How to generate wind with the Sphinx

[matsuura]

I want to generate wind on the simulation.
I referred to the following page.

https://developer.parrot.com/docs/sphinx/pluginctrl_howtos.html

I added the wind plugin in my world file.
Then I try to run the following command, but I don’t know which path <path/to/unix/socket> to choose.

Please tell me what to do

*$ echo ‘{“jsonrpc”: “2.0”, “method”: “SetParameter”, *
*“params”: {“machine”:“world”, “object”:“wind/wind”, “parameter”:“magnitude_mean”, “value”:“5.0”}, “id”: 1}’*

• | nc -U <path/to/unix/socket> | python -m json.tool*

[ocrave]

Hello,

<path/to/unix/socket> is http://localhost:8383 in most cases.

[matsuura]

Thank you for your reply.
Based on your opinion, I entered the following command after launching the Sphinx.

$ echo '{“jsonrpc”: “2.0”, “method”: “SetParameter”, *
“params”: {“machine”:“world”, “object”:“wind/wind”, “parameter”:“magnitude_mean”, “value”:“5.0”}, “id”: 1}’*
** | nc -U http://localhost:8383 | python -m json.tool

And I got the following error:

nc: unix connect failed:No such file or directory
No JSON object could be decoded

I thought about it myself, but I don’t know what to do.
I would like to know what to do next.

[ocrave]

When you start sphinx, you should see these lines:

[Msg] Connected to gazebo master @ http://127.0.0.1:11345
[Msg] Publicized address: 127.0.0.1
[Msg] created parameter server on http:8383

In this example, the server is listening on http://127.0.0.1:8383 and the following command should work:

echo '{"jsonrpc": "2.0", "method": "SetParam", "params": {"machine":"world", "object":"wind/wind", "parameter":"magnitude_mean", "value":"5.0"}, "id": 1}' | curl -d @- http://127.0.0.1:8383 | python -m json.tool

[matsuura]

Thank you very much. I was able to enter that command to generate a wind of 5m / s in the x direction.

I have another question,is it possible to generate wind from the y direction or two-dimensional wind (wind from the x and y directions)?

[ocrave]

If you are not happy with the expressions used to compute the wind in Tuning of drone internals at runtime - Howtos — Parrot-Sphinx 1.2.1 documentation you can change either of them with the same kind of JSON-RPC command. For example, this command simplifies the magnitude_expr expression to its bare minimum value:

echo '{"jsonrpc": "2.0", "method": "SetParam", "params": {"machine":"world", "object":"wind/wind", "parameter":"magnitude_expr", "value":"val"}, "id": 1}' | curl -d @- http://127.0.0.1:8383 | python -m json.tool

If you want to get rid of this plugin and use the default wind from gazebo, you have to remove it from your .world file. If you do that, the wind velocity can be set this way:

parrot-gz topic -p "~/wind" -m linear_velocity{x:5\ y:0\ z:0}

[matsuura]

I was able to generate a two-dimensional wind using the second command.

Is it possible to change the wind over time?
For example, 5 m / s for 1 minute, then 3 m / s.

I’m sorry to ask you a lot of questions.

[ocrave]

Because the real-time factor is probably less than 1, you have to take into account the simulation time that comes from telemetry. A bash script like this should work:

#!/bin/bash

function get_sphinx_time
{
    echo $(tlm-data-logger inet:127.0.0.1:9060 9061 | grep -m1 time_anafi.SimTimestamp_ns | sed 's/time_anafi.SimTimestamp_ns://g')
}

function set_wind_vel
{
    local x=$1
    local y=$2
    local z=$3
    echo "set wind velocity to [$x $y $z]"
    parrot-gz topic -p "~/wind" -m linear_velocity{x:$x\ y:$y\ z:$z}
}

function wait_for
{
    local wait_time=$1
    time0=$(get_sphinx_time)
    while true; do
        now=$(get_sphinx_time)
        diff=$(($now - $time0))
        if [ $diff -ge $wait_time ]; then
            break
        fi
        echo "remaining time: $(($(($wait_time - $diff)) / 1000000000)) seconds"
        sleep 1
    done
}

set_wind_vel 5 5 0
wait_for $((1000000000 * 60 * 1)) # nanoseconds
set_wind_vel 3 3 0

[matsuura]

By writing a script based on your opinion, I was able to change the wind over time.

Thank you for answering the question many times.