Build and install Intel's librealsense for the NVIDIA Jetson Nano Developer Kit
These scripts are for installing the Intel RealSense SDK librealsense on the NVIDIA Jetson Nano Developer Kit. The current release of librealsense (V2) supports the D415, D435, D435i and T265 cameras. All of the cameras need to have librealsense installed. CUDA support is disabled by default. The CLI flag ( -w | --build_with_cuda ) will build with CUDA support. installLibrealsense.sh
Building without CUDA support will take signficantly less time. Because the Jetson Nano uses CUDA 10, it requires CMake version 3.11+. The version in the Ubuntu repository does not meet the requirement. The installLibrealsense script will build CMake if a CUDA build is indicated, and use that version to configure the librealsense build.
During compilation, the script will run out of memory on the Nano You will need either to: * Enable swap memory OR: * Modify the script to 'make' with only 1 processor
If you are using the T265 Tracking Camera exclusively, it is worth considering compiling without CUDA support. The librealsense CUDA support is for the Depth cameras. To build and installLibrealsense with CUDA support:
$ ./installLibrealsense.sh --build_with_cuda
To build and install librealsense WITHOUT CUDA support:
Note: If you are using a RealSense T265 exclusively, this is the only installation necessary. You do not need to patch the modules and kernels as noted below.
The patchUbuntu.sh patches kernel modules and installs them to support the RealSense cameras. These patches are for the Depth cameras. The patches include adding the different image formats to the uvcvideo module and timestamping. For the D435i in particular, support for HID interface for the IMU. In addition, a Nano specific patch is added addressing a USB throughput issue. After updating and installing the modules, the script rebuilds the kernel and installs it. To patch the kernel and modules, then install them: patchUbuntu.sh
If you run one of the Depth cameras without addressing these issues, you will see the issues listed on the console about not being able to recognize the image formats, along with time stamp issues and so on. You will also notice that the system log is flooded with OOPs related to not being able to recognize the camera formats from the kernel.
Note: If you are building from a USB or some other device than the SD card, you will need to copy the Image file to the /boot directory on the SD card which you will be booting from.
April 30, 2019
Switch CLI argument to build_with_cuda ; Build with CUDA takes a lot more time because CMake needs to be rebuilt. Default is to build without CUDA support
previous commit Add CLI argument build_no_cuda ; script defaults to build with CUDA.
D435i is not recognized by RealSense applications, but shows up in Cheese webcam viewer.
April 29, 2019
Add CUDACXX flag for compilation using CUDA
Add USE_CUDA flag to script (Defaults to YES)
If using only RealSense T265 camera, this is the only installation necessary
If using the T265, you probably don't need CUDA (still needs to be tested); Set USE_CUDA to false. Saves compilation time
During compilation, the script will run out of memory on the Nano You will need either to:
- Enable swap memory
- Modify the script to 'make' with only 1 processor
patchUbuntu will patch all of the needed modules for librealsense, build the modules, and then install the modules. The kernel Image is then built and installed in /boot/Image.
Note: If you are building from a USB or some other device than the SD card, you will need to copy the Image file to the /boot directory on the SD card.