From reed at mobilerobots.com Fri Aug 1 14:50:51 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 01 Aug 2008 14:50:51 -0400 Subject: [Aria-users] Convert SONARNL -> ARNL In-Reply-To: <20080731183539.1335.qmail@hm1079.locaweb.com.br> References: <20080731183539.1335.qmail@hm1079.locaweb.com.br> Message-ID: <48935B0B.2030900@mobilerobots.com> Hi Bruno, I don't think I understand you fully, can you say more about what you need to do? Are you referring to a log of robot positions? ARNL and SONARNL basically navigate on the fly (after the initial [first, basic] path is made). They don't save a detailed log of robot positions or motion commands, it continuously calculates the next one. However if you want to use the initial path for something, you can get it from the ArPathPlanningTask as a vector of ArPose objects. > In order to load the map sailed to the data from ARNL therefore is more accurate ... Do you mean that you want to use ARNL to generate a map, then use the map with SONARNL? This is possible and it's the best way to get a map for use with SONARNL, but it requires a robot with a laser. Reed From reed at mobilerobots.com Fri Aug 1 18:13:29 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 01 Aug 2008 18:13:29 -0400 Subject: [Aria-users] Convert SONARNL -> ARNL In-Reply-To: <20080801190401.31261.qmail@hm1079.locaweb.com.br> References: <48935B0B.2030900@mobilerobots.com> <20080801190401.31261.qmail@hm1079.locaweb.com.br> Message-ID: <48938A89.9070909@mobilerobots.com> Thanks, I see now. You can use Mapper3-Basic to create the map. Use the "line" tool to draw lines where walls are. It will show the position in the status bar at the bottom of the screen (line endpoints and length). You can also double-click a line to exactly set its position. Use this information to make the lines match the walls in your space that you measure. In View->Units you can choose mm, meters, or feet/inches. It is possible to use sonar readings to create a map, like Mapper3 does (by applying SLAM to a log of laser readings), but it's much more difficult, and produces less reliable results, than with the more precise laser, and we haven't implemented it. Maybe others on this list know of any resources about that. > Has the navigation by sonar no log that I can use the navigated points to construction of a map, even if it is not as precise when the laser? You could add your own code or write a seperate program to do this logging. ARNL uses a class called ArSickLogger to log laser readings -- this class is in ARIA so it's source code is available for you to look at and use (GPL). Reed From blomba at unione.com.br Mon Aug 4 07:13:42 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Mon, 4 Aug 2008 08:13:42 -0300 Subject: [Aria-users] Mapping with SONARNL Message-ID: <20080804111342.30965.qmail@hm1079.locaweb.com.br> Hi ARIA-USERS, I know that it?s possible to use sonar readings to create a map, but it?s much more difficult and produces less reliable results than with the more precise laser. I?d like to know if somebody on this list has implemented it or know any resources about that????? Thanks Bruno Lomba From jyotikab83 at gmail.com Thu Aug 7 17:08:08 2008 From: jyotikab83 at gmail.com (Jyotika Bahuguna) Date: Fri, 8 Aug 2008 02:38:08 +0530 Subject: [Aria-users] Wait for thread created by runAsync Message-ID: Hi, I am using Aria with MobileSim. I have a Qt thread (other than the main thread) that calls move and rotate commands. I call runAsync(true) just after connecting to the robot. My application requires me to start the QT thread once it completes the rotating/moving. Since runAsync() spawns a new thread, i require my this thread to wait for thread spawned by runAsync(). How can i do that ? Without this, my QT thread seems to return even when the runAsync thread has not finished its job and hence I restart the QT thread , and try to run the robot again, while it is already running. I tried using run() instead of runAsync(), but the robot does not move at all, which might be because there might be two QT threads at a time, which might be calling run(). Thanks in advance, Jyotika -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080808/69eda0fb/attachment.html From albertocorralesgarcia at gmail.com Thu Aug 7 18:10:44 2008 From: albertocorralesgarcia at gmail.com (Alberto Corrales Garcia) Date: Fri, 8 Aug 2008 00:10:44 +0200 Subject: [Aria-users] Wait for thread created by runAsync In-Reply-To: References: Message-ID: <1672c7d80808071510n389d4470xbd7a17d03fa80459@mail.gmail.com> Hello Jyotika, to synchronize asynchronous tasks Aria provide ArMutexand ArCondition. See "threadExample", it's a good and easy example. regards, Alberto. To 2008/8/7 Jyotika Bahuguna > Hi, > I am using Aria with MobileSim. > I have a Qt thread (other than the main thread) that calls move and rotate > commands. > I call runAsync(true) just after connecting to the robot. > My application requires me to start the QT thread once it completes the > rotating/moving. > Since runAsync() spawns a new thread, i require my this thread to wait for > thread spawned by runAsync(). > How can i do that ? > Without this, my QT thread seems to return even when the runAsync thread > has not finished its job and hence I restart the QT thread , and try to run > the robot again, while it is already running. > > I tried using run() instead of runAsync(), but the robot does not move at > all, which might be because there might be two QT threads at a time, which > might be calling run(). > > > Thanks in advance, > Jyotika > > > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com > -- -=-=-=-=-=-=-=-=-=-=-=-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-= Alberto Corrales Garc?a Coordinador DotNetClub Albacete blog: http://labloguera.net/blogs/alberto/ web: http://albertocorralesgarcia.googlepages.com PFC: http://pfc-albertocorralesgarcia.blogspot.com/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080808/ec3a7940/attachment.html From work.chandan at gmail.com Sun Aug 10 07:20:06 2008 From: work.chandan at gmail.com (chandan datta) Date: Sun, 10 Aug 2008 15:20:06 +0400 Subject: [Aria-users] ARNL:Sending motion commands to guiServer from goalClient Message-ID: <2457b8930808100420gf07f666vf23c6c50e4e2512@mail.gmail.com> Hi, Using ARNL I can send a direct motion command like setvel(10) or setrotvel() from the server program ,but if I'm using the guiServer and the goalClient to give goals to the robot ,what is the function to give a velocity command to the robot from the goalClient while its goal to the goal by a dwa planned path ? -- Regards, Chandan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080810/ebf73858/attachment.html From work.chandan at gmail.com Sun Aug 10 08:39:28 2008 From: work.chandan at gmail.com (chandan datta) Date: Sun, 10 Aug 2008 16:39:28 +0400 Subject: [Aria-users] ARNL:Sending motion commands to guiServer from goalClient In-Reply-To: <2457b8930808100420gf07f666vf23c6c50e4e2512@mail.gmail.com> References: <2457b8930808100420gf07f666vf23c6c50e4e2512@mail.gmail.com> Message-ID: <2457b8930808100539j74652204wbdf6269939cdf25e@mail.gmail.com> Hi, I checked the clientDemo.cpp example in Arnetworking examples to set the ratiodrive request ,so I wrote another client program which I can call and ask the robot to stop while the goalClient in ARNL has given the command to move to a goal in the map.I used the following code: ArNetPacket packet; packet.doubleToBuf(0.0); packet.doubleToBuf(0.0); packet.doubleToBuf(50); // use half of the robot's maximum. if(!client.dataExists("ratioDrive") ) printf("Warning: server does not have ratioDrive command, can not use drive commands!\n"); else { client.sendPacketTcp( &packet); client.requestOnce("ratioDrive", &packet); But after I stop the robot and start it again the guiServer complains about packet errors: => Sending goal "g7" to server... ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 Bad char in sync1 12 Bad char in sync1 11 Bad char in sync1 0 Bad char in sync1 70 Bad char in sync1 1 Bad char in sync1 103 Bad char in sync1 55 Bad char in sync1 0 Bad char in sync1 1 Bad char in sync1 154 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 Bad char in sync1 12 Bad char in sync1 11 Bad char in sync1 0 Bad char in sync1 70 Bad char in sync1 1 Bad char in sync1 103 Bad char in sync1 55 Bad char in sync1 0 Bad char in sync1 1 Bad char in sync1 154 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 ArNetPacketReceiverTcp::receivePacket: bad packet, bad checksum on packet 26369 Bad char in sync1 12 Bad char in sync1 11 Bad char in sync1 0 Bad char in sync1 70 Bad char in sync1 1 Bad char in sync1 103 Bad char in sync1 55 Bad char in sync1 0 Bad char in sync1 1 Bad char in sync1 154 getCommand is 0 when it probably shouldn't be What can be a possible solution to give a stop the robot from another client ( just like mobileEyes does while running it with goalClient) ? On Sun, Aug 10, 2008 at 3:20 PM, chandan datta wrote: > Hi, > Using ARNL I can send a direct motion command like setvel(10) or > setrotvel() from the server program ,but if I'm using the guiServer and the > goalClient to give goals to the robot ,what is the function to give a > velocity command to the robot from the goalClient while its goal to the goal > by a dwa planned path ? > > -- > Regards, > Chandan > -- Regards, Chandan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080810/a580b4b4/attachment-0001.html From ericavery at epix.net Sun Aug 10 15:14:48 2008 From: ericavery at epix.net (Eric Avery) Date: Sun, 10 Aug 2008 15:14:48 -0400 Subject: [Aria-users] Confused about ArRobot::findDeltaHeadingTo results Message-ID: <002101c8fb1d$5bee0cd0$0202a8c0@flowbenew> I am trying to calculate the delta heading from the robots current pose to a pose that is basically directly in front of the robot using findDeltaHeadingTo. The results I am getting are not what I'm expecting, I expect to see a very small angle i.e. +- 2 degrees and I'm getting a very large angle. Here's a code snippet and the values I am seeing at runtime. Code: ArPose *goalPose = new ArPose(goalX, goalY, goalTh); double angle = m_Robot->findDeltaHeadingTo(*goalPose); Runtime values just after the angle is computed: goalPose { myX=97.709089517593384 myY=219.53247487545013 myTh=0.00000000000000000 } m_Robot->myGlobalPose { myX=97.709086924642008 myY=219.53247674546819 myTh=94.556264125856657 } angle -130.35513765725017 I'm assuming the angle returned from findDeltaHeadingTo should be the delta heading to the goal pose (i.e. the number of degrees to apply to the heading of the robot to point in the direction of the goal pose). Am I doing something wrong or assuming the wrong coordinate system for the angle returned? Thanks Eric Avery -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080810/bdb0f0e2/attachment.html From unreal at internet.lu Mon Aug 11 06:44:05 2008 From: unreal at internet.lu (unreal at internet.lu) Date: Mon, 11 Aug 2008 12:44:05 +0200 Subject: [Aria-users] currentReadingPolar and Java Message-ID: 1022487829-473381198-1943887510@INC_SMTP_SERVER Hi, I have problems retrieving the angle information using the currentReadingPolar() method in Java - I consistently get 0 when trying to retrieve the angle of the nearest measurement. The distances get measured and returned correctly and I assume that the angle variable simply never gets written to. (I am using Aria 2.5) Any clues on this? Some more detail on the code: //Declaration of the variable to contain the angle: SWIGTYPE_p_double laserReadingAngle = new SWIGTYPE_p_double(); //get a reading in an arc specified by the from and to variables double laserReading = laser.currentReadingPolar(from, to, laserReadingAngle); //Print the result: System.out.println(laserReadingAngle.getCPtr(laserReadingAngle)); --> The result is always 0. Best, Christian From blomba at unione.com.br Mon Aug 11 06:46:08 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Mon, 11 Aug 2008 07:46:08 -0300 Subject: [Aria-users] Mapping with SONARNL Message-ID: <20080811104608.4123.qmail@hm1079.locaweb.com.br> Hi ARIA-USERS, I know that it?s possible to use sonar readings to create a map, but it?s much more difficult and produces less reliable results than with the more precise laser. I?d like to know if somebody on this list has implemented it or know any resources about that????? Thanks Bruno Lomba From work.chandan at gmail.com Mon Aug 11 08:38:36 2008 From: work.chandan at gmail.com (chandan datta) Date: Mon, 11 Aug 2008 16:38:36 +0400 Subject: [Aria-users] Mapping with SONARNL In-Reply-To: <20080811104608.4123.qmail@hm1079.locaweb.com.br> References: <20080811104608.4123.qmail@hm1079.locaweb.com.br> Message-ID: <2457b8930808110538m557bfee0pabe73c6d2a86555f@mail.gmail.com> I had worked in developing a Sonar mapping driver for the P/S/G framework back in 2006 for the HRS UGV ( http://www.hitechroboticsystemz.com/prod_ugv_agv.html ) but unfortunately the work was proprietary .There was another guy in Italy who worked in parallel to develop a plugin sonar mapping driver. here's his thesis: http://www.dcs.qmul.ac.uk/~paladini/tesimarco.pdfand you can see the Navigation architecture in page 53 . The traditional occupancy grid-based map-building algorithm with laser mapping uses a fixed probability distribution function of the laser readings and so doesn't work on sonars very well taking in account the sparse measurements. Is your sonar a Murata MA40B8R/S model with a beam aperture of approximately 45 , minimum detection range of 20cm, and maximum detection range of 6m. I'm working on a self-funded map generation web-service that will allow robot developers to obtain a map generated from their own sonar/laser/odom sensor data (standard logfile format).. let me know if people are interested !! We can also use Gutmann's algo ( proprietary ) supplied with ARNL as a benchmark for mapping accuracy... -- Regards, Chandan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080811/f57b5023/attachment.html From fernandoauat at gmail.com Mon Aug 11 09:33:52 2008 From: fernandoauat at gmail.com (Fernando Auat) Date: Mon, 11 Aug 2008 10:33:52 -0300 Subject: [Aria-users] Question about Aria In-Reply-To: <5d328b0808081313o1f61dc14q691c2bdde1b3996d@mail.gmail.com> References: <5d328b0808081313o1f61dc14q691c2bdde1b3996d@mail.gmail.com> Message-ID: <5d328b0808110633m4f62fe51h3b7150b154a568e1@mail.gmail.com> Dear people form ActivMedia, I?m working with a Pioneer 3DX bougth by Professor Dr. Mario Sarcinelli at the University of Espirito Santo, Brazil. When I try to use the laser (Sick), the compilation works well but when excecuting the program, it shows an issue. In the pdf file there is picture of that problem. I?m using the Aria 2.5.1 with the Visual C++ 2008 and the "SickRobotExample" code provided by Aria's examples folder. Would you tell me what is the problem and how do I deal with it? Sinceresly yours. Fernando. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080811/eb74d9aa/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: ParaActivMedia.pdf Type: application/pdf Size: 106951 bytes Desc: not available Url : http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080811/eb74d9aa/attachment-0001.pdf From fernandoauat at gmail.com Mon Aug 11 10:06:54 2008 From: fernandoauat at gmail.com (Fernando Auat) Date: Mon, 11 Aug 2008 11:06:54 -0300 Subject: [Aria-users] Question about Aria In-Reply-To: <5d328b0808081313o1f61dc14q691c2bdde1b3996d@mail.gmail.com> References: <5d328b0808081313o1f61dc14q691c2bdde1b3996d@mail.gmail.com> Message-ID: <5d328b0808110706h26c2f00av7ddf0f246688bd80@mail.gmail.com> Dear people form ActivMedia, I?m working with a Pioneer 3DX bougth by Professor Dr. Mario Sarcinelli at the University of Espirito Santo, Brazil. When I try to use the laser (Sick), the compilation works well but when excecuting the program, it shows an issue. In the pdf file there is picture of that problem. I?m using the Aria 2.5.1 with the Visual C++ 2008 and the "SickRobotExample" code provided by Aria's examples folder. Would you tell me what is the problem and how do I deal with it? Sinceresly yours. Fernando. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080811/5e68337f/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: ParaActivMedia.pdf Type: application/pdf Size: 106951 bytes Desc: not available Url : http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080811/5e68337f/attachment-0001.pdf From aaron123 at ksu.edu Mon Aug 11 12:09:36 2008 From: aaron123 at ksu.edu (Aaron Chleborad) Date: Mon, 11 Aug 2008 11:09:36 -0500 Subject: [Aria-users] Confused about ArRobot::findDeltaHeadingTo results In-Reply-To: <002101c8fb1d$5bee0cd0$0202a8c0@flowbenew> References: <002101c8fb1d$5bee0cd0$0202a8c0@flowbenew> Message-ID: <1218470976.48a06440d6d96@webmail.ksu.edu> Hi Eric. The solution lies in knowing that the angle you get back is in radians. If you do a conversion to degrees: (-130.35513765725017) * (PI / 180) you get -2.27512635 which is what you are expecting. Hope that helps. -- Aaron Chleborad Kansas State University Department of Computing and Information Sciences Research Assistant http://macr.cis.ksu.edu/ Quoting Eric Avery : > I am trying to calculate the delta heading from the robots current > pose to a > pose that is basically directly in front of the robot using > findDeltaHeadingTo. The results I am getting are not what I'm > expecting, I > expect to see a very small angle i.e. +- 2 degrees and I'm getting a > very > large angle. Here's a code snippet and the values I am seeing at > runtime. > > > > Code: > > ArPose *goalPose = new ArPose(goalX, goalY, goalTh); > > > > double angle = m_Robot->findDeltaHeadingTo(*goalPose); > > > > Runtime values just after the angle is computed: > > > goalPose > > { myX=97.709089517593384 myY=219.53247487545013 > myTh=0.00000000000000000 } > > > > m_Robot->myGlobalPose > > { myX=97.709086924642008 myY=219.53247674546819 > myTh=94.556264125856657 } > > > > angle > > -130.35513765725017 > > > > I'm assuming the angle returned from findDeltaHeadingTo should be the > delta > heading to the goal pose (i.e. the number of degrees to apply to the > heading > of the robot to point in the direction of the goal pose). > > Am I doing something wrong or assuming the wrong coordinate system > for the > angle