Ytai: Why can’t I use:
strobe.write(true); strobe.write(false); input.getDuration(); Vic > On Apr 12, 2016, at 2:57 PM, Ytai Ben-Tsvi <[email protected]> wrote: > > Good question. Currently the only way to ensure this level of sync is the > motion control API. Alternatively, if your time constants are coarse enough, > you can open multiple PWM channels running at slow rates and insert delays > between the open command (order of 10's of ms as you have seen with your > experiments). The ideal solution would have been a one-shot function, which > is easy to implement using an output compare module, but this doesn't > currently exist. > > On Apr 12, 2016 11:07, "Vic Wintriss" <[email protected] > <mailto:[email protected]>> wrote: > How would you work the PWM with multiple sensors? > > On Tuesday, April 12, 2016 at 10:54:34 AM UTC-7, Ytai wrote: > No threads involved and nothing asynchronous. You set the PWM once during > setup(), then you getDuration() whenever you please. > > On Apr 12, 2016 10:52 AM, "Vic Wintriss" <[email protected] <>> wrote: > I would rather run the ultrasonics asynchronously. Also since the kids may > not be be that advanced in our curriculum I prefer to stay away from threads. > The last code that I posted works fine. > > On Tuesday, April 12, 2016 at 10:37:52 AM UTC-7, Ytai wrote: > Use PWM output to generate periodic, precisely timed trigger pulses. > > On Apr 12, 2016 10:36, "Vic Wintriss" <[email protected] <>> wrote: > I thought I was following your instructions. I am using the 4-pin sensor as > suggested and code that I thought was standard. What did you tell me to do > that I am not doing? > > On Monday, April 11, 2016 at 1:28:48 PM UTC-7, Ytai wrote: > I feel like I gave you my best advice, but you seem to insist on doing it in > a way that's not supposed to work. > > On Apr 11, 2016 1:03 PM, "Vic Wintriss" <[email protected] <>> wrote: > When I use this code: > ioio.beginBatch(); > rightStrobe.write(true); > rightStrobe.write(false); > ioio.endBatch(); > I get this beautiful result: > > <https://lh3.googleusercontent.com/-qyhO2zqtmoY/VwwBnL2vE7I/AAAAAAAAAAg/xyOLBnQBnuIvFhdp0IdprRb_qVvx6Hckw/s1600/bad.PNG> > > > But the code hangs up and I get no duration readings. > > If I use this code: > rightStrobe.write(true); > SystemClock.sleep(10); > rightStrobe.write(false); > rightDistance = (int) (rightInput.getDuration() * CONVERSION_FACTOR); > I get this awful situation, but it works. Anything less than a 10 millisec > sleep fails. > > <https://lh3.googleusercontent.com/-WEm_XYa5PF8/VwwBBYuVQ9I/AAAAAAAAAAY/15jv4SAWC3ouxIcAW9EyO7GC-SEZgtSZA/s1600/works.png> > > On Monday, April 4, 2016 at 9:25:36 PM UTC-7, Ytai wrote: > Sorry, I meant "output" from the IOIO's point of view. > Let me make sure I understand you correctly: are you claiming that you're > emitting a shorter-than-10-ms trigger pulse, the pulse gets emitted correctly > as verified by a scope, the echo pulse gets emitted correctly as verified by > a scope, but your app is blocked on waitPulseGetDuration()? If that's the > case, the only thing I can imagine is that the wait*() method is called after > the pulse has already comes back. > > Have you considered using PwmOutput as trigger, as per my suggestion? Then > you can use getDuration() and never have to block. I think this is a rather > elegant solution. > > On Mon, Apr 4, 2016 at 10:51 AM, <[email protected] <>> wrote: > The output pulse (echo pulse returning from the ultrasonic sensor) is > correct...it is milli secs long and varies proportionately with the distance. > I am not using Bluetooth...the Android is directly connected to the ioio > board. It is the trigger pulse that must be long...more than 10 > millisec...for the program to run. It hangs up with smaller trigger pulses. > I can easily generate 10 microsec trigger pulses...as observed on scope...but > then the program hangs up. Ideas? > > On Friday, April 1, 2016 at 9:55:17 PM UTC-7, Ytai wrote: > When you look at the waveforms, what do you see as the output pulse? My guess > would be that it is very narrow, in the order of a couple of usec perhaps, > and if that's indeed the case, it is likely because both messages comprising > the pulse got collapsed into the same packet over one of the slower > connections such as Bluetooth. In short, trying to achieve precise timing of > any sort at this time scale is a bad idea. It would be possible to add a > one-shot feature to the IOIO to achieve that. If you want this reading > periodically, consider using a PwmOutput to keep triggering the sensor > without your intervention at very precise pulse widths and period. Another > option would be to use a version of those sensors that emits an analog > voltage corresponding to the distance. > > On Apr 1, 2016 9:38 PM, <[email protected] <>> wrote: > It's a new iARoC (International Autonomous Robot Competition) year (2016) and > we are back to the ultrasonic sensor