Add part of VSLAM-demo code

Signed-off-by: jike5 <[email protected]>

Signed-off-by: jike5 <[email protected]>

Signed-off-by: jike5 <[email protected]>

Signed-off-by: jike5 <[email protected]>

Signed-off-by: jike5 <[email protected]>

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/README.md

@@ -0,0 +1,116 @@
+# Implement VSLAM algorithm based on KubeEdge
+
+This project aims to develop a Visiual SLAM application on an edge-side robot and implementing compute offloading to reduce the compute and memory resource pressure on edge-side.
+
+## Prerequisites
+
+1. Kubeedge  1.10.1+
+2. Edgemesh  1.11.0+
+
+## Steps to reproduce
+
+1. Make sure your cluster is set up and [edgemesh](https://kubeedge.io/en/docs/advanced/edgemesh/) is enabled.
+
+2. Clone the community/sig-robotics respository.
+
+   ```bash
+   git clone https://github.com/kubeedge/community
+   ```
+
+3. Prepare your RGBD dataset or RGBD-sensors on the edge-side.
+
+   > We use [TUM Datasets](https://vision.in.tum.de/data/datasets/rgbd-dataset/download) and its format in our development and test process.
+
+4. With all nodes running on the cluster, we create the deployment by using the yaml file below.
+
+   > For example, we have a cloud node and two edge nodes, named edge01 and edge02
+
+   `cloud-deployment.yaml`
+
+   ```yaml
+   spec:
+     template:
+       spec:
+         nodeName: edge02
+   ```
+
+   `edge-deployment.yaml`
+
+   ```yaml
+   spec:
+     template:
+       spec:
+         nodeName: edge01
+   ```
+
+   Download [TUM Datasets](https://vision.in.tum.de/data/datasets/rgbd-dataset/download) to your edge node, and mount it within k8s.
+
+   ```yaml
+   spec:
+     template:
+       spec:
+         - name: data-mount
+           hostPath: 
+             path: [your_absolute_path]/workspace/data
+   ```
+
+5. Make sure you have replace the `PATH_TO_SEQUENCE_FOLDER` ,`ASSOCIATIONS_FILE` in [edge-deployment.yaml](edge-deployment.yaml) and `TUN_NAME` in [cloud-deployment.yaml](cloud-deployment.yaml)to your own parameters.
+
+   ```yaml
+   env:
+   - name: PATH_TO_SEQUENCE_FOLDER
+     value: /workspace/data/YOUR_DATA_FOLDER
+   - name: ASSOCIATIONS_FILE
+     value: /workspace/Edgeslam/Examples/RGB-D/YOUR_CONFIG.yaml
+   ```
+
+   ```yaml
+   env:
+   - name: TUM_NAME
+     value: TUM2
+   ```
+
+6. Create the applications deployment.
+
+   ```bash
+   kubectl apply -f edge-deployment.yaml
+   kubectl apply -f cloud-deployment.yaml
+   ```
+
+7. If your edge-side node support X11 GUI, you will see a frame windows and a map window. The trajectory will be saved under `/workspace/edgeslam/Examples`  folder.
+
+## Validation
+
+The works we has used for reference including [ORB_SLAM2](https://github.com/raulmur/ORB_SLAM2), and combined the performance on [TUM datasets](https://vision.in.tum.de/data/datasets/rgbd-dataset/download). We illustrate the overall framework of ours in the chart and figures.
+
+|    RMSE    | ORBSLAM2 | Ours |
+| :--------: | :------: | :--: |
+|  fr2_desk  |   0.05   | 0.12 |
+| fr3_office |   0.02   | 0.04 |
+
+<img src="images/results.png" alt="results" style="zoom:50%;" />
+
+**Conclusion**: Through the cloud-edge collaboration, the computational pressure on edge devices is reduced while ensuring better localization accuracy results.
+
+## Development Tools
+
+Due to the different application scenarios of robots, for cases that customer require, we provide some tools and environments to accelerate development process.
+
+You can locally build a mutually isolated cloud and side container. To simulate edgemesh network communication, we need to first define a docker-network:
+
+```bash
+docker network create orbslam-cloud
+```
+
+Start the containers:
+
+```bash
+sudo ./scripts/run_container_cloud.sh
+sudo ./scripts/run_container_edge.sh
+```
+
+## Reference
+
+* [raulmur/ORB_SLAM2](https://github.com/raulmur/ORB_SLAM2)
+* [droneslab/edgeslam](https://github.com/droneslab/edgeslam)
+* [turlucode/ros-docker-gui](https://github.com/turlucode/ros-docker-gui)

