Replikating GPS-less Navigation and Obstacle Avoidance with Tracking and Depth Camera project with jetson nano

Im working on a project similar to Thien Nguyen
GPS-less navigation

The main difference is the companion computer I’m using. Nvidia Jetson Nano instead of x86 based one. Im still waiting for the intel real sense cameras but I have installed all the necessary libs on the nano. Will update the post when I test the cameras.

Intel® RealSense™ Depth Camera D435 & Tracking Camera T265 bundle.


Looking forward to seeing your results!

1 Like

On the jetson nano, I have followed the steps from

([GSoC 2020] Enhancements of GPS-less Navigation and Obstacle Avoidance with Tracking and Depth Camera) but changed some things.

I have Increased Jetson Nano swap file.


pip3 was not working so first i have installed:

sudo apt-get install libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran
sudo apt-get install python3-pip
sudo pip3 install -U pip
sudo pip3 install -U pip testresources setuptools numpy==1.16.1 future==0.17.1 mock==3.0.5 h5py==2.9.0 keras_preprocessing==1.0.5 keras_applications==1.0.8 gast==0.2.2 futures protobuf pybind11

sudo apt install python3-opencv

For T265 and D4xx script

pip3 install pyrealsense2 <-- On the jetson Nano we skip this step instead we use.

pip3 install transformations
pip3 install dronekit
pip3 install apscheduler
pip3 install pyserial # For serial connection
pip3 install opencv-python

For using both scripts at the same time

pip3 install dronekit-sitl -UI
pip3 install MAVProxy

Install ROS

Clone the package in the ROS workspace or any location if you don’t use ROS

cd ~/catkin_ws/src
git clone
cd vision_to_mavros/script <-- Here we have typo, the dir is /scripts
chmod +x
chmod +x
chmod +x
chmod +x
chmod +x