problem. This year I am using the 4-pin > version. The following code works: > > public void readRight() throws ConnectionLostException, InterruptedException > {//This code works with the 4-pin sensor...don't change anything! > rightStrobe.write(true); > SystemClock.sleep(10); > rightStrobe.write(false); > rightDistance = (int) (rightInput.getDuration() * CONVERSION_FACTOR); > } > > > But it will not work with less than a 10 ms long trigger pulse. I don't > understand this, because the ultrasonic sensor spec says that the trigger > pulse should be around 10 micro sec. Do you have any idea why I need such a > long trigger pulse. I verified the waveforms with a scope. It seems to be > the case with both the 4-pin and the 3-pin version. > > On Thursday, April 23, 2015 at 9:43:50 PM UTC-7, Ytai wrote: > You can add a TimerTask to interrupt the thread after some timeout. Search > the forum for examples if you need them. If you don't want to delay all the > other 5 if one is stuck you can easily run 6 sensors on 6 different threads. > > On Wed, Apr 22, 2015 at 2:42 PM, Daniel Brown <[email protected] <>> wrote: > if one PulseInput pin is open or has not received a pulse and you call > getDuration() it will block you from reading all other PulseInput pins until > it has received at least one pulse. We have found that one bad sensor will > stop you from reading any other sensor. > > > On Friday, April 10, 2015 at 6:59:52 PM UTC-6, Daniel Brown wrote: > We have a project that uses 6 distance finders, when we try to call > .getDuration() on more then one all further reading freezes. > > protected void setup() throws ConnectionLostException { > > // front left pulse and echo > echoPinfl_ = ioio_.openPulseInput(6, > PulseMode.POSITIVE); > triggerPinfl_ = ioio_.openDigitalOutput(7); > > // front middle pulse and echo > echoPinfm_ = ioio_.openPulseInput(2, > PulseMode.POSITIVE); > triggerPinfm_ = ioio_.openDigitalOutput(4); > > ... > > public void loop() throws ConnectionLostException, InterruptedException { > > > // updates front left sensor > try{ > triggerPinfl_.write(false); > Thread.sleep(5); > triggerPinfl_.write(true); > Thread.sleep(1); > triggerPinfl_.write(false); > echosecondsfl_ = (int)(echoPinfl_.getDuration() * 1000 > * 1000); > echoDistanceCmfl_ = echosecondsfl_ / 29 / 2; > }catch (ConnectionLostException e) { throw e;} > > > // updates front middle sensor > try{ > triggerPinfm_.write(false); > Thread.sleep(5); > triggerPinfm_.write(true); > Thread.sleep(1); > triggerPinfm_.write(false); > // echosecondsfm_ = (int)(echoPinfm_.getDuration() * 1000 > * 1000); //if this line is commented in all further input stops > // echoDistanceCmfm_ = echosecondsfm_ / 29 / 2; > }catch (ConnectionLostException e) { throw e;} > > ... > > On Friday, May 16, 2014 at 11:03:20 AM UTC-6, Ytai wrote: > Thanks for sharing! > > > On Thu, May 15, 2014 at 9:25 PM, <[email protected] <>> wrote: > Here is some code that works great with the Parallax Ping))) with the circuit > in the file attached (thanks to Ytai): > > private int read(DigitalOutput strobe, PulseInput input, int inputPin) throws > ConnectionLostException, InterruptedException // Order of following > statements is very important...do not change > > > { > int distance = 0; > > > ioio.beginBatch(); > > > strobe.write(true); > > > input = ioio.openPulseInput(inputPin, PulseMode.POSITIVE); > > > ioio.endBatch(); > > > SystemClock.sleep(40); > > > strobe.write(false); > > > distance += (int) (input.getDuration() * CONVERSION_FACTOR); > > > input.close(); > > > return distance; > } > > On Wednesday, April 30, 2014 10:25:27 AM UTC-7, [email protected] <> > wrote: > I'll clean up the class and post it. Hope other people can use it. Works > very solidly with the Parallax Ping))) ultrasonic module. > Great suggestion to use only one pin for both input and output. It would > even have saved me a part on the pc boards. > > On Sunday, April 27, 2014 8:24:19 PM UTC-7, Ytai wrote: > Cool. If you can create a nice class for using the PING (where the ctor gets > a IOIO and two pin numbers, and there's a getDistance() method) and share it > on this forum, it would be a decent return for our efforts :) As a side note, > since you're doing the open-close trick, you might as well use a single pin > and open-close the digital output as well and not need the driver chip, so > you can even further simplify both your code and your circuit. > > > On Sun, Apr 27, 2014 at 8:19 PM, Vic Wintriss <[email protected] <>> wrote: > I finally got it to work using your last suggestion. This code works that > you recommended works perfectly. Thanks so much for sticking with us. We've > got 30 or 40 kids using this code at the International Autonomous Robot > Competition (iARoC 2014) coming up at the end of June. Check out iaroc.org > <http://iaroc.org/>. > > triggerHigh(); > in = openPulseInput(); > sleep(20ms); // wait until the pulse input module is ready