I have a question regarding a problem that seems easy, but after already doing some internet searches, seems not to be that easy to solve. I have a non-moving test setup in a large room where I have to measure the 3D coordinates of some sensors, and I need a fast and easy way of measuring their 3D position coordinates - without having to use a ruler. The sensors are quite small and mostly installed inside a machine. Because of the installation inside a machine, typically visual methods are not possible. The location has to be determined only once, and the reason to automate this is because there are quite a lot of sensors on locations that are not always easy to measure.

The required precision would be in the order of 1 cm. The solution should be fast, but does not have to be wireless. The size of the test setup itself could be several meters.

Does someone have an idea what type of sensors would be suitable for this kind of application? Is a dead-reckoning solution with accelerometers that have suitable low-frequent sensitity a good and cost-effective solution, regarding the given precision? And if so, should I use some noise filters?

  • \$\begingroup\$ Is this an optical table or something? If they're installed inside a machine, what access do you have? How do you propose to use dead-reckoning - track the movement of the sensor at the time of its installation? \$\endgroup\$ – pjc50 Mar 6 '15 at 13:32
  • \$\begingroup\$ This is a mechanical design problem, not an electronic one. You'll probably get a better response on Engineering.SE. \$\endgroup\$ – Dave Tweed Mar 6 '15 at 14:09
  • \$\begingroup\$ Do you know the location of the sensor relative to the outside of the machine so that only the machine has to be located? If not, it will be very difficult to locate a non-accessible sensor that doesn't emit any energy. \$\endgroup\$ – Barry Mar 6 '15 at 15:42
  • \$\begingroup\$ @DaveTweed no, this is a robotics/localisation issue. It would be best on Robotics SE \$\endgroup\$ – KyranF Mar 6 '15 at 16:37
  • \$\begingroup\$ Hardev, dead reckoning doesn't work unless you are moving, positioning at rest and initial conditions for dead reckoning are always starting at 0, unless you align to a known map. These concepts are for mobile robotics, not a static test emplacement. Without visual or sonic access, it will be hard to measure your position in 3D space to 1cm accuracy. \$\endgroup\$ – KyranF Mar 6 '15 at 16:41

Your Answer

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

Browse other questions tagged or ask your own question.