Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

Capture the main elements from the review meeting:

• neighbor link probability, taking into account distance and obstacles
• packet drop probability
• comms blackout probability / duration

For now, put this in the broker plugin. Consider migrating it to an ns-3 model later.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The original report had attachments: below_ground.png

This can be recreated by using the tutorial_7 controller. The lost person initial x,y position can also be set to 0,0 and it appears that the lost person starts about 20m below the surface.

Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

• Declare the world's size in each world file.
• Expose that size information via new API in RobotPlugin.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The addresses in ground_simple_2.world are now 192.168.3.1 and .2. After making the change in tutorial_3, there was still no communication.

I had some difficulties in my own controller so I am going back to a simple case to debug.

Tested on 2413d2f (default)

Original report (archived issue) by Nate Koenig (Bitbucket: Nathan Koenig).

Have the BOO send a message back to the reporting robot in the case of an incorrect detection.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

This is very often available to robots and can be used for example with task allocation - we could use this as part of our cost function.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

after merging pull request #90 I am getting an undefined symbol on line 277 of BooPlugin.cc. Here is the error:

/usr/bin/gzserver: symbol lookup error: /usr/local/lib/x86_64-linux-gnu/libSwarmBooPlugin.so.0: undefined symbol: _ZN5swarm4msgs9BooReportC1Ev

Original report (archived issue) by Anonymous.

Hello,
I'm using Ubuntu 16.04, ROS Kinetic.
When I try to clone the Swarm repo:
hg clone #!/osrf/swarm

I get the following response:
destination directory: swarm
abort: Permission denied: '/swarm'

Can someone help me resolve this issue?

Original report (archived issue) by Pat Cappelaere (Bitbucket: cappelaere).

I am trying to run on Mac OSX 10.11.2
I built the swarm client library and I am trying to test the installation

gzserver --verbose worlds/ground_simple_2.world
Gazebo multi-robot simulator, version 6.5.1
Copyright (C) 2012-2015 Open Source Robotics Foundation.
Released under the Apache 2 License.
http://gazebosim.org

[Msg] Waiting for master.
[Msg] Connected to gazebo master @ http://127.0.0.1:11345
[Msg] Publicized address: 192.168.1.10
[Msg] BrokerPlugin::ReadSwarmFromSDF: 3 swarm members found

It just hangs... and does not display the Neighbors messages.
Am I missing something else to run in the background?
[Note: Install says ns-3 is not required yet. But then says to follow the instructions described in the ns-3 installation tutorial. I went back and built ns-3-allinone/ns-3-dev... hummm]
Thanks,
Pat.

Original report (archived issue) by Pat Cappelaere (Bitbucket: cappelaere).

Just did a download on Ubuntu virtualbox
Was trying to run tests to verify install

#!python

# gzserver --verbose worlds/ground_simple_2.world
Gazebo multi-robot simulator, version 6.5.1
Copyright (C) 2012-2015 Open Source Robotics Foundation.
Released under the Apache 2 License.
http://gazebosim.org

[Msg] Waiting for master.
[Msg] Connected to gazebo master @ http://127.0.0.1:11345
[Msg] Publicized address: 172.17.0.2
[Err] [RenderEngine.cc:701] Can't open display:
[Wrn] [RenderEngine.cc:103] Unable to create X window. Rendering will be disabled
[Wrn] [RenderEngine.cc:302] Cannot initialize render engine since render path type is NONE. Ignore this warning ifrendering has been turned off on purpose.
[Wrn] [ModelDatabase.cc:339] Getting models from[http://gazebosim.org/models/]. This may take a few seconds.
[Err] [SystemPaths.cc:410] File or path does not exist[""]
[Err] [HeightmapShape.cc:199] Unable to find heightmap[file://media/materials/textures/swarm_terrain_257_random.png]
Segmentation fault

Where do I get that file from?
my /media directory is empty.
Thanks,
Pat

Original report (archived issue) by Nate Koenig (Bitbucket: Nathan Koenig).

If a false positive occurs in the logical camera, we could also assign a duration for the false positive.

Original report (archived issue) by Pat Cappelaere (Bitbucket: cappelaere).

sudo make install fails

#!python

-- Installing: /usr/lib/libTeamControllerPlugin.dylib
CMake Error at cmake_install.cmake:31 (file):
  file INSTALL cannot copy file
  "/Users/patricecappelaere/Development/swarm/example/build/libTeamControllerPlugin.dylib"
  to "/usr/lib/libTeamControllerPlugin.dylib".

Wouldn't it be better to copy to /usr/local/lib?

I am not sure I want to turn off System Integrity Protection on the Mac.
Thanks,
Pat.

Original report (archived issue) by Nate Koenig (Bitbucket: Nathan Koenig).

Make sure the Boo logs the result of a Found message and function call.

Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

• Add gps sensor to model, similar to:

<sensor name="gps" type="gps">
  <gps/>
  <always_on>1</always_on>
 </sensor>

• Add spherical coordinates to each world, similar to (but look up some real coords):

 <spherical_coordinates>
  <latitude_deg>0</latitude_deg>
  <longitude_deg>0</longitude_deg>
  <elevation>0</elevation>
  <heading_deg>0</heading_deg>
</spherical_coordinates>

• Expose this information via new API in RobotPlugin.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The original report had attachments: terrain.png, elevation.png, terrain.py, terrain_debug.cc, terrain_debug.hh, buildings_complete_10.png, default_gzclient_camera(1)-2015-11-19T09_52_21.565514.jpg, default_gzclient_camera(1)-2015-11-19T09_52_34.246116.jpg

I wrote a simple plugin that writes the output of MapQuery to a csv file. An associated python file plots it with a colormap with code (0 = PLAIN, 1 = FOREST, 2 = BUILDING). There are a few problems I noticed:

1. Buildings don't show up but they are part of the world when viewing it with the gazebo gui
2. The forest does not match the layout shown by the gazebo gui
3. The elevation doesn't seem to match the 2d image of terrain_camp_robots.sdf

This can be recreated with the attached cc and hh files.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The lost person controller needs a bounds check since it is a random walk. Otherwise the robots will not be able to find it.

Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

I see the following tests fail on default:

[  FAILED  ] 4 tests, listed below:
[  FAILED  ] CommsTest.HalfMsgsDrop
[  FAILED  ] CommsTest.TemporaryOutage
[  FAILED  ] CommsTest.PermanentOutage
[  FAILED  ] CommsTest.OutagesAndDrops

The problem seems to be that expected message counts, which determined randomly, but which should be deterministic given a constant seed (which is set), don't agree. I get the following values reliably:

/home/gerkey/code/swarm/swarm/test/integration/comms_plugin.cc:173: Failure
Value of: expectedNumMsgs
  Actual: 347
Expected: this->numMsgsRecv
Which is: 354
[Dbg] [ServerFixture.cc:118] ServerFixture::Unload
[  FAILED  ] CommsTest.HalfMsgsDrop (1859 ms)

/home/gerkey/code/swarm/swarm/test/integration/comms_plugin.cc:173: Failure
Value of: expectedNumMsgs
  Actual: 470
Expected: this->numMsgsRecv
Which is: 442
[Dbg] [ServerFixture.cc:118] ServerFixture::Unload
[  FAILED  ] CommsTest.TemporaryOutage (1854 ms)

/home/gerkey/code/swarm/swarm/test/integration/comms_plugin.cc:173: Failure
Value of: expectedNumMsgs
  Actual: 296
Expected: this->numMsgsRecv
Which is: 229
[Dbg] [ServerFixture.cc:118] ServerFixture::Unload
[  FAILED  ] CommsTest.PermanentOutage (1859 ms)

/home/gerkey/code/swarm/swarm/test/integration/comms_plugin.cc:173: Failure
Value of: expectedNumMsgs
  Actual: 426
Expected: this->numMsgsRecv
Which is: 429
[Dbg] [ServerFixture.cc:118] ServerFixture::Unload
[  FAILED  ] CommsTest.OutagesAndDrops (1855 ms)

Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

I think that we're going to want both the source and dest addresses passed to users in the OnDataReceived() callback, to assist with relaying logic.

Original report (archived issue) by Nate Koenig (Bitbucket: Nathan Koenig).

Make sure the random number seed works correctly.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

If I change the following lines in worlds/master.erb:

$ground_count = 2

$fixed_count = 0

$rotor_count = 0

to

$ground_count = 2

$fixed_count = 0

$rotor_count = 0

then rebuild/install the lost person starts at seemingly random altitudes. In the latest run I did it started at z = 150,547.083247.

This can be recreated on 25cc4e5 in tutorial_7.

Original report (archived issue) by Brian Gerkey (Bitbucket: Brian Gerkey, GitHub: gerkey).

We need to offer orientation and velocity (linear and angular) data. The easiest way is probably to add an IMU sensor and forward that info through the plugin API, like we did with GPS.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The original report had attachments: TeamControllerPlugin.cc.patch, complete_10.world.patch, RobotPlugin.cc.patch, RobotPlugin.hh.patch, tutorial5_CMakeLists.txt.patch, grid_lines.cc, grid_lines.hh, above_terrain.png, below_terrain.png, camp_roberts.png

When querying the center of the scenario of the 10x10km world I am getting an elevation of 994. From the gui where the boo is located the correct number appears to be 785.

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

The original report had attachments: fixed_wing_err.cc, fixed_wing_err.hh

Original report (archived issue) by Eric Squires (Bitbucket: esquires).

This will allow the flying aircraft to maintain a constant height over the ground. Right now we only have distance above sea level and with the added terrain the vehicles fly into the ground sometimes.

Original report (archived issue) by Carlos Agüero (Bitbucket: caguero, GitHub: caguero).

Originally we envisioned the motion API using velocity control. Right now, we're directly applying the target velocities to the models. This allows potential unrealistic motion (e.g.: going full speed forward and in the next cycle going full speed backwards).

Two ideas:

1. Change from a velocity control to an acceleration control. You're only allowed to increase/decrease a short amount of speed each cycle.

2. Keep using velocity control but with a PID in the middle.