Put the bluetooth module in a tin can with the opening forward. The can will reject most signals from back and lateral of its opening, so strongest signal will be when the beacon is somewhere ahead.
See cantenna cantenna - Google Search
It doesn't need to be precisely the tuned size/dimensions, any tin container will give more or less forward directivity.
azahari:
The year 2020 is next year, why does the average home doesn't have a robot that can move heavy items from one place to another with a push of a button?
I have to question your premise. How often do you need to move heavy items around the home? How heavy? A ton? 90lbs? Is it the size of a couch or the size of a Sony Wega TV? What will this robot do that a $20 furniture dolly or a $50 hand truck can't?
I mean, putting technology aside for a moment (something the users of this site hate to do), what are you really trying to accomplish here? Might want to address that before getting bogged down in Bluetooth modules.
I'm not trying to be negative: I spend a lot of time on a weekly/daily basis moving heavy things and I've given the matter a lot of thought. But I live on a farm and that's where the heavy things I need to move are, not in the house. There are lots of existing solutions, but many of them are lacking in one way or another depending on your individual situation.
Well, to answer your original question, I can think of a few.
RSSI is not reliable for distance, even less for direction. You can never be sure that the strongest signal you find looking around is the direction the beacon is in, it can very well be a reflection.
other objects that are in the way and the robot has to navigate around, keeping the dimensions of this "heavy object" in mind.
the need of placing this "heavy object" on the robot in a safe way, so it won't fall off or hit other objects en route.
the great variety of sizes and shapes "heavy object"s tend to come in.
obstacles like stairs (even an issue for highly advanced Daleks!), thresholds/doorsteps, etc.
I'm sure there are more reasons.
Why do you think "follow me" luggage is also still not there? That's a simpler problem to solve, as the moving object itself is so well defined.
wvmarle:
Well, to answer your original question, I can think of a few.
RSSI is not reliable for distance, even less for direction. You can never be sure that the strongest signal you find looking around is the direction the beacon is in, it can very well be a reflection.
other objects that are in the way and the robot has to navigate around, keeping the dimensions of this "heavy object" in mind.
the need of placing this "heavy object" on the robot in a safe way, so it won't fall off or hit other objects en route.
the great variety of sizes and shapes "heavy object"s tend to come in.
obstacles like stairs (even an issue for highly advanced Daleks!), thresholds/doorsteps, etc.
I'm sure there are more reasons.
Why do you think "follow me" luggage is also still not there? That's a simpler problem to solve, as the moving object itself is so well defined.
For now, I'm focusing on coding this
Pseudocode:
otomoov switched on
rotate 360 degrees. measures and records BLE signal strength
360 degrees rotation complete
determine angle with highest signal strength(eg: 30 degrees)
turns 30 degrees.
end
To all coders who comprehend what I'm saying, feel free PM me your price.
as you can see, this is what I imagine how it would find its direction. There's a distance between the center of the robot and the receiver. When it is switched on, it rotates 360degress finding which angle has the highest signal strength. Then after it did a full rotation, it then turns to the angle which has the highest signal strength
You'll have to get that done, first. Then do experiments: rotate it by hand, record RSSI values in various environments. See whether it's even a reliable way of determining the direction (you WILL see peaks in odd directions due to reflections in a normal home type environment with furniture, walls, people, and other stuff present).
Sounds highly unreliable. Do experiment with it, get the data.
accelerometer module, do you have any recommendation?
Perfectly unreliable for measuring quantity of rotation. Use a magnetic compass. Dead reckoning using encoders on the wheels also works quite well for a two-wheeled robot. You can use the same encoders to keep the thing going more or less in a straight line.
Max 5 meters, 360 data points, about ~10cm error.
Then your direction has to be accurate to 1°. I'd be seriously surprised if you can reliably find the direction within 10°. You'll have to stop and search a few times as the signal drops because your angle is getting too large as you move.
robot kit + Rigado BMD-360-EVAL Evaluation Kit + iTag Bluetooth tracker
This is the setup I've decided to drive this project forward.
About Rigado BMD-360-EVAL Evaluation Kit
Rigado BMD-360-EVAL Evaluation Kit is a great starting point for Bluetooth 5.1 Low Energy projects. The kit allows easy access to all features of the BMD-360. A simple USB connection provides power, programming, and a virtual COM port.
Four user buttons, four LEDs and a reset button are available. All GPIO are available on headers and are compatible with the Arduino form factor. This allows easy use of any existing Arduino shields. Current sense resistors allow for measuring current into the BMD-3xx module and into the shield. https://my.mouser.com/new/rigado/rigado-bmd-360-eval-kit/
Now to the topic of $$$
How much? PM me your price. Video evidence will be prioritized.