returned? > > > > Thanks > > Eric Avery > > From reed at mobilerobots.com Mon Aug 11 13:23:34 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Mon, 11 Aug 2008 13:23:34 -0400 Subject: [Aria-users] currentReadingPolar and Java References: 1022487829-473381198-1943887510@INC_SMTP_SERVER Message-ID: <48A07596.8090209@mobilerobots.com> Hi Christian, I've never tried using this function in Java or Python, or any function that returns a value in a pointer argument actually. My guess is that SWIG's automatically generated wrapper function just isn't writing the result value to your laserReadingAngle variable... it's probably creating a temporary variable, assigning laserReadingAngle to it, and passing that to the C++ method, but stops there. I'll have to look at more deeply sometime, or write a separate method for use in Java that returns a small object with the two return values or something like that. Reed unreal at internet.lu wrote: > Hi, > I have problems retrieving the angle information using the currentReadingPolar() method in Java - I consistently get 0 when trying to retrieve the angle of the nearest measurement. The distances get measured and returned correctly and I assume that the angle variable simply never gets written to. (I am using Aria 2.5) > > Any clues on this? > > Some more detail on the code: > > //Declaration of the variable to contain the angle: > SWIGTYPE_p_double laserReadingAngle = new SWIGTYPE_p_double(); > > //get a reading in an arc specified by the from and to variables > double laserReading = laser.currentReadingPolar(from, to, laserReadingAngle); > > //Print the result: > System.out.println(laserReadingAngle.getCPtr(laserReadingAngle)); > > --> The result is always 0. > > > Best, > > Christian > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com > From reed at mobilerobots.com Mon Aug 11 13:23:42 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Mon, 11 Aug 2008 13:23:42 -0400 Subject: [Aria-users] Confused about ArRobot::findDeltaHeadingTo results In-Reply-To: <002101c8fb1d$5bee0cd0$0202a8c0@flowbenew> References: <002101c8fb1d$5bee0cd0$0202a8c0@flowbenew> Message-ID: <48A0759E.7070806@mobilerobots.com> Eric Avery wrote: > I am trying to calculate the delta heading from the robots current pose > to a pose that is basically directly in front of the robot using > findDeltaHeadingTo. The results I am getting are not what I?m expecting, > I expect to see a very small angle i.e. +- 2 degrees and I?m getting a > very large angle. Here?s a code snippet and the values I am seeing at > runtime. > > Hmm, this function expands to this, basically: fixAngle(radToDeg(atan2(y1-y2, x1-x2)) - theta); Where x1,y1 is your goalPose and x2,y2,theta is the robot's global pose, and fixAngle normalizes the angle to (180,-180) and radToDeg(r) is (r*180/pi) So it should be in degrees. It looks like your goal pose is right on top of the robot? (Or, the robot is now on top of the goal, rather?) If so that may be the source of the problem. I don't know what atan2(0, 0) will return, but it may be undefined, and it may produce not-very-useful values as its arguments approach 0. I'm not sure. I just ran a little test program and it seems to work out correctly: #include #include #include "Aria.h" void printDelta(const ArPose& p1, const ArPose& p2, double exp) { printf("delta angle (%.0f,%.0f,%.0f) -> (%.0f,%.0f,%.0f): %f (expected %f)\n", p1.getX(), p1.getY(), p1.getTh(), p2.getX(), p2.getY(), p2.getTh(), ArMath::subAngle(p1.findAngleTo(p2), p1.getTh()), exp); } int main(int argc, char **argv) { printf("atan2(0,0)=%f\n", atan2(0,0)); printf("atan2(0.0002,0.001)=%f\n", atan2(0.0002,0.001)); ArPose p1(0, 0, 0); ArPose p2(20, 0, 0); ArPose p3(20, 20, 0); ArPose p4(0, 20, 0); ArPose p5(20, 0, 90); ArPose p6(0, 0, 45); ArPose p7(0, -20, 0); printDelta(p1, p2, 0); printDelta(p1, p3, 45); printDelta(p1, p4, 90); printDelta(p1, p5, 0); printDelta(p6, p5, -45); printDelta(p1, p7, -90); } atan2(0,0)=0.000000 atan2(0.0002,0.001)=0.197396 delta angle (0,0,0) -> (20,0,0): 0.000000 (expected 0.000000) delta angle (0,0,0) -> (20,20,0): 45.000000 (expected 45.000000) delta angle (0,0,0) -> (0,20,0): 90.000000 (expected 90.000000) delta angle (0,0,0) -> (20,0,90): 0.000000 (expected 0.000000) delta angle (0,0,45) -> (20,0,90): -45.000000 (expected -45.000000) delta angle (0,0,0) -> (0,-20,0): -90.000000 (expected -90.000000) From fernandoauat at gmail.com Tue Aug 12 09:07:17 2008 From: fernandoauat at gmail.com (Fernando Auat) Date: Tue, 12 Aug 2008 10:07:17 -0300 Subject: [Aria-users] General questions and comments of interest to the ARIA community Message-ID: <5d328b0808120607j79971a03k428de6aded953f83@mail.gmail.com> Dear people Aria users, I?m working with a Pioneer 3DX bougth by Professor Dr. Mario Sarcinelli at the University of Espirito Santo, Brazil. When I try to use the laser (Sick), the compilation works well but when excecuting the program, it shows an issue. In the pdf file there is picture of that problem. I?m using the Aria 2.5.1 with the Visual C++ 2008 and the "SickRobotExample" code provided by Aria's examples folder. Would you tell me what is the problem and how do I deal with it? Sinceresly yours. Fernando. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080812/a9147c17/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: ParaActivMedia.pdf Type: application/pdf Size: 106952 bytes Desc: not available Url : http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080812/a9147c17/attachment-0001.pdf From unreal at internet.lu Thu Aug 14 12:32:12 2008 From: unreal at internet.lu (unreal at internet.lu) Date: Thu, 14 Aug 2008 18:32:12 +0200 Subject: [Aria-users] Fwd: FW: currentReadingPolar and Java Message-ID: 1505098528-885843844-1084467277@INC_SMTP_SERVER Thanks for your reply Reed, sorry I get back to you so late - I didnt get your reply at first and a friend only now forwarded it to me - not sure whats going on with my email there... Anyway, yes, I have the same feeling about the "not passing" of the actual value. I have written my own code now which simply samples the laser at every (sub) degree which is not ideal though. Do you have an idea when this functionality could be implemented? Many functions in Aria write to a pointer - not having access to them is quite a significant drawback (ie no angle for the laser readings...)? I know you are inundated with requests but just adding my bit ;) Thanks for the reply! Best, Christian -----Original Message----- From: aria-users-bounces at lists.mobilerobots.com [mailto:aria-users-bounces at lists.mobilerobots.com] On Behalf Of Reed Hedges Sent: 11 August 2008 18:24 Cc: aria-users at lists.mobilerobots.com Subject: Re: [Aria-users] currentReadingPolar and Java Hi Christian, I've never tried using this function in Java or Python, or any function that returns a value in a pointer argument actually. My guess is that SWIG's automatically generated wrapper function just isn't writing the result value to your laserReadingAngle variable... it's probably creating a temporary variable, assigning laserReadingAngle to it, and passing that to the C++ method, but stops there. I'll have to look at more deeply sometime, or write a separate method for use in Java that returns a small object with the two return values or something like that. Reed unreal at internet.lu wrote: > Hi, > I have problems retrieving the angle information using the currentReadingPolar() method in Java - I consistently get 0 when trying to retrieve the angle of the nearest measurement. The distances get measured and returned correctly and I assume that the angle variable simply never gets written to. (I am using Aria 2.5) > > Any clues on this? > > Some more detail on the code: > > //Declaration of the variable to contain the angle: > SWIGTYPE_p_double laserReadingAngle = new SWIGTYPE_p_double(); > > //get a reading in an arc specified by the from and to variables > double laserReading = laser.currentReadingPolar(from, to, laserReadingAngle); > > //Print the result: > System.out.println(laserReadingAngle.getCPtr(laserReadingAngle)); > > --> The result is always 0. > > > Best, > > Christian > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com > _______________________________________________ Aria-users mailing list Aria-users at lists.mobilerobots.com http://lists.mobilerobots.com/mailman/listinfo/aria-users To unsubscribe visit the above webpage or send an e-mail to: aria-users-leave at lists.mobilerobots.com --------------------------------------- From reed at mobilerobots.com Fri Aug 15 17:37:46 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 15 Aug 2008 17:37:46 -0400 Subject: [Aria-users] ARNL:Sending motion commands to guiServer from goalClient In-Reply-To: <2457b8930808100539j74652204wbdf6269939cdf25e@mail.gmail.com> References: <2457b8930808100420gf07f666vf23c6c50e4e2512@mail.gmail.com> <2457b8930808100539j74652204wbdf6269939cdf25e@mail.gmail.com> Message-ID: <48A5F72A.2020208@mobilerobots.com> chandan datta wrote: > ArNetPacket packet; > packet.doubleToBuf(0.0); > packet.doubleToBuf(0.0); > packet.doubleToBuf(50); // use half of the robot's maximum. > if(!client.dataExists("ratioDrive") ) > printf("Warning: server does not have ratioDrive command, can not > use drive commands!