Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Designing a rf based tracking robot

  1. Dec 17, 2012 #1
    i am trying to design a robot which can track a transmitter which produces rf signals in a given range of frequencies(example:- 400-900 MHz). i want to know the following details.

    1. i have a design of wireless robot with A433MHz RF modules. i need to change the frequency of transmission without changing the main components like antenna(assume the tracking frequency is within the antenna' bandwidth).is it possible to design such transmitters and receivers in RF range. Then how to choose the antenna types for such designs if possible.

    2. i plan to train it initially to a particular frequency in the given range which is supposed to be my training phase. then the robot should track my transmitter's change in frequencies within lock in range and should be used for finding the transmitter's location.i plan to have a buzzer which sounds when the bot reaches say within 1 meter radius of the transmitter.

    3.please suggest me some key concepts and prerequisites for my extensive reading to implement this stuff.

    4. suggest me the feasibility of this design like how far this is possible for implementation. if not what are the modifications needed.

    thank you all.
  2. jcsd
  3. Dec 17, 2012 #2
    I was given this task as a college design project and it turned out to be some what complicated. I might be able to point you in the right direction but many of my project's complications were never resolved.

    In my design I attempted to use 2 directional antennas spaced a half meter apart to measure the phase difference of the received signal on each. I was using an Arduino to control the robot so I could not directly sample the RF signal. To resolve this I tryed using a phase detector chip but could never get it yo work correctly.

    If you care to look at this method you can read more about it here:

    Rather than this approach I recommend using 2 antennas and down mixing the received signals so that it can be directly sampled independently. Then you are free to use either phase of amplitude comparison monopulse to determine the direction of the beacon.

    If you want to build a variable frequency beacon I suggest using a tunable VCO and amplifier with an omni-directional dipole antenna.

    If your looking for RF parts Mini-Circuits is some of the best but allot of the components can be expensive.
  4. Dec 20, 2012 #3
    - Wideband antennas are difficult and the wrong option. Have one antenna per frequency.
    - Wideband HF electronics is difficult. Do you really need this? It does exist, this is the UHF TV band.
    - You plan to transmit within the TV band? Are you serious about that? Unless you live in Antarctica or within a Faraday shield, your attempt will be angrily stopped very soon.
    - HF needs many years to learn. What's your current knowledge, and when shall the project be finished? If the answers are "little" and "in 3 months" give up now.
  5. Dec 20, 2012 #4
    If you are really intending to hijack the UHF TV band illegally then why bother "designing" anything. Just pump 100mW onto an coat hanger, who cares if the VSWR is 25. If it doesn't have enough range to suit you, jack it up to 1W, 10W ...

    As the previous poster indicated, this would be a bad idea, you will be caught quickly. The point is that all of the stuff that we learn about RF design; picking the right antenna, tuning it, managing spurious response, etc. are necessary to design a radio that is compliant with regulatory requirements and has minimum power consumption, neither of which seem to be your goal.
    Last edited: Dec 20, 2012
  6. Dec 21, 2012 #5


    User Avatar
    Science Advisor
    Homework Helper

    Be careful in use of terms. HF is a specific range of radio spectrum (≈ 3 to 30MHz). The behavior of RF as well as the rules and regulations governing use of frequencies in this portion of the spectrum are quite different than the spectrum Issac plans to use; i.e. frequencies in the UHF band (≈ 300MHz to 3GHz).
    There is certainly bandwidth in the UHF spectrum that may be used for Issac's purpose. As long as the design (chosen frequency bandwidth, ERP, etc), meets regional spectrum management guidelines, no one will be shutting anybody down. The proposed 433 MHz module, is a good choice of spectrum, (see LPD433).
    I would not discourage anyone from learning RF design, no matter what their given knowledge or time frame. At completion the person will have increased their level of understanding.
    Last edited: Dec 21, 2012
  7. Dec 22, 2012 #6
    This project is definitely doable so don't get discouraged. For my junior design we were given the exact same task of finding a 433MHz beacon transmitting at 0dBm at 100m. None of our projects met with much success but we had nearly no RF experience and only a semester to achieve it, but got close near the end.
  8. Dec 23, 2012 #7
    hey i am planning to make a robot that can find the transmitter in a room, when it reaches about some radius near transmitter it produces a beep. Method am planning to make use of is that the antenna in the bot which is directional should be rotated.The antenna is connected to analog to digital converter through a low pass filter. The micro-controller i am using is MSP430 which has inbuilt ADC. The values are compared to find the direction of the transmitter(direction with maximum value) and then the antenna is fixed in that direction.

    This process continues periodically to converge in to a radius where transmitter is available(A predefined ADC value is used to find this condition). Then a buzzer sounds. As of now the assumption is that the robot has no obstacles in its path.

    Further developments like GPS could be used to find the robot position and image sensor interface can be used to guide the bot in proper direction.
  9. Dec 23, 2012 #8
    1. Is this a school project?

    2. Are you planning to track the frequency from 400 MHz to 900 MHz or just over narrow ranges in the 400 MHz band or 900 MHz band? I really doubt you will be able to track the signal from 400 MHz to 900 MHz. How would you prevent your receiver from locking onto a high power TV station?

    3. You didn't mention any amplifier or receiver between the antenna and the A/D. Were you planning to use one?

    4. How will you determine when the robot is within 1 meter of the transmitter?
  10. Dec 23, 2012 #9
    Keeping your radio in the 433MHz license free band and not modifying the radios is the way to go.

    For direction finding you really don't need to know much about RF design, assume your TX and RX modules are handling all of the dirty details.

    Here is what you need to know:

    There are two ways to do DF:

    1 - Rotate receiving antenna(s) and look at RSSI. You can look for peak or null. This approach can suffer severly from multipath. If you are in a multipath rich environment you will be getting constructive and destructive interference which can make the received pattern appear to not make any sense. For example at a particular distance you can get a reflection from the floor that makes your RSSI go to 0 when aimed right at your transmitter. At that same distance you can be getting a reflection from a metal file cabinet that reinforces the signal at your antenna. You can find yourself on a wild goose chase.

    2 - Doppler method. This involves rotating the antenna rapidly a circular orbit (while pointed in a given direction) and looking for the doppler shift in the received frequency. This is a much more reliable method, but more difficult to build.

    Assuming that you are using RSSI, I would suggest that you do a "site survey" of you area first just to see what you are up against. Do you have access to a spectrum analyzer?
    Last edited: Dec 23, 2012
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook