Nano-backpack for NAO robot

Nano-backpack is an open hardware add-on for the humanoid robot NAO. It adds a Nvidia Jetson Nano development board to the robot and thereby enhances the computational capabilities of the system. The corresponding software architecture includes a webapp to control the robot and allows an easy start into the development of further applications.

This project was part of the Robotics Practical class (MScTI_ROBP) in summer term 2020 @ Heidelberg University.

Setup:

Install and setup linux-wifi-hotspot on Jetson Nano

Install dependencies:

sudo apt install -y libgtk-3-dev build-essential gcc g++ pkg-config make hostapd

Clone repo and build:

git clone https://github.com/lakinduakash/linux-wifi-hotspot
cd linux-wifi-hotspot

# build binaries
make

# install
sudo make install

Open application and create hotspot with rndis0 preset
Run linux-wifi-hotspot on every startup:
```
systemctl enable create_ap
```

Clone this repo and build Docker Image

Clone:

git clone https://github.com/roydenwa/nano-backpack.git

Navigate to nano-backpack/src/app/ and build Docker image with:
```
sudo docker build -t "nano-backpack" .
```

Install and run dnsmasq

Install dnsmasq:
```
apt install dnsmasq
```
Run provided bash script to setup a subnet:
```
sudo bash run_dnsmasq.sh
```

Run Docker image and connect to wifi-hotspot to see webapp with other devices

Start Docker image with interactive shell: sudo docker run --rm -it -p 80:80 "nano-backpack" The webapp should now be visible at 192.168.12.1 if you are connected to the previously created wifi hotspot.

License and credits

This work is licensed under a Creative Commons Attribution 4.0 International License.

This repo is built upon other open-source contributions:

CAD models are partially based on NAO-backpack.
The wifi hotspot is created with linux-wifi-hotspot.
The unfinished concept for monocular depth estimation is based on DenseDepth.

roydenwa/nano-backpack

Nano-backpack for NAO robot

Setup:

License and credits

TODO: