DWM1001-Dev: tight proxmity

Wireless Connectivity Ultra-Wideband
,
1

Hi,

My robotics project is on quadcopters (drones) and I am exploring the use of DWM1001-Dev for indoor localization. Specifically, I am trying to bring the drone to a 1 m^3 cube zone within a 3 m radius, where the cube zone is the only placement location for anchors / tags. My questions regarding the DWM1001-Dev:

  1. What is the minimum distance between anchors? In other words, if anchors are spaced 1 m apart from one another, is it possible to attain cm level position accuracy in a 2-tags, 4-anchors system (assuming constant LOS)?

  2. Can relative distance be measured in a 1-anchor (moving) + multi-tags (stationary) system? I understand 1A + 1T system provides only relative distance measurement between the two nodes and usually the anchor is fixed in place. I was wondering if I can get multiple distance measurements between a moving anchor and multiple stationary tags?

  3. Can distance be measured in a 1-anchor (stationary) + multi-tags (moving) system? What if the system is similar to 1A + 1T, where the anchor is fixed in place, can relative distance be measured between an anchor and multiple moving tags?

  4. What is the minimum distance between an anchor and a tag?

Thank you in advance for your help and comments. You are greatly appreciated.

2

Hi Anthony,

Please have a look at the DWM1001 (PANS) system overview. It should answer most of your questions.

  1. What is the minimum distance between anchors?

It is recommended to space anchors 20 to 25 meters apart when placed in a square grid. I don’t think there is a minimal distance. In theory, a smaller distance will increase received power, which is good for accuracy, but it might also make trilateration less precise. Anyway, for 1m accuracy this shouldn’t matter. Be sure to use a box (or other 3D) arrangement in order to get correct 3D positioning.

is it possible to attain cm level position accuracy in a 2-tags, 4-anchors system

It is possible to get 10 cm level accuracy with PANS. More anchors will increase accuracy. Theoretically better performance can be achieved with the DWM1001 system using more advanced ranging techniques and better calibration, but this is beyond the scope of PANS.

  1. Can relative distance be measured in a 1-anchor (moving) + multi-tags (stationary) system?

In theory, yes, but the absolute positioning calculation will not make sense. Why not use multiple anchors? Tags and anchor are just names, an anchor is chosen as name for a stationary node, a tag for a node that moves. I’m not sure what you are trying to achieve by reversing this. You could have a system with many tags and one anchor, or a system with many anchors and one tag.

  1. Can distance be measured in a 1-anchor (stationary) + multi-tags (moving) system?

Yes. The PANS RTLS system uses TWR between tags and anchors. Calculating the absolute position on the tags won’t work off course.

  1. What is the minimum distance between an anchor and a tag?

In theory, 0mm. Depends how well the antenna delay is calibrated.

I suggest placing at least 4 anchors on the outside of your 3m radius area (sphere?) in an equal sided tetrahedron (or pyramid, cube, …) formation and have your drone(s) act as tags.

Your biggest obstacle will probably not be accuracy, but rather latency. the max location rate is of PANS is 10 HZ. A drone can move a lot off distance in 1/10 of a second. I suggest creating your own firmware to achieve better results or to augment the position data with e.g an accelerometer.

3

Hi Seppe,

Thank you for getting back to me with such detailed response and recommendation. I apologize if my questions were ambiguous, allow me to rephrase since some of my questions were not yet fully addressed:

  • The reason I asked if there is a minimum distance between anchor nodes is because I only have a small confined space (1 m^3) to place the anchors. On top of the PANS system documentation, I also reviewed Source of error in TWR and it appears that the Range Bias Effect might affect the location measurements between anchors and as the tag come close to the anchor. That’s why I was wondering if cm level location accuracy (precise trilateration) can be achieved in a 2-T, 4-A system where anchors are placed 1 m apart.

Yes, my drone has other components for position data, assuming 10 Hz is sufficient, what will be the position accuracy (precise trilateraion) or distance measurement

Thank you again for your input and I look forward to your response.