Shop OBEX P1 Docs P2 Docs Learn Events
IR vs. Sonar — Parallax Forums

IR vs. Sonar

ArchiverArchiver Posts: 46,084
edited 2003-04-16 01:37 in General Discussion
I am looking to add "Distance Detection" to my next Stamp II powered
robot. Which is better at determining the distance of an object...
Sharp IR Detectors (like the kind that Acroname sells) or Sonar
systems like the Devantech SRF04?

Which one is more accurate?
Which one is less susceptible false readings?
Which one works faster?

There are lots of articles about each, but I have not found much that
compares and contrast the advantages/disadvantages of each type.

Thanks,

Steve

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-04-15 22:55
    I wrote an article for N&V april 2002 regarding the Devantech srf04:
    discussing the pros and cons of IR sensors.
    IR will tend to give false readings depending on wall colors, fabric, light
    sources etc. The sharp series is also not good at detecting small objects
    such as chair legs.
    Speed of detection is faster with the devantech sonar as well.
    I highly recommend the devantech srf04 sonar. I use it for my main obstacle
    detection on my robot. It will pick up all small objects such as chair legs,
    or small trees outdoors. The price is not that much more then a sharp
    either. For a real world speed of detection. I used a yardstick, and passed
    it across the Devantech at a moderate speed, at a range of 1 foot. It would
    easily pick up the moving yardstick everytime. Approximatly the speed a
    flagman waves a flag at you on the road. The sharp series would be hard
    pressed to detect one movement let alone every one. One negative point of
    the devantech is that it has a wide field of view. 40 degrees facing forward
    in a cone shape. You may get a detection at a long range, and not know
    exactly where the object was detected. For example was it on the left
    forward, or right forward of the robot. I had a track robot setup to swing
    the robot left 45 degrees, back to center, then right 45 degrees on an
    obkect detection. This way I knew exactly where the object was. I later
    placed a sharp IR sensor on a servo head. I detect an object with the
    devantech. Then I rotate the servo/sensor head to find out exactly where the
    object is. The narrow field of view of the sharp sensor allows me to do this.
    Good luck, and have fun.
    -Kerry



    At 09:22 PM 4/15/03 -0000, you wrote:
    >I am looking to add "Distance Detection" to my next Stamp II powered
    >robot. Which is better at determining the distance of an object...
    >Sharp IR Detectors (like the kind that Acroname sells) or Sonar
    >systems like the Devantech SRF04?
    >
    >Which one is more accurate?
    >Which one is less susceptible false readings?
    >Which one works faster?
    >
    >There are lots of articles about each, but I have not found much that
    >compares and contrast the advantages/disadvantages of each type.
    >
    >Thanks,
    >
    >Steve
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >
    >
    Very respectfully, your obedient servant.
    Kerry
    Admin@M...
    WWW server hosting [url=Http://mntnweb.com]Http://mntnweb.com[/url]
    Kerry Barlow
    p.o. box 21
    kirkwood ny
    13795
    607-775-1575
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-15 23:22
    Steve,
    I am currently using both products, a sharp GPD05 and the Devantech
    system for a science project, to find out exactly what your asking. So
    far the Devantech system is the better of the two because it gives a
    more accurate reading (in my findings). Even though the IR is 13$
    cheaper, the Devantech system is the way to go. You can see my bot on
    the yahoogroups website, under the folder Kirby which employs both, but
    is pictured with the Devantech finder.

    -Mike

    Original Message
    From: Kerry Barlow [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=MT8vbRIK_YgT-uKlcHlYaVQt6hVJOkTXR5MV0_QxN45kbvOEmxh3InaMf9D8qZdM07GrUafjSQ8]admin@m...[/url
    Sent: Tuesday, April 15, 2003 5:56 PM
    To: basicstamps@yahoogroups.com
    Subject: Re: [noparse][[/noparse]basicstamps] IR vs. Sonar

    I wrote an article for N&V april 2002 regarding the Devantech srf04:
    discussing the pros and cons of IR sensors.
    IR will tend to give false readings depending on wall colors, fabric,
    light
    sources etc. The sharp series is also not good at detecting small
    objects
    such as chair legs.
    Speed of detection is faster with the devantech sonar as well.
    I highly recommend the devantech srf04 sonar. I use it for my main
    obstacle
    detection on my robot. It will pick up all small objects such as chair
    legs,
    or small trees outdoors. The price is not that much more then a sharp
    either. For a real world speed of detection. I used a yardstick, and
    passed
    it across the Devantech at a moderate speed, at a range of 1 foot. It
    would
    easily pick up the moving yardstick everytime. Approximatly the speed a
    flagman waves a flag at you on the road. The sharp series would be hard
    pressed to detect one movement let alone every one. One negative point
    of
    the devantech is that it has a wide field of view. 40 degrees facing
    forward
    in a cone shape. You may get a detection at a long range, and not know
    exactly where the object was detected. For example was it on the left
    forward, or right forward of the robot. I had a track robot setup to
    swing
    the robot left 45 degrees, back to center, then right 45 degrees on an
    obkect detection. This way I knew exactly where the object was. I later
    placed a sharp IR sensor on a servo head. I detect an object with the
    devantech. Then I rotate the servo/sensor head to find out exactly where
    the
    object is. The narrow field of view of the sharp sensor allows me to do
    this.
    Good luck, and have fun.
    -Kerry