\n"); > else > { > client.sendPacketTcp( &packet); > client.requestOnce("ratioDrive", &packet); > Hello, does it work if you only call requestOnce, rather than both sendPacketTcp() and requestOnce()? (requestOnce sends the packet with the correct ID for the name "ratioDrive", as well as the correct checksum.) (Also, note, the robot won't drive with the 0.0 values your code uses, the third value is an overall amount (and should be between 0 and 1, so use 0.5), while the first two give the actual translation and rotation ratios within that portion of the robot's maximums.) Reed From orkun_alp at hotmail.com Sat Aug 16 07:14:58 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sat, 16 Aug 2008 14:14:58 +0300 Subject: [Aria-users] a problem about the commands of robot position Message-ID: Hi everybody, I am interested in the commands below. That I cant understand about the commands is how to calculate the change of location and angle of robot when it moves ??? I mean that in the simulation there will be a map when the robot wanders in it so wrt the map, global locations of robot can be determined (wrt to ARIA) or is it performed by calculating robot wheel's velocity or something like that. What are the calculating principles of x,y,theta when they are called in the program ???? I need to know this information because I have to use robot pose but I will find them by using odometry and kalman filter for localization. Thanks a lot... myRobot->getX(), ?????????? myRobot->getY(), ????????? myRobot->getTh(), ???????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080816/6caeb075/attachment.html From orkun_alp at hotmail.com Sun Aug 17 09:34:28 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sun, 17 Aug 2008 16:34:28 +0300 Subject: [Aria-users] a problem about the commands of getx(), gety(), getth() Message-ID: Hi everybody, I am interested in the commands below. That I cant understand about the commands is how to calculate the change of location and angle of robot when it moves ??? I mean that in the simulation there will be a map when the robot wanders in it, wrt the map global locations of robot can be determined (wrt to ARIA) or is it performed by calculating robot wheel's velocity or something like that. What are the calculating principles of x,y,theta when they are called in the program ???? I need to know this information because I have to use robot pose but I will find them by using odometry and kalman filter for localization. Thanks a lot... myRobot->getX(), ?????????? myRobot->getY(), ????????? myRobot->getTh(), ???????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080817/b9fbcbd2/attachment.html From orkun_alp at hotmail.com Sun Aug 17 14:27:51 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sun, 17 Aug 2008 21:27:51 +0300 Subject: [Aria-users] a question about the sonar firing rate Message-ID: Hi, Is there anyone who knows how the sonar firing rate is changed (default 40 ms per sonar) ?? And I cant understand that I use the 16 sonars in my project and I did not change the firing rate (i use the default rate ) and I get range samples from enviroment at a rate of 300 ms (the values are saved in a text file by a rate of 300 ms) but the time required to fire the all sonars overall is 16*40ms=640 ms . But I read the ranges of 16 sonars correctly when using the Aria platform. In the real word is it possible to read range of 16 sonars when it is sampled by a lower rate than 640 ms ??? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080817/6e23b777/attachment.html From jyotikab83 at gmail.com Wed Aug 20 06:47:54 2008 From: jyotikab83 at gmail.com (Jyotika Bahuguna) Date: Wed, 20 Aug 2008 16:17:54 +0530 Subject: [Aria-users] About wakeAllRunExitWaitingThreads Message-ID: Hi, As far i understood wakeAllRunExitWaitingThreads() function wakes up all the threads which explicitly called waitForRunExit(). What about the thread that called runAsync() ? The implementation of runAsync() shows that it creates a new AsyncTask which is joinable and is of higher priority. Which i understans means, that the thread calling runAsync would wait for the robot's sync loop to finish/return/exit. If i use wakeAllRunExitWaitingThreads() function, would it also wake up the thread that called runAsync ? Thanks in advance, Jyotika -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080820/ebf58644/attachment.html From reed at mobilerobots.com Wed Aug 20 11:44:18 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 20 Aug 2008 11:44:18 -0400 Subject: [Aria-users] a problem about the commands of robot position In-Reply-To: References: Message-ID: <48AC3BD2.8020903@mobilerobots.com> Orkun ALP wrote: > I am interested in the commands below. That I cant understand about the > commands is how to calculate the change of location and angle of robot > when it moves ??? I mean that in the simulation there will be a map > when the robot wanders in it so wrt the map, global locations of > robot can be determined (wrt to ARIA) or is it performed by > calculating robot wheel's velocity or something like that. What > are the calculating principles of x,y,theta when they are called in the > program ???? I need to know this information because I have to use robot > pose but I will find them by using odometry and kalman filter for > localization. The robot's microcontroller is calculating these based on wheel encoder ticks. More about that is in the robot operations manual. In the case of MobileSim, it doesn't simulate encoders, instead it just extrapolates a new position for the robot continuously based on its velocities and acceleration, and also applies some artificial error to the position of the simulated robot as it moves. The X, Y and Theta values are sent by the robot in the SIP packet and received by ArRobot, which uses them to update its robot pose. (Although it doesn't just provide those values, it compares them to the previous values to prevent discontinuities when the robot's values reach the 16 bit limit.) If you can, I recommend adapting your algorithm to work with the pose values, since they are easier to work with and may also include corrections using a gyro if your robot has one. But if you need the raw encoder values, it is possible to specially request them from the robot. See the requestEncoderPackets(), getLeftEncoder() and getRightEncoder() methods of ArRobot. (MobileSim can't provide this data of course since it doesn't simulate encoders.) Reed -- Reed Hedges Developer Software Support MobileRobots Inc. Find downloads, documentation, FAQ, mailing lists and more at: From reed at mobilerobots.com Wed Aug 20 11:51:54 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 20 Aug 2008 11:51:54 -0400 Subject: [Aria-users] a question about the sonar firing rate In-Reply-To: References: Message-ID: <48AC3D9A.4090007@mobilerobots.com> It is possible to change the order and rate of the sonar by reconfiguring the robot's firmware. Use the firmware configuration tool for your robot to do so (for example, if your robot has an SH microcontroller running ARCOS, use ARCOScf). See the robot operations manual for details. Increasing the firing rate of the sonar will reduce the range of detection of the sonar. (Note, I think it may also be possible for adjacent sonar to interfere with each other if the rate is very high as well, but if you think this is happening you can change the firing order.) The sonar data stored in ArRobot and ArSonarDevice in ARIA are the most recently received value for each transducer. (And partial data may be sent by the robot to ARIA in a SIP, e.g. one SIP may contain data for transducers 1-8 and the next 8-16.) So if ARIA is receiving data at 100hz, and you are checking asynchronously at a faster rate, then some values may have been received in the most recent SIP packet, some in the packet before that, etc. There is a flag that indicates whether a particular value is from the most recent packet or not (isSonarNew()). Reed Orkun ALP wrote: > Hi, > > Is there anyone who knows how the sonar firing rate is changed (default > 40 ms per sonar) ?? And I cant understand that I use the 16 sonars in > my project and I did not change the firing rate (? use the default rate > ) and I get range samples from enviroment at a rate of 300 ms (the > values are saved in a text file by a rate of 300 ms) but the > time required to fire the all sonars overall is 16*40ms=640 ms . But I > read the ranges of 16 sonars correctly when using the Aria platform. In > the real word is it possible to read range of 16 sonars when it is > sampled by a lower rate than 640 ms ??? > > > From reed at mobilerobots.com Wed Aug 20 12:01:28 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 20 Aug 2008 12:01:28 -0400 Subject: [Aria-users] About wakeAllRunExitWaitingThreads In-Reply-To: References: Message-ID: <48AC3FD8.3080605@mobilerobots.com> Hello, Can you explain more about what you are trying to do? You should normally not need to call wakeAllRunExitWaitingThreads(). runAsync()'s thread does end when the robot's loop exits or aborts, which you can