Skip to content

Commit b43d997

Browse files
committed
docs: add jetson setup
Issue: #43
1 parent 40d9312 commit b43d997

File tree

1 file changed

+89
-0
lines changed

1 file changed

+89
-0
lines changed

README.md

Lines changed: 89 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,10 @@
1818
- [Apply Fixes](#242-apply-fixes)
1919
- [2.5. Bazel Buildifier](#25-bazel-buildifier)
2020
- [How to Run](#251-how-to-run)
21+
- [3. Hardware](#3-hardware)
22+
- [3.1. Jetson Nano](#31-jetson-nano)
23+
- [Flashing](#311-flashing)
24+
- [Remote Connection](#312-remote-connection)
2125

2226
## 1. Development Environment
2327

@@ -187,3 +191,88 @@ To run Buildifier execute the following command:
187191
```bash
188192
bazel run //bazel:buildifier
189193
```
194+
195+
## 3. Hardware
196+
197+
### 3.1. Jetson Nano
198+
199+
#### 3.1.1. Flashing
200+
201+
The approach described in the [official documentation](https://www.waveshare.com/wiki/JetRacer_AI_Kit#Support) did not work for this setup. Instead, the solution is to use the [NVIDIA SDK Manager with Docker containers](https://docs.nvidia.com/sdk-manager/docker-containers/index.html).
202+
203+
1. **Install NVIDIA SDK Manager with Docker**:
204+
205+
During installation, choose the Docker image that uses Ubuntu 18.04, as the Jetson Nano is not compatible with newer versions.
206+
207+
2. **Install Required Dependencies**:
208+
209+
Ensure that the following dependencies are installed:
210+
211+
```bash
212+
sudo apt install qemu-user-static binfmt-support
213+
sudo update-binfmts --enable
214+
```
215+
216+
3. **Load the Docker Image**:
217+
218+
Load the Docker image from the tar file:
219+
220+
```bash
221+
docker load -i ./sdkmanager-2.2.0.12021-Ubuntu_18.04_docker.tar.gz
222+
```
223+
224+
4. **Tag the Docker Image**:
225+
226+
Tag the image as latest:
227+
228+
```bash
229+
docker tag sdkmanager:2.2.0.12021-Ubuntu_18.04 sdkmanager:latest
230+
```
231+
232+
5. **Put the Jetson Nano into Recovery Mode**:
233+
234+
Before running the container, connect the pin "fcrec" and ground to put the board into recovery mode, needed to flash the os.
235+
Insert the SD card into the Jetson Nano and connect the micro-USB cable.
236+
237+
6. **Run the Docker Container**:
238+
239+
Execute the following command to run the SDK Manager container:
240+
241+
```bash
242+
sudo docker run -it --privileged \
243+
-v /dev/bus/usb:/dev/bus/usb/ \
244+
-v /dev:/dev \
245+
-v /media/$USER:/media/nvidia:slave \
246+
--network host --name jetson_flash sdkmanager --cli
247+
```
248+
249+
7. **Follow the Steps**:
250+
251+
- Select Action: Install
252+
- Select Product: Jetson
253+
- Select System Configuration: Host Machine [Ubuntu 18.04 - x86_64]
254+
- Select Target Hardware: Jetson Nano module
255+
- Choose manual configuration and **wait for the download to finish before selecting the Flash option**. Else flashing will not work.
256+
257+
For more details on Jetson Nano pins, refer to the Jetson Nano 2GB DevKit User Guide.
258+
259+
#### 3.1.2. Remote Connection
260+
261+
After flashing is complete, turn off the Jetson board and remove the connected pins.
262+
Ensure peripherals are connected to the board before powering it up.
263+
264+
1. **Check IP Address**:
265+
266+
On the Jetson Nano, check its IP address by running:
267+
268+
```bash
269+
ipconfig
270+
```
271+
272+
2. **SSH into the Jetson Nano**:
273+
274+
Connect to the device using SSH:
275+
276+
```bash
277+
ssh <user>@<wlan_inet>
278+
```

0 commit comments

Comments
 (0)