# Sample applications

Source: [https://docs.qualcomm.com/doc/80-88500-6/topic/Sample-applications-overview.html](https://docs.qualcomm.com/doc/80-88500-6/topic/Sample-applications-overview.html)

Note: Not supported on the Linux Embedded
      systems.

Sample applications for the Qualcomm Robotics RB5 Development Kit are available on [GitHub](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x).

The sample applications are grouped into categories and each category has a corresponding folder. See the README file in the respective sample application folders for instructions to build, deploy, and test the sample application. The following is a list of the available sample applications:

- **[AI-ML-apps](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps)**
          
    This category includes applications that enable AI workloads on the Qualcomm Robotics RB5 development kit and applications that implement selected AI use cases.

    - **[snpe_classification](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps/snpe_classification)**
              
        This application sets up the environment, compiles, and runs object classification using a DLC format (generated from Qualcomm® Neural Processing SDK) TFLite model as an input.
    - **[snpe_detection](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps/snpe_detection)**
              
        This application sets up the environment, compiles, and runs object detection using a DLC format (generated from Qualcomm Neural Processing SDK) TFLite model as an input.
    - **[tflite_classification](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps/tflite_classification)**
              
        This application sets up the environment, compiles, and runs object classification using a TFLite model as an input and implementing the GStreamer pipeline.
    - **[tflite_posenet](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps/tflite_posenet)**
              
        This application sets up the environment, compiles, and runs pose estimation using a TFLite model as an input and implementing the GStreamer pipeline.
    - **[tflite_segmentation](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/AI-ML-apps/tflite_segmentation)**
              
        This application sets up the environment, compiles, and runs image segmentation using a TFLite model as an input and implementing the GStreamer pipeline.
- **[Codec2-video](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/Codec2-Video)**
          
    This category contains GStreamer-based application for H.264 video encoding and decoding. You can compile and test the H.264 video encoding and decoding capabilities for different file formats.
- **[GStreamer-apps](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp)**
          
    This category contains a set of Python and C sample applications that show the use of GStreamer plugins on the Qualcomm Robotics RB5 development platform.

    - **[gst-alsa](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_alsa)**
              
        This application implements the playback and recording functions of GStreamer and ASLA on the RB5 development platform.
    - **[gst_audio](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_audio)**
              
        This application demonstrates audio recording and playback with PulseAudio for WAV, AAC, and MP3 file formats.
    - **[gst_camera](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_camera)**
              
        This application demonstrates camera live view on display. There are two samples: one is display from the ISP camera (main onboard camera) and the other is display from the USB camera.
    - **[gst_encode_decode](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_encode_decode)**
              
        This application demonstrates how to encode camera video into an MP4 file and play back from the file. There are two samples: one is encoding from the ISP camera (main onboard camera) and the other is playback of the MP4 file.
    - **[gst_streaming](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_streaming)**
              
        This application demonstrates two ways to stream video on the network. The first
                sample creates a TCP streaming server with GStreamer. The second sample demonstrates
                how to run a Real-Time Streaming Protocol (RTSP) server on the board.
    - **[gst_tflite](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_tflite)**
              
        This application shows how to set up the environment, compile, and run object classification using a TFLite model as an input and implementing the GStreamer pipeline.
    - **[gst_transform](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/GStreamer-apps/cpp/gst_transform)**
              
        This application shows how to transform an ISP camera video to the Weston display.
- **[ROS-apps](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/ROS-apps)**
          
    This category contains applications that show the use of ROS middleware for building robotics applications.

    - **[ROS-Caffe](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/ROS-apps/ROS-Caffe)**
              
        This application integrates the Caffe framework into ROS on Qualcomm Robotics RB5 development platform.
    - **[ROS-TFLite](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/ROS-apps/ROS-TFLite)**
              
        This application integrates the Tensorflow framework into ROS with object detection API on the Qualcomm Robotics RB5 development platform.
- **[OpenCL-Application](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/OpenCL-Application)**
          
    This application demonstrates OpenCL sample code using the following examples:

    - FFT shows the use of OpenCL for fast fourier transform.
    - Benchmark shows the reading and writing rate when the memory unit is Byte, KB, or
              MB, respectively.
    - Matrix\_multiply shows the multiplication of two 20 × 20 matrices and prints the
              output of the input matrices and multiplication of the two matrices on the
              screen.
- **[platform-bringup](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/platform-bringup)**
          
    This category contains applications to fetch device information, GPIO samples, Weston-Client, and Wi-Fi hotspots.

    - **[Device-info](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/platform-bringup/Device-info)**
              
        This application provides the following information: list of peripherals to which the device is connected, status of subsystem through kernel node, CPU and GPU information, display information, and memory size.
    - **[GPIO-samples](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/platform-bringup/GPIO-samples)**
              
        This application handles the following functions: LED light control, GPIO input and output, receive GPIO interrupt events, button event catching, simple PWM output, and read the SoC thermal temperature.
    - **[WiFi-OnBoarding](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/platform-bringup/WIFI-OnBoarding)**
              
        Wi-Fi onboarding includes scanning the surrounding hotspots in the STA mode, creating a new Wi-Fi connection, and obtaining the SSID and PSK of the currently connected Wi-Fi. After switching to the SAP mode, you can create a new hotspot and get the SSID and PSK of the current hotspot.
    - **[Weston-Client-Application](https://github.com/quic/sample-apps-for-robotics-platforms/tree/master/RB5/linux_kernel_5_x/platform-bringup/Weston-Client-Application)**
              
        The Weston sample code shows how to compile and run a simple client in Weston.

Last Published: Sep 11, 2025

[Previous Topic
Key documents](https://docs.qualcomm.com/bundle/publicresource/80-88500-6/topics/RB5-documents.md)