for a pulse. > triggerLow(); // kick off a measurement (falling edge does not trigger the > pulse input) > duration = in.getDuration(); > in.close(); > return duration; > > On Sunday, April 27, 2014 11:36:53 AM UTC-7, Ytai wrote: > Let me shed some light on the internals of PulseInput: > A timer is running on the IOIO set to 200Hz (5ms). > As soon as the pulse input module is opened, and every time after a pulse is > captured the module will be put in the "ready" state. > Every time the 200Hz timer triggers, each "ready" module will be activated, > i.e. put in a state where it is waiting for a new pulse. > Once activated, the module will wait forever (or until closed) for a pulse, > then measure its duration, then sends the result to the Android. > So this means that effectively no matter what the actual pulse rate is, you > will never get more than 200 reports per second. This has been done on > purpose to prevent a high frequency pulse train from saturating the > connection between the IOIO and the Android. This also means that there is a > potential "dead time" of up to 5ms after opening or between pulses, during > which a pulse would not be detected. > On the Android side, every pulse report finds its way to your PulseInput > object. You can then read it in one of three ways: > getDuration() will return the last report. It will generally not block, then > only exception is until the first report arrives. > getDurationSync() will always block until a new report comes in, then return > it. So you can be sure that the report is new. > getDurationBuffered() pulls pulses one by one from a queue. When the queue > becomes empty it behave like getDurationSync(), i.e. waits until a new report > comes in. > The Arduino approach cannot be directly applied to the IOIO API, since you > have to take into account that because of the communication between the > Android and the IOIO, much of the IOIO API has been designed to be > asynchronous in nature. If you were to bake the PING driver directly into the > IOIO firmware, you can use a similar approach to Arduino's (although you'd > probably want to implement it in a non-blocking way, since the IOIO allows > everything to be used concurrently). The reason why I have not done that is > because I tried to focus on generic use-cases rather than on one peculiar > sensor interface. > > > > On Sun, Apr 27, 2014 at 10:57 AM, Duane DeSieno <[email protected] <>> wrote: > My confusion is maybe over the way getDuration works. > Does it delay 5ms before looking for the pulse or does it look immediately? > I put 555 timer on the input pin and set it up for 3.15hz and 25.4% duty > cycle or an 80.6ms pulse every 317.4ms. > I called getDuration roughly every 100ms. It did not block and returned the > same value several times before the next pulse occurred(not what I expected). > > Looked at the Arduino approach to using the PING and they use just one pin, > changing from output after sending a pulse to ping to an input for their get > duration call. > Since they don't impose a 5ms delay, they get the duration of the return > pulse. > > Thanks for you help on this. > > Duane > > > > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > > Visit this group at http://groups.google.com/group/ioio-users > <http://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at http://groups.google.com/group/ioio-users > <http://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at http://groups.google.com/group/ioio-users > <http://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at http://groups.google.com/group/ioio-users > <http://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <>. > To post to this group, send email to [email protected] <>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > -- > You received this message because you are subscribed to the Google Groups > "ioio-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] > <mailto:[email protected]>. > To post to this group, send email to [email protected] > <mailto:[email protected]>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > -- > You received this message because you are subscribed to a topic in the Google > Groups "ioio-users" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/ioio-users/3MDLEEKtejY/unsubscribe > <https://groups.google.com/d/topic/ioio-users/3MDLEEKtejY/unsubscribe>. > To unsubscribe from this group and all its topics, send an email to > [email protected] > <mailto:[email protected]>. > To post to this group, send email to [email protected] > <mailto:[email protected]>. > Visit this group at https://groups.google.com/group/ioio-users > <https://groups.google.com/group/ioio-users>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. -- You received this message because you are subscribed to the Google Groups "ioio-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/ioio-users. For more options, visit https://groups.google.com/d/optout.