    At 09:22 PM 4/15/03 -0000, you wrote:
    >I am looking to add "Distance Detection" to my next Stamp II powered
    >robot. Which is better at determining the distance of an object...
    >Sharp IR Detectors (like the kind that Acroname sells) or Sonar
    >systems like the Devantech SRF04?
    >
    >Which one is more accurate?
    >Which one is less susceptible false readings?
    >Which one works faster?
    >
    >There are lots of articles about each, but I have not found much that
    >compares and contrast the advantages/disadvantages of each type.
    >
    >Thanks,
    >
    >Steve
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject
    and
    Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
    >
    >
    >
    Very respectfully, your obedient servant.
    Kerry
    Admin@M...
    WWW server hosting [url=Http://mntnweb.com]Http://mntnweb.com[/url]
    Kerry Barlow
    p.o. box 21
    kirkwood ny
    13795
    607-775-1575




    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-15 23:49
    Well,

    The SRF04 has more range and returns a value that can be directly converted
    to distance. It has a fairly wide pattern and would be good for object
    avoidance. With some work it is also good for object detection.

    The Sharp sensors have a much shorter range and return a non-linear voltage
    or value that would have to be converted to get the distance. They have a
    much narrower pattern, which is better for object detection. You would need
    to scan them to use them for object avoidance.

    I have two GP2D02's and an SRF04 on my big robot, all mounted on a disk that
    will turn 90 degrees. Everything is run off a BS2SX, and It works for wall
    following, object detection, etc....

    Original Message

    > I am looking to add "Distance Detection" to my next Stamp II powered
    > robot. Which is better at determining the distance of an object...
    > Sharp IR Detectors (like the kind that Acroname sells) or Sonar
    > systems like the Devantech SRF04?
    >
    > Which one is more accurate?
    > Which one is less susceptible false readings?
    > Which one works faster?
    >
    > There are lots of articles about each, but I have not found much that
    > compares and contrast the advantages/disadvantages of each type.
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-16 01:37
    I've used both. I find them to be pretty similar in characteristics and
    ease of use. Both can be fooled by certain materials and shapes. They are
    both pretty accurate within their ranges. The SRF04 and the Gp2D02 take
    about the same time to make a reading.

    The GP2d02 is less expensive and uses less power. The annoying thing about
    it is that it has a very narrow field of view. I use two of them facing
    slightly towards each other or I use a servo to swivel one back and forth.
    There is also a version of these that outputs different voltage to show
    distance. The Stamp doesn't do ADC, but a you could use a ADC chip to
    convert the voltage. These analog versions are also power hogs and use
    about the same current as the SRF04.

    The SRF04 is nice, but a bit pricey I think. It's also uses a lot of power.
    The SRF08 uses less juice, but is twice the cost.

    If you are adventurous, you might look into Circuit Cellar's PSoC Sonar
    ranger project. It uses a Cypress PSoC chip and two sonar transducers to
    make a sonar ranger. The initial outlay of the price of the programmer is
    $25 [noparse][[/noparse]Digikey], but the chips themselves are pretty inexpensive and the
    software is free. Once you get the hang of the PSoC you could build a Sonar
    ranger for $14 ~ $18 ($7 for the PSoC [noparse][[/noparse]Digikey], $7 for the transducers +
    plus misc components).

    check out:
    http://stage.cypressmicro.com/

    chris in napa
    Original Message
    From: "sjoblin001" <steven_joblin@h...>
    To: <basicstamps@yahoogroups.com>
    Sent: Tuesday, April 15, 2003 2:22 PM
    Subject: [noparse][[/noparse]basicstamps] IR vs. Sonar


    > I am looking to add "Distance Detection" to my next Stamp II powered
    > robot. Which is better at determining the distance of an object...
    > Sharp IR Detectors (like the kind that Acroname sells) or Sonar
    > systems like the Devantech SRF04?
    >
    > Which one is more accurate?
    > Which one is less susceptible false readings?
    > Which one works faster?
    >
    > There are lots of articles about each, but I have not found much that
    > compares and contrast the advantages/disadvantages of each type.
    >
    > Thanks,
    >
    > Steve
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >
Sign In or Register to comment.