Go Down

Topic: Follow me bot (ultraonic tracking system ) -help !!!!!!!!!!11 (Read 2453 times) previous topic - next topic

I am building a follow me robot.I have decided to use ultrasonic tracking system.The user will be carrying an ultrasonic transmitter which sends out bursts at definite intervals  and the bot will be having three receivers.Based on the reception time at three receivers the source can be located .I have some doubts in building the transmitter and receiver .As i need an omnidirectional tx and rx ,i have decided to use prowave-400fs080http://in.element14.com/prowave/400fs080/ultrasonic-txr-40khz-11mm/dp/1007365?Ntt=400FS080 .I need help in constructing the transmitter which gives out bursts at definite intervals and the receiver which detects the signal and sends the signal to the controller(arduino) .I am not able to determine whether prowave-400fs080 is just a transducer or a transmitter by itself i.e . there is an inbuilt circuitry which generates a 40 khz signal .I don't know how to verify this .could anybody help me please

PaulS

Quote
Based on the reception time at three receivers the source can be located

Really? How, pray tell? Just how far apart are these sensors going to be?

Think about the distance from the center receiver to the transmitter. Then, think about the distance from the left receiver to the transmitter. The distance from the left and right receivers, when they form right angles, is a function of the distance from the center receiver to the transmitter and the distance between the left or right receiver and the center receiver. What, exactly, is the increase in distance between the left or right receiver over the distance to the center receiver. Given the speed of sound, what does that increased distance translate to in increase in time? Do you think that you can measure that delta time?

zoomkat

The below might be of interest.

http://revision3.com/tbhs/robotluggage
Google forum search: Use Google Advanced Search and use Http://forum.arduino.cc/index in the "site or domain:" box.


HI PaulS ,
I thought of tracking the source as depicted in the picture .Here e1 is the time interval between reception at rx1 and rx2,e2 is the time interval between reception at rx2 and rx3.
Do you think this is possible ?

PaulS

Quote
Here e1 is the time interval between reception at rx1 and rx2,e2 is the time interval between reception at rx2 and rx3.

What is that time interval?

As the matter in consideration is sound ,the time interval between the reception must be in micro seconds.

PaulS

Quote
As the matter in consideration is sound ,the time interval between the reception must be in micro seconds.

Provide some sample calculations. Suppose that the tracker is 10 feet (or 3 meters) from the subject being tracked (D1 = 10 feet/3 meters). How long does it take the signal to reach the center receiver (Rx1). If the tracker is 45 degrees off, so D2 is minimal, how long does it take the signal to reach Rx2? How long does it take the signal to reach Rx3?

Keep in mind that time is measured in whole microseconds, so be sure to express your answers as whole numbers.

I have given the calculations in the diagram for the values you mentioned.I was wrong abt the time interval.Its only in milliseconds not microseconds. .
If I could determine the time interval as t1~t2 =0.7 ms ,t1~t3 =0.7 ms  ,I could bring out the distance and angle as follows .
velocity = distance/time.here velocty of sound is 340m/s .
d1 = v .t1=340 .t1
d2=v.(t1+(t1~t2))=340.(t1+0,7ms)
d3=v.(t1+(t1~t3)) =340.(t1+0.7 ms)
consider the rx1 to be the origin in a 2dimmensional coordinate system .this leaves us with (x1,y1) =(0 ,0)  ,(x2,y2)=(0.33,0) ,(x3,y3) =(0,0.33) .(x4,y4 ) the coordinates of tx which is unknown .so by using the distance formula for d1,d2 and d3 in the above equations . x4 and y4 can be determined .from which the angle can be calculated .
Is there any error in this method ?

Oops!! :smiley-sweat: I can find the mistake now.when the values are substituted ,I am left with two equations and three unknowns(x,y,t) .Is there any other way tracking ?

PaulS

If you can determine the time that a pulse arrives at each of the three sensors, you can tell if you are facing approximately the correct direction, because the time of flight to the center sensor will be the smallest of the three. Once you know that you are facing in approximately the right direction, the time of flight to the other two sensors will be equal if you are facing in exactly the right direction. The difference between the time of flight to the outer sensors will give you both the direction to turn (because the difference is either positive or negative) and the relative amount (large value == way off; small value == almost right). The exact angle doesn't really matter, as a few milliseconds later, you'll check and adjust again.

This all assumes that the receivers know WHEN the sender sent a pulse, which is where I think you'll find that are missing something.

So to synchronize the source and receiver,I will use an rf tx and rx .The time interval can be found by starting a timer when rf-tx initiates the ultrasonic transmission  and stopping the timer when the Us-rx detects the signal. this cycle can be repeated at particular intervals.will I be able to use the Timers available in the arduino board for this purpose?How do I do it?

Go Up