stop by calling stopRunning(), or can automatically happen if the robot connection fails or is closed (unless runAsync(false) is used). The only thing inside ArRobot that uses waitForRunExit() is the stopRunning() method itself. (As far as I can tell, the "wake" methods signal ArCondition objects, and do not affect the state of the actual native OS threads; the "wake" terminology is not related to sleeping/woken threads e.g. in Java or Windows.) Does this explain it? I can find out more about how this aspect of ARIA works if you want. Reed Jyotika Bahuguna wrote: > Hi, > As far i understood wakeAllRunExitWaitingThreads() function wakes up all > the threads which explicitly called waitForRunExit(). > What about the thread that called runAsync() ? The implementation of > runAsync() shows that it creates a new AsyncTask which is joinable and > is of higher priority. Which i understans means, that the thread calling > runAsync would wait for the robot's sync loop to finish/return/exit. > If i use wakeAllRunExitWaitingThreads() function, would it also wake up > the thread that called runAsync ? > From jyotikab83 at gmail.com Wed Aug 20 15:22:03 2008 From: jyotikab83 at gmail.com (Jyotika Bahuguna) Date: Thu, 21 Aug 2008 00:52:03 +0530 Subject: [Aria-users] Fwd: About wakeAllRunExitWaitingThreads In-Reply-To: References: Message-ID: ---------- Forwarded message ---------- From: Jyotika Bahuguna Date: Wed, Aug 20, 2008 at 4:17 PM Subject: About wakeAllRunExitWaitingThreads To: aria-users Hi, As far i understood wakeAllRunExitWaitingThreads() function wakes up all the threads which explicitly called waitForRunExit(). What about the thread that called runAsync() ? The implementation of runAsync() shows that it creates a new AsyncTask which is joinable and is of higher priority. Which i understans means, that the thread calling runAsync would wait for the robot's sync loop to finish/return/exit. If i use wakeAllRunExitWaitingThreads() function, would it also wake up the thread that called runAsync ? Thanks in advance, Jyotika -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080821/4842279e/attachment.html From phano002 at gmail.com Fri Aug 22 11:09:55 2008 From: phano002 at gmail.com (Gridsada Phanomchoeng) Date: Fri, 22 Aug 2008 10:09:55 -0500 Subject: [Aria-users] how to use Arcos Message-ID: Hello All, I am a new user for AT-3. I have read AT-3 manuals. I understand that we have two options to program the robot. The first one is Aria (program by visual C++) and the next one is ARCOS. However, I do not understand how to program the robot by ARCOS. Pioneer 3 Operations Manual shows only commands to control the robot. I do not know how to implement those commands. Do I need any software to write a program? Where can I get the software? Could you show me a program example? Gridsada Phanomchoeng -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080822/03e08b71/attachment.html From phano002 at gmail.com Fri Aug 22 11:45:36 2008 From: phano002 at gmail.com (Gridsada Phanomchoeng) Date: Fri, 22 Aug 2008 10:45:36 -0500 Subject: [Aria-users] How to use Arcos Message-ID: Hello All, I am a new user for AT-3. I have read AT-3 manuals. I understand that we have two options to program the robot. The first one is Aria (program by visual C++) and the next one is ARCOS. However, I do not understand how to program the robot by ARCOS. Pioneer 3 Operations Manual shows only commands to control the robot. I do not know how to implement those commands. Do I need any software to write a program? Where can I get the software? Could you show me a program example? Gridsada -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080822/bad619fb/attachment.html From reed at mobilerobots.com Fri Aug 22 14:44:02 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 22 Aug 2008 14:44:02 -0400 Subject: [Aria-users] Wait for thread created by runAsync In-Reply-To: References: Message-ID: <48AF08F2.5020703@mobilerobots.com> Jyotika Bahuguna wrote: > Hi, > I am using Aria with MobileSim. > I have a Qt thread (other than the main thread) that calls move and > rotate commands. > I call runAsync(true) just after connecting to the robot. > My application requires me to start the QT thread once it completes the > rotating/moving. > Since runAsync() spawns a new thread, i require my this thread to wait > for thread spawned by runAsync(). > How can i do that ? To wait for ArRobot::runAsync() to finish, call ArRobot::waitForRunExit(). To end runAsync(), call ArRobot::stopRunning(). > I tried using run() instead of runAsync(), but the robot does not move > at all, which might be because there might be two QT threads at a time, > which might be calling run(). Probably what is happening here is that execution is not continuing past the call to run(), which blocks until stopRunning() is called from another thread, or the connection is closed or has an error. Reed From aria-users at mobilerobots.com Fri Aug 22 15:04:52 2008 From: aria-users at mobilerobots.com (aria-users at mobilerobots.com) Date: Fri, 22 Aug 2008 15:04:52 -0400 (EDT) Subject: [Aria-users] *****SPAM-E***** RE: SALE 84% OFF Message-ID: <20080822140446.5303.qmail@host-79-164-109-247.qwerty.ru> An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080822/996ba7dc/attachment.html From reed at mobilerobots.com Fri Aug 22 15:05:21 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 22 Aug 2008 15:05:21 -0400 Subject: [Aria-users] How to use Arcos In-Reply-To: References: Message-ID: <48AF0DF1.20603@mobilerobots.com> Hello, ARCOS is the firmware running on the robot's internal microcontroller. It receives those commands over a serial port (RS-232) connection, and it also sends data packets back using that connection. If your robot has an onboard computer, it is already connected (on COM1 or /dev/ttyS0). If your robot does not have an onboard computer, but has a wireless-serial bridge device (Lantronix WiBox), then that device is connected to the microcontroller, and you can use a TCP connection over a wireless network to send commands and receive data. If your robot has neither, use the SERIAL port on the top control panel. ARIA is a library that runs on a seperate computer connected to the robot through that serial RS-232 connection. The classes in ARIA open that serial port connection and send and receive data to and from it using the protocol described in the manual. So if you are using ARIA, you just need to set up the connection (use ArSimpleConnector) and then you can use the methods in ArRobot and other classes, and they will take care of sending commands and receiving and storing data. (If you are using a wireless-serial bridge, you must use a command-line option to tell ARIA what the IP address is of the device on the network. See the Quick Start chapter of the robot's operations manual.) If you have an onboard computer, ARIA has been installed (in C:\Program Files\MobileRobots\Aria or the start menu on Windows, or /usr/local/Aria on Linux). Or you can download it from http://robots.mobilerobots.com/ARIA Example programs and more documentation is included with ARIA. Hope this helps, Reed Gridsada Phanomchoeng wrote: > Hello All, > > I am a new user for AT-3. I have read AT-3 manuals. I understand that we > have two options to program the robot. The first one is Aria (program by > visual C++) and the next one is ARCOS. > > However, I do not understand how to program the robot by ARCOS. Pioneer 3 > Operations Manual shows only commands to control the robot. I do not know > how to implement those commands. Do I need any software to write a program? > Where can I get the software? Could you show me a program example? > > Gridsada > > From orkun_alp at hotmail.com Fri Aug 22 17:41:59 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sat, 23 Aug 2008 00:41:59 +0300 Subject: [Aria-users] Fw: a question about the sonar firing rate Message-ID: 90 50 30 0 -10 -30 -50 -90 -90 -110 -130 -170 ---->angle 16321 1181 1351 2402 2159 2283 5000 5000 16321 585 2865 ---->ranges 170 -150 -130 90 ------------>angles 1039 1558 5000 5000 ---------> ranges Hi, The values above are samples saved in a file by sonar range (sonars mounted on pioneer 3dx) according to the enviroment. What I cant understand is what the 16321 means?? What I mean, as far as I know the max range (when the rate of sonar firing is 40 ms ) which can be measured from the furthest point is 5000 mm and 16321 is out of range . And in the continue of my file the value of 16312 (or similar values) repeats a couple of times. In which conditions does this wrong reading occur?? Has anybody encounter such a situation before??? What do you suggest me to do??? -------------------------------------------------- From: "Reed Hedges" Sent: Wednesday, August 20, 2008 6:51 PM Cc: Subject: Re: [Aria-users] a question about the sonar firing rate > > It is possible to change the order and rate of the sonar by reconfiguring > the > robot's firmware. Use the firmware configuration tool for your robot to do > so > (for example, if your robot has an SH microcontroller running ARCOS, use > ARCOScf). See the robot operations manual for details. Increasing the > firing > rate of the sonar will reduce the range of detection of the sonar. (Note, > I > think it may also be possible for adjacent sonar to interfere with each > other if > the rate is very high as well, but if you think this is happening you can > change > the firing order.) > > The sonar data stored in ArRobot and ArSonarDevice in ARIA are the most > recently > received value for each transducer. (And partial data may be sent by the > robot > to ARIA in a SIP, e.g. one SIP may contain data for transducers 1-8 and > the next > 8-16.) So if ARIA is receiving data at 100hz, and you are checking > asynchronously at a faster rate, then some values may have been received > in the > most recent SIP packet, some in the packet before that, etc. There is a > flag > that indicates whether a particular value is from the most recent packet > or not > (isSonarNew()). > > Reed > > > > Orkun ALP wrote: >> Hi, >> >> Is there anyone who knows how the sonar firing rate is changed (default >> 40 ms per sonar) ?? And I cant understand that I use the 16 sonars in >> my project and I did not change the firing rate (? use the default rate >> ) and I get range samples from enviroment at a rate of 300 ms (the >> values are saved in a text file by a rate of 300 ms) but the >> time required to fire the all sonars overall is 16*40ms=640 ms . But I >> read the ranges of 16 sonars correctly when using the Aria platform. In >> the real word is it possible to read range of 16 sonars when it is >> sampled by a lower rate than 640 ms ??? >> >> >> > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com From orkun_alp at hotmail.com Fri Aug 22 17:44:47 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sat, 23 Aug 2008 00:44:47 +0300 Subject: [Aria-users] a question about the sonar firing rate Message-ID: > 90 50 30 > -10 -30 -50 -90 -90 -110 > -130 -170 ---->angle > 16321 1181 1351 2402 2159 2283 5000 5000 > 16321 585 2865 1092 ---->ranges > > 170 -150 -130 90 ------------>angles > 1039 1558 5000 5000 ---------> ranges > > Hi, > > The values above are samples saved in a file by sonar range (sonars > mounted on pioneer 3dx) according to the enviroment. What I cant > understand is what the 16321 means?? What I mean, as far as I know the max > range (when the rate of sonar firing is 40 ms ) which can be measured > from the furthest point is 5000 mm and 16321 is out of range . And in the > continue of my file the value of 16312 (or similar values) repeats a > couple of times. In which conditions does this wrong reading occur?? Has > anybody encounter such a situation before??? What do you suggest me to > do??? > > > -------------------------------------------------- > From: "Reed Hedges" > Sent: Wednesday, August 20, 2008 6:51 PM > Cc: > Subject: Re: [Aria-users] a question about the sonar firing rate > >> >> It is possible to change the order and rate of the sonar by reconfiguring >> the >> robot's firmware. Use the firmware configuration tool for your robot to >> do so >> (for example, if your robot has an SH microcontroller running ARCOS, use >> ARCOScf). See the robot operations manual for details. Increasing the >> firing >> rate of the sonar will reduce the range of detection of the sonar. (Note, >> I >> think it may also be possible for adjacent sonar to interfere with each >> other if >> the rate is very high as well, but if you think this is happening you can >> change >> the firing order.) >> >> The sonar data stored in ArRobot and ArSonarDevice in ARIA are the most >> recently >> received value for each transducer. (And partial data may be sent by >> the robot >> to ARIA in a SIP, e.g. one SIP may contain data for transducers 1-8 and >> the next >> 8-16.) So if ARIA is receiving data at 100hz, and you are checking >> asynchronously at a faster rate, then some values may have been received >> in the >> most recent SIP packet, some in the packet before that, etc. There is a >> flag >> that indicates whether a particular value is from the most recent packet >> or not >> (isSonarNew()). >> >> Reed >> >> >> >> Orkun ALP wrote: >>> Hi, >>> >>> Is there anyone who knows how the sonar firing rate is changed (default >>> 40 ms per sonar) ?? And I cant understand that I use the 16 sonars in >>> my project and I did not change the firing rate (? use the default rate >>> ) and I get range samples from enviroment at a rate of 300 ms (the >>> values are saved in a text file by a rate of 300 ms) but the >>> time required to fire the all sonars overall is 16*40ms=640 ms . But I >>> read the ranges of 16 sonars correctly when using the Aria platform. In >>> the real word is it possible to read range of 16 sonars when it is >>> sampled by a lower rate than 640 ms ??? >>> >>> >>> >> _______________________________________________ >> Aria-users mailing list >> Aria-users at lists.mobilerobots.com >> http://lists.mobilerobots.com/mailman/listinfo/aria-users >> >> To unsubscribe visit the above webpage or send an e-mail to: >> >> aria-users-leave at lists.mobilerobots.com > From orkun_alp at hotmail.com Fri Aug 22 17:46:52 2008 From: orkun_alp at hotmail.com (Orkun ALP) Date: Sat, 23 Aug 2008 00:46:52 +0300 Subject: [Aria-users] a question about the sonar firing rate Message-ID: 16321 1181 1351 2402 2159 2283 5000 5000 16321 585 2865 1092 1039 1558 5000 5000 Hi, >> >> The values above are samples saved in a file by sonar range (sonars >> mounted on pioneer 3dx) according to the enviroment. What I cant >> understand is what the 16321 means?? What I mean, as far as I know the >> max range (when the rate of sonar firing is 40 ms ) which can be >> measured from the furthest point is 5000 mm and 16321 is out of range . >> And in the continue of my file the value of 16312 (or similar values) >> repeats a couple of times. In which conditions does this wrong reading >> occur?? Has anybody encounter such a situation before??? What do you >> suggest me to do??? >> -------------------------------------------------- >> From: "Reed Hedges" >> Sent: Wednesday, August 20, 2008 6:51 PM >> Cc: >> Subject: Re: [Aria-users] a question about the sonar firing rate >> >>> >>> It is possible to change the order and rate of the sonar by >>> reconfiguring the >>> robot's firmware. Use the firmware configuration tool for your robot to >>> do so >>> (for example, if your robot has an SH microcontroller running ARCOS, use >>> ARCOScf). See the robot operations manual for details. Increasing the >>> firing >>> rate of the sonar will reduce the range of detection of the sonar. >>> (Note, I >>> think it may also be possible for adjacent sonar to interfere with each >>> other if >>> the rate is very high as well, but if you think this is happening you >>> can change >>> the firing order.) >>> >>> The sonar data stored in ArRobot and ArSonarDevice in ARIA are the most >>> recently >>> received value for each transducer. (And partial data may be sent by >>> the robot >>> to ARIA in a SIP, e.g. one SIP may contain data for transducers 1-8 and >>> the next >>> 8-16.) So if ARIA is receiving data at 100hz, and you are checking >>> asynchronously at a faster rate, then some values may have been received >>> in the >>> most recent SIP packet, some in the packet before that, etc. There is a >>> flag >>> that indicates whether a particular value is from the most recent packet >>> or not >>> (isSonarNew()). >>> >>> Reed >>> >>> >>> >>> Orkun ALP wrote: >>>> Hi, >>>> >>>> Is there anyone who knows how the sonar firing rate is changed >>>> (default >>>> 40 ms per sonar) ?? And I cant understand that I use the 16 sonars in >>>> my project and I did not change the firing rate (? use the default rate >>>> ) and I get range samples from enviroment at a rate of 300 ms (the >>>> values are saved in a text file by a rate of 300 ms) but the >>>> time required to fire the all sonars overall is 16*40ms=640 ms . But I >>>> read the ranges of 16 sonars correctly when using the Aria platform. In >>>> the real word is it possible to read range of 16 sonars when it is >>>> sampled by a lower rate than 640 ms ??? >>>> >>>> >>>> >>> _______________________________________________ >>> Aria-users mailing list >>> Aria-users at lists.mobilerobots.com >>> http://lists.mobilerobots.com/mailman/listinfo/aria-users >>> >>> To unsubscribe visit the above webpage or send an e-mail to: >>> >>> aria-users-leave at lists.mobilerobots.com >> From blomba at unione.com.br Sat Aug 23 16:07:41 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Sat, 23 Aug 2008 17:07:41 -0300 Subject: [Aria-users] Communication - Interface x MobileEyes Message-ID: <20080823200741.19021.qmail@hm1079.locaweb.com.br> Hi ARIA/Pionner users, Could you help me please? I?m developing an software that loads the site of Books?s Library of my university. The user to select a book, can send a request to the robot go to this particular book... I created the Library?s map "by hand", and added the appropriate goals... I?d like to know How the communication of my application with MobileEyes (I send the x,y robot positions of respective goal) ... I don?t mind how that communication, or how to submit these data for the position of the robot and do it with what he will automatically