0
\$\begingroup\$

Is that possible? I'd like to track the amplitude of a horse leg/foot height for a show.

The many many difficulties:

  • constantly moving! So ultrasound way too slow
  • the floor is irregular, made of shavings
  • seems that an accelerometer isn't reliable for measuring distance, as I read here.

The best would be a sensor I can put on the horse's leg/foot rather than video tracking (i.e. openCV) because the stage is round, and I'd love to keep the process light (not putting cameras all around etc).

\$\endgroup\$
  • 1
    \$\begingroup\$ Even though you prefer not to use image processing, that would be the most reliable solution (with existing tools to process your data). \$\endgroup\$ – Eugene Sh. Feb 12 at 19:22
  • 1
    \$\begingroup\$ If you start with the assumption of a mostly flat floor, you may do pretty well with an IMU (with emphasis on the "may" -- this will at least take some playing around with the math to see if it's feasible). Because the hoof rotates you'll need a full six-channel IMU -- an accelerometer won't cut it. The horse's lateral motion will also make the signal processing more complicated. \$\endgroup\$ – TimWescott Feb 12 at 19:26
  • \$\begingroup\$ Look at this:azosensors.com/article.aspx?ArticleID=43 \$\endgroup\$ – Fredled Feb 12 at 22:36
2
\$\begingroup\$

Actually, because of the cyclical motion, and the fact that each gait ends on the floor, an accelerometer might be ideal. More specifically, a 6 degree of freedom motion sensor unit.

The problem with using accelerometers to measure distance is accumulated errors. Since the gait always ends at the ground, you can remove errors at the end of each step.

\$\endgroup\$
  • 1
    \$\begingroup\$ This has already been demonstrated on humans. It's simple, cheap, lightweight and easily adapted to horses. A search for the phrase "zero velocity update" will find all kinds of papers about exactly this method. \$\endgroup\$ – Dave Tweed Feb 12 at 22:18
0
\$\begingroup\$

You could use a Lidar, which can very rapidly determine the distance from a given object. They are frequently used in robotics and autonomous vehicles to determine their distance from walls, floor, etc. Some of them are pretty small and could possibly fit on a horse's leg, and are relatively inexpensive.

\$\endgroup\$
  • \$\begingroup\$ I doubt there are lidars for ranges from zero up. \$\endgroup\$ – Eugene Sh. Feb 12 at 20:19
  • \$\begingroup\$ @EugeneSh. I'm not sure what you mean "from zero up" - could you clarify? \$\endgroup\$ – ConcernedHobbit Feb 12 at 20:20
  • \$\begingroup\$ Lidars are usually not working well for short ranges. Here we are talking about ranges between zero and up to 1-2 meters. \$\endgroup\$ – Eugene Sh. Feb 12 at 20:22
  • \$\begingroup\$ @EugeneSh. Good point. If he is mounting the lidar to the horses leg (what would be the horse's ankle) it might work, since it would never be fully reaching the ground, but if he is mounting it directly to the horse's foot, that wouldn't work. \$\endgroup\$ – ConcernedHobbit Feb 12 at 20:24
  • \$\begingroup\$ It'd be virtually impossible to make sure that a 1-D LIDAR is measuring vertical distance and not some slant distance. On the other hand, a 3-D LIDAR mounted off to the side, that the horse moves past, could possibly do motion capture with enough precision to be useful. Not cheap, though. \$\endgroup\$ – Dave Tweed Feb 12 at 22:16
0
\$\begingroup\$

That is a complicated task, which may require a lot of post-processing. For the sensor part, I would give a shot to barometers. Despite the air movement, an array of barometers could be used for maximum SNR. As of 2019, it's typical to find high-precision barometers having 0.01 hPa resolution, such as LPS25HB. It's enough to track altitude delta of centimeters.

\$\endgroup\$
  • 1
    \$\begingroup\$ Wouldn't barometers be highly affected by the rapid movement? Bernoulli principle and such... \$\endgroup\$ – Eugene Sh. Feb 12 at 19:45
  • 1
    \$\begingroup\$ I think will depend on the electronics final packaging because of airflow. Also, placed in different positions, and supported by accelerometers it could be processed. \$\endgroup\$ – A. Ayres Feb 12 at 20:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.