I wanted to know if it is possible to determine the relative position of a point relative to the arduino card. I'm told you need an accelerometer gyroscope. Ok. But I would like to xyz of a point outside the board. Possibly? I would simply like to communicate to a program the coordinates of a point.
There is a point exactly 1m away from my Arduino. What can you do with that point? There might be air or it might be under the table. Knowing the coordinates of the point tells me nothing about the point.
If you give us more information about what you are doing then we can help you more.
feste:
I wanted to know if it is possible to determine the relative position of a point relative to the arduino card. I'm told you need an accelerometer gyroscope. Ok. But I would like to xyz of a point outside the board. Possibly? I would simply like to communicate to a program the coordinates of a point.
Gyro+accelerometer can only give orientation long-term. Position cannot be figured out except in the very short term before integration drift completely destroys the meaning of the position data.
Outdoors you can use GPS to cancel the drift in position, though this isn't very precise. Barometer can do vertical
position but will also drift due to atmospheric changes (but much much less quickly than accelerometer drift).
Indoors is a hard problem. UWB ToF systems like Pozyx are expensive but are contenders for this.
I apologize for the delay, but I have not received reply notifications. I would like to determine the coordinates with respect to a base of a point in space, without connection to the base. If it were connected with wire would it be easier? I don't want GPS because it's inaccurate. I look for precision in enclosed spaces like a room.
How do you connect a wire to a point?
Another sensor that might be useful is Pixy.
How about a Laser measure? Take 4 readings, each 90 degrees apart, or 8 readings 45 degrees apart, and do some calculating from there to see how far you are from 2 walls.
I know the laser measurements, but I'm not trying to measure a room, but I would like to know the xyz position of a point with respect to a base.
Interesting pixy, but only detects what it sees.
Mine for now is just a theory. We have a gyroscope ... so a corner. With an accelerometer we should be able to determine the distance AB between sensor and "base" (?). When I have an angle and a distance - if I assign coordinates to my base X0 Y0 Z0 - I must always get relative positions of the point I want to detect. the base is X0 Y0 Z0 the point is XYZ variable. Everything here is what I want to find. How? Initially I like it even with a thread if there is no other way, but the point would be that it really was a point ... not another base, but very small.
With an accelerometer we should be able to determine the distance AB between sensor and "base" (?)
No. See this explanation for why not.
I'm told you need an accelerometer gyroscope.
Having read the above, you will be able to explain to whomever told you that, why they are wrong.
You keep saying "point". What is the point? Is there an object there you wish to measure?
Your last post makes it sound like you aren't measuring points but you wish to measure the distance travelled in XYZ space for your (handheld?) Arduino device. Start it at some "home position" and then move it along an unspecified path of unspecified length and the result should be XYZ relative to the start. Am I right?
As jremington's link shows, this is a spectacularly difficult task with only inertial navigation. It works in the physics textbooks but does not work in real life unless you have a million-dollar budget and access to military technology.
Tell us more about what you are really doing and maybe we can help simplify the problem to a solvable one.
Are y
feste:
I know the laser measurements, but I'm not trying to measure a room, but I would like to know the xyz position of a point with respect to a base.
Interesting pixy, but only detects what it sees.
Mine for now is just a theory. We have a gyroscope ... so a corner. With an accelerometer we should be able to determine the distance AB between sensor and "base" (?). When I have an angle and a distance - if I assign coordinates to my base X0 Y0 Z0 - I must always get relative positions of the point I want to detect. the base is X0 Y0 Z0 the point is XYZ variable. Everything here is what I want to find. How? Initially I like it even with a thread if there is no other way, but the point would be that it really was a point ... not another base, but very small.
It is difficult to determine what you are trying to achieve. However, my interpretation is that you are going to keep the IMU (accel/gyro) in the "base" and get the position of another "object". if that is correct, then you could do this with an accel/gyro and distance sensor. Add math (vectors and translations using the rotation of the base and distance to the object) and you have a way determine the offset location of the "object" relative to the "base".
From reading a tiny bit, the thought of iso-center comes to mind. You would want to find the iso-center of the room, that would be point 0,0,0. You can use strings from corner to corner to find the intersecting line that would become the iso-center. That would be the starting point. From there you'd do math to determine the x,y,z thingies.
Camen:
Are y
It is difficult to determine what you are trying to achieve. However, my interpretation is that you are going to keep the IMU (accel/gyro) in the "base" and get the position of another "object". if that is correct, then you could do this with an accel/gyro and distance sensor. Add math (vectors and translations using the rotation of the base and distance to the object) and you have a way determine the offset location of the "object" relative to the "base".
Yes. That's it. Can you also do with more points than the same base?
feste:
Yes. That's it. Can you also do with more points than the same base?
"than"???? Translation error?
Did you mean "from"? Yes, it should not be hard to measure more than one point from the same base.
Did you mean "instead"? Yes, you could have multiple bases, although I don't know how that might help. Traditional surveying uses multiple bases to measure the position of trees, buildings, mountains or whatever via triangulation.
I apologize for translation errors. As for my request, I was referring to triangulation, but more points than the single base. But the points ... how much are "points"? If the point I am looking for is a sensor X x Y cm (unknown measurements) ... where is the point?
to have a base and two points to determine which products should I look for concretely? Can you show me an example with a link?
I am also willing to consider more bases and then triangulation as you said ... even if I then have to relate the two bases in some way: they must have a geometric relationship between them.
In theory, if you know the origin location position (x,y,z), rotation (a,b,y) when pointing at new position (u) and the distance between the two locations (D) then you can find the new position (x2,y2,z2). Adding more origins (additional bases) only serves to validate and give more precision to the system.