get this Goal, if selected as the GO TO menu of Mobile Eyes on an automatic (with the data from my interface) Regards Bruno Lomba From blomba at unione.com.br Mon Aug 25 22:51:31 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Mon, 25 Aug 2008 23:51:31 -0300 Subject: [Aria-users] Problem - Compiling source code Message-ID: <20080826025131.2648.qmail@hm1079.locaweb.com.br> Hi Aria-users... I?m not able to generate a solution of my source code .. Has somebody any idea of what can I be? Look at the file Img1 attached that I can compile, but I can?t Buil Solution in Img2... It's error in .EXE .. Thank you Bruno Lomba -------------- next part -------------- A non-text attachment was scrubbed... Name: Img1.jpg Type: image/jpeg Size: 187066 bytes Desc: not available Url : http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080825/3182db46/attachment-0002.jpg -------------- next part -------------- A non-text attachment was scrubbed... Name: Img2.jpg Type: image/jpeg Size: 164956 bytes Desc: not available Url : http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080825/3182db46/attachment-0003.jpg From reed at mobilerobots.com Wed Aug 27 17:37:14 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 27 Aug 2008 17:37:14 -0400 Subject: [Aria-users] Fw: a question about the sonar firing rate In-Reply-To: References: Message-ID: <48B5C90A.3080908@mobilerobots.com> Orkun ALP wrote: > 90 50 30 > 0 -10 -30 -50 -90 -90 > -110 -130 -170 ---->angle > 16321 1181 1351 2402 2159 2283 5000 5000 16321 585 2865 ---->ranges > > 170 -150 -130 90 ------------>angles > 1039 1558 5000 5000 ---------> ranges > > Hi, > > The values above are samples saved in a file by sonar range (sonars mounted > on pioneer 3dx) according to the enviroment. What I cant understand is what > the 16321 means?? Not sure, unless it's coming from an uninitialized variable, or it's due to a known bug in ARCOS version 1.5 - try upgrading if you have that version or lower. (You can always treat it the same as 5000 in your code if it only happens for sonars not detecting anything.) Reed From reed at mobilerobots.com Wed Aug 27 17:39:36 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 27 Aug 2008 17:39:36 -0400 Subject: [Aria-users] Communication - Interface x MobileEyes In-Reply-To: <20080823200741.19021.qmail@hm1079.locaweb.com.br> References: <20080823200741.19021.qmail@hm1079.locaweb.com.br> Message-ID: <48B5C998.8090305@mobilerobots.com> Hello, in the example programs (guiServer), new goal requests from MobileEyes are handled by the ArServerModeGoto class. This automatically gives ARNL the new goal. The list of goals comes from the map. If ArServerModeGoto doesn't do what you want, you can make your own class to use instead, but which has the same ArNetworking interface. The code for ArServerModeGoto is included with ARNL in ArServerClasses.h/cpp for reference. Reed Bruno Teixeira Lomba wrote: > Hi ARIA/Pionner users, > > Could you help me please? > > I?m developing an software that loads the site of Books?s Library of my university. The user to select a book, can send a request to the robot go to this particular book... > > I created the Library?s map "by hand", and added the appropriate goals... > > I?d like to know How the communication of my application with MobileEyes (I send the x,y robot positions of respective goal) ... > > I don?t mind how that communication, or how to submit these data for the position of the robot and do it with what he will automatically get this Goal, if selected as the GO TO menu of Mobile Eyes on an automatic (with the data from my interface) > > Regards > > Bruno Lomba > From reed at mobilerobots.com Wed Aug 27 17:43:08 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Wed, 27 Aug 2008 17:43:08 -0400 Subject: [Aria-users] Problem - Compiling source code In-Reply-To: <20080826025131.2648.qmail@hm1079.locaweb.com.br> References: <20080826025131.2648.qmail@hm1079.locaweb.com.br> Message-ID: <48B5CA6C.3070206@mobilerobots.com> Bruno Teixeira Lomba wrote: > Hi Aria-users... > > I?m not able to generate a solution of my source code .. > > Has somebody any idea of what can I be? > > Look at the file Img1 attached that I can compile, but I can?t Buil Solution in Img2... > > It's error in .EXE .. I don't know exactly what's wrong with your program but I'd start by rechecking the project settings (General, Linker and Debugging sections especially). What is the second image the result of? Building? Starting the program in the debugger? Reed From mjohnson at ihmc.us Wed Aug 27 21:15:10 2008 From: mjohnson at ihmc.us (Matthew Johnson) Date: Wed, 27 Aug 2008 20:15:10 -0500 Subject: [Aria-users] Robot flash In-Reply-To: <27AED437-C0A0-47C2-8A6E-2893DEA74A39@ihmc.us> References: <27AED437-C0A0-47C2-8A6E-2893DEA74A39@ihmc.us> Message-ID: <4FE9D8DDD61147C1B612F8D128038C9A@mjdell> Wayne, I am familiar with what you are talking about, but have never attempted to do it myself. The best place to get this question answered is the aria users group (aria-users at mobilerobots.com) on the MobileRobots website. There is probably a post about this already. If not post one and they will get back to fairly quickly. This is the kind of issue they are good at answering. I am not sure the calibration will improve your accuracy that much (a good question for the user group). Odometry is well known for its unbounded error growth. Calibrating may slow the process, but I predict not by much. The big problem is wheel slippage and inaccurate turns. Once you heading is off by a bit, obviously your XY will develop a big error proportional to your distance. The more you turn, the greater the heading error. You can reduce this error considerably by adding a stabilizing gyro and/or compass. MobileRobots sells them or you could get them from elsewhere. In general localization is ALWAYS a problem. There are a number of possible solutions (each with varying accuracy). Your domain of operation, desired accuracy, and budget will determine what would be best. If you provide me a description of what you are trying to do, I can suggest some options. Matt _____ From: Marco Carvalho [mailto:mcarvalho at ihmc.us] Sent: Wednesday, August 27, 2008 7:43 PM To: wakraus at rockwellcollins.com Cc: Marco Arguedas; Carlos A. Perez; Kelli Hoback; Matthew Johnson Subject: Re: Robot flash Hi Wayne, You're not referring to the linksys radios are you? I'm not really familiar with that problem but I'm copying Matt Johnson here - he is the very familiar with the robots and have some suggestions for you. Matt, have you seen that problem before? Any suggestions? Thanks, Marco Marco Carvalho, Ph.D. Research Scientist Institute for Human and Machine Cognition www.ihmc.us On Aug 27, 2008, at 9:56 AM, wakraus at rockwellcollins.com wrote: Guys I was looking into the possibility of calibrating these robots to get better accuracy. I tried using the ARCOScf routine. You work in a temporary ram database and then tell it to save to flash. When I tried this the program failed and returned a message saying that the flash erase failed and quit without trying to save the values. However, it appears that the erase did in fact work and the flash has been wiped out. Have you guys ever tried to use this software? Have you seen this behavior? At this point I need working software just to restore the defaults and get this robot operational again. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080827/f8e0e984/attachment.html From blomba at unione.com.br Thu Aug 28 06:45:51 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Thu, 28 Aug 2008 07:45:51 -0300 Subject: [Aria-users] Problem - Compiling source code In-Reply-To: <48B5CA6C.3070206@mobilerobots.com> References: <48B5CA6C.3070206@mobilerobots.com> Message-ID: <20080828104551.12022.qmail@hm1079.locaweb.com.br> Hello Reed, I'm trying to compile a source code to run in the Simulator... I think that it??s necessary to generate a . EXE, don??t? How can I get run a program (.cpp)? Just as we have the examples in the portfolios of ARIA ...??? Bruno Lomba Wed, 27 Aug 2008 17:43:08 -0400, Reed Hedges escreveu: > Bruno Teixeira Lomba wrote: > > Hi Aria-users... > > > > I??m not able to generate a solution of my source code .. > > > > Has somebody any idea of what can I be? > > > > Look at the file Img1 attached that I can compile, but I can??t Buil Solution in Img2... > > > > It's error in .EXE .. > > I don't know exactly what's wrong with your program but I'd start by rechecking > the project settings (General, Linker and Debugging sections especially). > > What is the second image the result of? Building? Starting the program in the > debugger? > > Reed > > > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com > From reed at mobilerobots.com Thu Aug 28 10:05:02 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Thu, 28 Aug 2008 10:05:02 -0400 Subject: [Aria-users] Problem - Compiling source code In-Reply-To: <20080828104551.12022.qmail@hm1079.locaweb.com.br> References: <48B5CA6C.3070206@mobilerobots.com> <20080828104551.12022.qmail@hm1079.locaweb.com.br> Message-ID: <48B6B08E.1070605@mobilerobots.com> Bruno Teixeira Lomba wrote: > Hello