Relative Location Guidance

Hello everyone, I would like some suggestions on how best to approach this. Now this is the situation.

There is a robot companion that moves and utilizes GPS to get its current location to +/-1.5m at the moment, which isn't the best, but it is also within our budget. Regardless, we also have a wearable that someone is obviously wearing. This wearable does not have GPS built on to it. The question is, is it possible to calculate the relative locations of the robot companion and the wearable using something like an IMU? We don't want the wearable and the companion to get too far from one another.

I appreciate all suggestions!

BilalTroll.

Hello
Take a search engine of your choice and ask the WWW for " Arduino Dead reckoning".

Have a nice day and enjoy programming in C++ and learning.
MIND THE GAP

No. An IMU tells you nothing about your position in the world.

One possible cheap solution is ultrasonic ranging. You can modify a pair of $2 ultrasonic rangefinders to make one a sender and the other a receiver. You need a way to trigger them 'simultaneously' (radio or infrared?). With two receivers you can calculate relative angle as well as distance.

or GPS, since it's already on one device... oh, wait, no GPS on the wearable...

that's a pretty good GPS if you stay always within that range

Actually, I wonder if the 1PPS signal is synchronized across multiple GPS receivers within a few microseconds. Does the 1PPS signal continue when the GPS can't see the satellites?

IIRC the PPS spec on my receivers is +/-10us. Actually you may have to check the serial for a lock condition, because AFAIK it does continue to run whether there is a lock or not.

Take that as 99% sure, because I haven't played with them in a while. I remember that when I was using PPS, it took longer to achieve PPS lock than to just acquire a fix - up to several minutes. My observation, the on board LED was a good indication of a PPS lock.

I was actually planning to pull them out and try again, last week. Never got around to it. Useless information here, I guess, without testing. The scenario of locking, and then losing lock, I really never tested at all.

Seems to me, the PPS lock to 10us would be easy to build in, since you really need that accuracy just to obtain a fix.

I wonder if the GPS on the robot is a red herring? What you are saying is that the wearable can't access GPS, because it's too small or not enough battery or whatever, or is inside. The problem really is how the robot can locate the wearable so it can follow it. You suggested an IMU, but at the current state of the art low cost IMUs and inertial navigation can't really give very good precision. That may well change but not for a year or two.

Where are the person and the robot located? Will they be indoors, outdoors, or a combination?

Precise relative location indoors is the kind of application where ultra-wideband radio (UWB) can be used. These people:

seem to use it for a similar application. A company specialising in UWB is

I'd love to see the smart suitcase go through the Xray machine security. :slight_smile: If nothing else, I believe you have to take batteries with you in the cabin, they are not allowed in the cargo bay. Oh, they said removable - but that would be a hassle. Still you would probably often be taken aside for questioning...

No, but the distance between the companion and the wearable can be measured using UWB modules, like these: ESP32 UWB(Ultra Wideband) | Makerfabs

With three or more of those modules, the relative position can be calculated as well. GitHub - jremington/UWB-Indoor-Localization_Arduino: Open source Indoor localization using Arduino and ESP32_UWB tags + anchors

no the PPS is from the satellites that is why the PPS from multiple GPS in geographical proximity to one another are so well correlated.

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.