course the solution might depend on your setup. If you already use a motion capture system, you might prefer my first proposal. If not, solution 2 would be best.

I am assuming that you are interested in linear velocity, only in the direction of the walkway.

SOLUTION 1:
In my lab, as I do have a VICON-MX system and most of the things I measure involve VICON, I use one marker at the sacrum or the mean of two pelvis markers, to calculate the subjects velocity. Although I can get the instantaneous velocity right after and even during the trial, I'm interested in the mean velocity in between of two marks (1m prior and after my force plates". After every capture a pipeline runs a routine which gives me the mean velocity of that marker and tells me if it is within the range (usually target velocity +- 5% allowance), so I know a few seconds after every trial if it can be accepted or has to be rejected. We are doing this using PECS and MATLAB and are willing to share it (it might need some cosmetics, but for ourselves it is fine). Sometimes I use this even when I do not need VICON (e.g. only plantar pressure registration) ... then I "misuse" the VICON for "only" checking velocity.

SOLUTION 2:
The next suggestion would be using a doppler (either ultrasound or light). It is very easy to implement and you get a direct relation between registered voltage and velocity. For the ultrasound, all you need is a buzzer with a constant frequency (we used hardly audible 50 Hz in the former lab) a microphone and an A/D board to get the signal into the PC (you could also do the electronics outside the PC). I've just googled a bit and found an interesting abstract on the net.

http://www.merl.com/papers/docs/TR2007-107.pdf

Good luck

Dr. Gaspar Morey Klapsing

Good afternoon -
I need to measure the velocity of subjects walking and running along a walkway of 10+ meters. Our lab brainstorming has produced several ideas, and we would like to hear about your experiences with the methods below, and how well they meet the criteria below. Any other ideas are also welcome!
Thank you,
Dr. Angie Berg Robertson

Ideal criteria:
- simple
- wireless (i.e., subject not tethered)
- can cover 10 meters
- measurements can be recorded and saved/exported
- precision of 0.1 m/s or better at walk/run speeds
- can provide instantaneous velocity data (though averages would suffice)
- not affected by subject posture or sway
- can synchronize with other equipment

Ideas that meet some/most of the above criteria:
- accelerometer
- cameras
- ultrasonic positioning system (e.g., http://en.souvr.com/product/200712/274.html#facts)
- indoor GPS (e.g., http://www.polhemus.com/?page=Motion_Liberty_Latus)
- laser range finder
- tow-line
- speed gun
- light gates
- your suggestion!

[Lee Atkins]

Re: Over-ground walk/run velocity measurement methods

Hello,

I realize it has been quite a while, however I just came upon this thread. As such, would you be willing to share your Matlab code from Solution 1 for averaging velocity (assuming you still have it)?

Thank you