Reed, > > I'm trying to compile a source code to run in the Simulator... > > I think that it??s necessary to generate a . EXE, don??t? > > How can I get run a program (.cpp)? Just as we have the examples in the portfolios of ARIA ...??? > Yes, there is a Visual Studio project file for each example. Open one of them to see what its settings are, and use those settings in a new Visual C++ project for your program. (There is also a summary of important settings in README.txt I think.) Reed From mzwinderman at gmail.com Thu Aug 28 11:43:28 2008 From: mzwinderman at gmail.com (Matthijs Zwinderman) Date: Thu, 28 Aug 2008 11:43:28 -0400 Subject: [Aria-users] PTU unit not panning fully to the right Message-ID: <750AB827-9443-4140-9375-7E348D99F759@gmail.com> Hi everybody, Currently we are working on a PeopleBot with a Bumblebee camera mounted on the standard Directed Perceptions PTU, version PTU 46-17.5. We are using Aria version 2.5.1 When running either the demo or the PTU example, we can control the pan and tilt of the PTU through Aria. However, we cannot pan the unit to the right for more than -51 degrees. During the initialization and the self-test, the unit can pan further than -51 degrees, so it is most likely not a hardware problem. There are no problems tilting the unit or panning it to the left. We think this is a very serious limitation. Has anyone experienced the same problem, and is there a solution for this? Thanks in advance! Matthijs Zwinderman From blomba at unione.com.br Thu Aug 28 06:45:51 2008 From: blomba at unione.com.br (=?ISO-8859-1?Q?Bruno Teixeira Lomba?=) Date: Thu, 28 Aug 2008 07:45:51 -0300 Subject: [Aria-users] [Pioneer-users] Problem - Compiling source code In-Reply-To: <48B5CA6C.3070206@mobilerobots.com> References: <48B5CA6C.3070206@mobilerobots.com> Message-ID: <20080828104551.12022.qmail@hm1079.locaweb.com.br> Hello Reed, I'm trying to compile a source code to run in the Simulator... I think that it??s necessary to generate a . EXE, don??t? How can I get run a program (.cpp)? Just as we have the examples in the portfolios of ARIA ...??? Bruno Lomba Wed, 27 Aug 2008 17:43:08 -0400, Reed Hedges escreveu: > Bruno Teixeira Lomba wrote: > > Hi Aria-users... > > > > I??m not able to generate a solution of my source code .. > > > > Has somebody any idea of what can I be? > > > > Look at the file Img1 attached that I can compile, but I can??t Buil Solution in Img2... > > > > It's error in .EXE .. > > I don't know exactly what's wrong with your program but I'd start by rechecking > the project settings (General, Linker and Debugging sections especially). > > What is the second image the result of? Building? Starting the program in the > debugger? > > Reed > > > _______________________________________________ > Aria-users mailing list > Aria-users at lists.mobilerobots.com > http://lists.mobilerobots.com/mailman/listinfo/aria-users > > To unsubscribe visit the above webpage or send an e-mail to: > > aria-users-leave at lists.mobilerobots.com > _______________________________________________ Pioneer-users mailing list Pioneer-users at lists.mobilerobots.com http://lists.mobilerobots.com/mailman/listinfo/pioneer-users To unsubscribe visit the above webpage or send an e-mail to: pioneer-users-leave at lists.mobilerobots.com From mzwinderman at gmail.com Thu Aug 28 13:50:46 2008 From: mzwinderman at gmail.com (Matthijs Zwinderman) Date: Thu, 28 Aug 2008 13:50:46 -0400 Subject: [Aria-users] PTU unit not panning fully to the right In-Reply-To: <750AB827-9443-4140-9375-7E348D99F759@gmail.com> References: <750AB827-9443-4140-9375-7E348D99F759@gmail.com> Message-ID: Hi everybody, Currently we are working on a PeopleBot with a Bumblebee camera mounted on the standard Directed Perceptions PTU, version PTU 46-17.5. We are using Aria version 2.5.1 When running either the demo or the PTU example, we can control the pan and tilt of the PTU through Aria. However, we cannot pan the unit to the right for more than -51 degrees. During the initialization and the self-test, the unit can pan further than -51 degrees, so it is most likely not a hardware problem. There are no problems tilting the unit or panning it to the left. We think this is a very serious limitation. Has anyone experienced the same problem, and is there a solution for this? Thanks in advance! Matthijs Zwinderman From wjxnet_126 at 126.com Fri Aug 29 07:56:52 2008 From: wjxnet_126 at 126.com (wjxnet_126) Date: Fri, 29 Aug 2008 19:56:52 +0800 (CST) Subject: [Aria-users] Why MobileSim always fails Message-ID: <9874977.365541220011012717.JavaMail.coremail@bj126app41.126.com> hello,everyone! Recently the MobileSim in my computer with Windows XP as OS always fails to open. The details of this error reports ModName:msvcrt.dll. I have tried to unstall the old one and reinstall a new MobileSim, the same error still happens. Other softwares work still well. Could you tell me how to fix it? Thanks! Jianxin Wu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.mobilerobots.com/pipermail/aria-users/attachments/20080829/25b582bf/attachment.html From reed at mobilerobots.com Fri Aug 29 08:48:00 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 29 Aug 2008 08:48:00 -0400 Subject: [Aria-users] Why MobileSim always fails In-Reply-To: <9874977.365541220011012717.JavaMail.coremail@bj126app41.126.com> References: <9874977.365541220011012717.JavaMail.coremail@bj126app41.126.com> Message-ID: <48B7F000.3010008@mobilerobots.com> Hello, wjxnet_126 wrote: > The details of this error reports ModName:msvcrt.dll. Can you send the exact, full error message displayed? Reed From reed at mobilerobots.com Fri Aug 29 08:51:05 2008 From: reed at mobilerobots.com (Reed Hedges) Date: Fri, 29 Aug 2008 08:51:05 -0400 Subject: [Aria-users] Robot flash In-Reply-To: <4FE9D8DDD61147C1B612F8D128038C9A@mjdell> References: <27AED437-C0A0-47C2-8A6E-2893DEA74A39@ihmc.us> <4FE9D8DDD61147C1B612F8D128038C9A@mjdell> Message-ID: <48B7F0B9.9080506@mobilerobots.com> Hello, The flash upload (save) process can be corrupted by any data transmission errors. Try to use a more reliable serial port connection; for example USB-serial devices have varying levels of reliability, similar for a few laptop serial ports. Let us know if trying a different computer or serial connection makes any difference or not. Reed Matthew Johnson wrote: > Wayne, > > > > I am familiar with what you are talking about, but have never attempted > to do it myself. The best place to get this question answered is the > aria users group (aria-users at mobilerobots.com > ) on the MobileRobots website. There > is probably a post about this already. If not post one and they will get > back to fairly quickly. This is the kind of issue they are good at > answering. > > > > I am not sure the calibration will improve your accuracy that much (a > good question for the user group). Odometry is well known for its > unbounded error growth. Calibrating may slow the process, but I predict > not by much. The big problem is wheel slippage and inaccurate turns. > Once you heading is off by a bit, obviously your XY will develop a big > error proportional to your distance. The more you turn, the greater the > heading error. You can reduce this error considerably by adding a > stabilizing gyro and/or compass. MobileRobots sells them or you could > get them from elsewhere. > > > > In general localization is ALWAYS a problem. There are a number of > possible solutions (each with varying accuracy). Your domain of > operation, desired accuracy, and budget will determine what would be > best. If you provide me a description of what you are trying to do, I > can suggest some options. > > > > Matt > > > > ------------------------------------------------------------------------ > > *From:* Marco Carvalho [mailto:mcarvalho at ihmc.us] > *Sent:* Wednesday, August 27, 2008 7:43 PM > *To:* wakraus at rockwellcollins.com > *Cc:* Marco Arguedas; Carlos A. Perez; Kelli Hoback; Matthew Johnson > *Subject:* Re: Robot flash > > > > Hi Wayne, > > > > You're not referring to the linksys radios are you? > > I'm not really familiar with that problem but I'm copying Matt Johnson > here - he is the very familiar with the robots and have some suggestions > for you. > > Matt, have you seen that problem before? Any suggestions? > > > > Thanks, > > Marco > > > > > > Marco Carvalho, Ph.D. > > Research Scientist > > Institute for Human and Machine Cognition > > www.ihmc.us > > > > > > > > On Aug 27, 2008, at 9:56 AM, wakraus at rockwellcollins.com > wrote: > > > > > Guys > > I was looking into the possibility of calibrating these robots to get > better accuracy. I tried using the ARCOScf routine. You work in a > temporary ram database and then tell it to save to flash. When I tried > this the program failed and returned a message saying that the flash > erase failed and quit without trying to save the values. However, it > appears that the erase did in fact work and the flash has been wiped > out. Have you guys ever tried to use this software? Have you seen this > behavior? > > At this point I need working software just to restore the defaults and > get this robot operational again. > > >