Gstreamer push rtsp stream to remote ip address

Is there a way to push rtsp stream to remote IP address with gstreamer on Jetson Nano ?

The similar example using ffmpeg might be (some parameters are omitted):
ffmpeg -re -i inputfile.mp4 -pix_fmt yuv420p -vsync 1 -vcodec libx264 -acodec aac -rtsp_transport udp -f rtsp rtsp://username:password@
We want the gstreamer way, eg. test-launch.

The usage is that the remote machine cannot get IP address of Nano (either remote machine is on public internet and Nano is behind NAT, or for other reasons) thus cannot do in ‘pull’ mode. We need a way to push rtsp stream from Nano to remote machine(known IP address and accessible from Nano).

Currently we are using ‘pull’ mode, on Jetson Nano we capture video, encode it, write to udpsink, then use test-launch play as rtsp server, the remote machine can use gst-launch or vlc or ffmpeg to play from Nano.
Here is illustrative example:

capture video send to udpsink

gst-launch-0.10 v4l2src ! video/x-raw-yuv,width=352,height=288 ! x264enc ! rtph264pay ! udpsink host= port=5500

rtsp server relay to rtsp stream

test-launch “( udpsrc port=5500 caps=“application/x-rtp,media=(string)video,encoding-name=(string)H264,clock-rate=(int)90000” ! queue ! rtph264depay ! rtph264pay name=pay0 pt=96 )” -p 8554

play on remote machine

gst-launch-1.0 playbin uri=rtsp://<Nano_IP_Address>:8554/test

Seems below topic does not solve my problem:

We have tried some cases. Please take a look at Jetson Nano FAQ
Q: Is there any example of running RTSP streaming?

Thanks for the response. What is under “Q: Is there any example of running RTSP streaming?” is exactly the way we are using right now.
I am looking for a different way that on Nano we ‘push’ rtsp stream to an IP address of client, this is different from playing from Nano URL on client.

We may not have enough experience to provide suggestion for this case. Could you go to gstreamer forum? If there is existing implementation in gstreamer, the users may have more experience to share suggestion. Once you have a working pipeline, you can replace with NVIDIA plugins to have hardware acceleration.