Detection of object movement & rotation in magnetic field

Question

I need to develop solution for monitoring movement & rotation of components in 3D space (integrated logistical and production workspace). Components have a size of circa 0.5 x 0.25 x 0.25 meters. Both movement as well as rotation of components need to be measured. Workspace where components are being moved & rotated is circa 10 x 10 x 4 meters. Total number of moves and rotations to each component can be high, counted in hundreds. Frequency is high as well, a few movements or rotations per second. Required precision: +/- 3 centimeters (movement), +/- 2 degrees (rotation). I am thinking about system consisting of two modules:

1. few (3?) dedicated “stations” emitting electromagnetic field with specific signature;
2. a detachable, small magnetic field sensors, which could be attached to each of the components in question.

Rough idea is following:

1. each of the “stations” is located in different corner of the workspace (at least one of them has to be on height different from the other two);
2. a to-be-monitored component is introduced to the workspace; a small magnetic field sensor is attached to the component and “calibrated” at place we can call “position 0”;
3. with each movement/ rotation of the component, the sensor detect change in magnetic field of each of the “stations”; information from the sensor is send via wi-fi to a dedicated computer;
4. based on information received, computer determines position/ rotation of the component in the production workspace.

A few questions to the above:

1. Does the general concept make any sense?
2. What type of magnetometers would be suggested for sensors (e.g. Hall? Reed? Any particular solutions/ models?)?
3. What components could be used for the “stations” emitting electromagnetic field?

Answer 1

This will not work based on a magnetic field: Everything will interact with everything. There can't be multiple stations (at most three orthogonal field vectors in 3D space means you couldn't even estimate vicinity to more than three stations if your object was floating in vacuum).

Also, positioning accuracy will be in the multiple meters in your description of the system, not centimeter, and also use kilowatts of power, because magnetic field strengths drops very sharply with distance from the magnet, and earth has its own magnetic field that you're competing with!

Hope you don't like tools made of steel, because they interact with magnetic fields, and thus can't be used if you're not willing to disturb your magnetic field sensors.

So, no, this isn't a viable approach, at all.

So, as Andy already said: your best guess is optic location. Cameras are cheap these day, even relatively high-resolving ones, and so is processing power. Especially if you can label your components, e.g. with colors or QR codes, this will get a lot easier; but even systems like specific coordinated times at which the components light up an infrared LED can work wonders.

With camera frame rates in the 10¹ to 10², your speed requirements are easily met, and optical resolution ability will definitely trump anything that you can build with a static magnetic field – which in the end can only be a rough proximity sensor, with the interaction problems mentioned above.