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/edgeslam-run/Dockerfile

@@ -0,0 +1,28 @@
+FROM jike5/orbslam2-dev:latest
+
+LABEL maintainer="[email protected]"
+USER root
+
+# # Clone code
+WORKDIR /workspace
+RUN git clone https://github.com/jike5/Edgeslam
+
+# Build
+RUN cd /workspace/Edgeslam && chmod +x build.sh \
+  && chmod +x build_ros.sh \
+  && rm -rf ./build \
+  && ./build.sh
+
+RUN cd /workspace/Edgeslam/Examples/ROS/Edge_SLAM \
+  && rm -rf build \
+  && mkdir build
+
+ENV ROS_PACKAGE_PATH "/opt/ros/melodic/share:/workspace/Edgeslam/Examples/ROS"
+
+WORKDIR /workspace/Edgeslam/Examples/ROS/Edge_SLAM/build
+RUN bash -c "source /opt/ros/melodic/setup.bash && export ROS_PACKAGE_PATH=/opt/ros/melodic/share:/workspace/Edgeslam/Examples/ROS && cmake .. -DROS_BUILD_TYPE=Release && make -j"
+
+RUN echo "source /workspace/Edgeslam/Examples/ROS/Edge_SLAM/build/devel/setup.bash" >> ~/.bashrc
+
+WORKDIR /workspace/Edgeslam/Examples
+RUN chmod +x cloud_run.sh edge_run.sh 

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/orbslam2-dev/Dockerfile

@@ -0,0 +1,65 @@
+FROM jike5/ros-melodic:cpu
+
+RUN apt-get update
+
+ARG DEBIAN_FRONTEND=noninteractive
+RUN apt-get install -y gnupg2 curl lsb-core vim wget python-pip libpng16-16 libjpeg-turbo8 libtiff5
+
+RUN apt-get install -y \
+        # Base tools
+        cmake \
+        build-essential \
+        git \
+        unzip \
+        pkg-config \
+        python-dev \
+        # OpenCV dependencies
+        python-numpy \
+        # Pangolin dependencies
+        libgl1-mesa-dev \
+        libglew-dev \
+        libpython2.7-dev \
+        # orbslam2 must use Eigen3.3.2, we will install another later
+        # libeigen3-dev \
+        apt-transport-https \
+        ca-certificates\
+        software-properties-common
+
+RUN apt-get update && apt-get install -y \
+	# websocketpp dependencies
+        libarmadillo-dev\
+        libasio-dev\
+        # network tools
+        iputils-ping net-tools iproute2 telnet
+
+# # build Eigen3
+RUN wget https://github.com/eigenteam/eigen-git-mirror/archive/refs/tags/3.3.2.zip \
+  && unzip 3.3.2 \
+  && cd eigen-git-mirror-3.3.2 \
+  && mkdir build \
+  && cd build \
+  && cmake .. \
+  && make install \
+  && cp -r /usr/local/include/eigen3/Eigen /usr/local/include
+
+# # Build Pangolin
+RUN git clone https://github.com/stevenlovegrove/Pangolin.git \
+  && cd Pangolin \
+  && git checkout 7987c9b \
+  && mkdir build \
+  && cd build \
+  && cmake .. \
+  && make \
+  && make install
+
+COPY docker/orbslam2-dev/ros_entrypoint.sh /ros_entrypoint.sh
+RUN chmod +x  /ros_entrypoint.sh
+ENV ROS_DISTRO melodic
+ENV LANG en_US.UTF-8
+
+ENTRYPOINT ["/ros_entrypoint.sh"]
+
+USER $USERNAME
+# terminal colors with xterm
+ENV TERM xterm
+CMD ["bash"]

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/orbslam2-dev/ros_entrypoint.sh

@@ -0,0 +1,16 @@
+# Copyright 2019 The KubeEdge Authors.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+set -e
+
+# setup ros environment
+source "/opt/ros/$ROS_DISTRO/setup.bash"
+exec "$@"

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/ros-melodic/Dockerfile

@@ -0,0 +1,84 @@
+# Refer: https://github.com/turlucode/ros-docker-gui
+
+FROM ubuntu:18.04
+
+LABEL Description="ROS-Melodic-Desktop (Ubuntu 18.04)"
+
+# Install packages without prompting the user to answer any questions
+ENV DEBIAN_FRONTEND noninteractive 
+
+# Install packages
+RUN apt-get update && apt-get install -y \
+locales \
+lsb-release \
+mesa-utils \
+git \
+subversion \
+nano \
+terminator \
+xterm \
+wget \
+curl \
+htop \
+libssl-dev \
+build-essential \
+dbus-x11 \
+software-properties-common \
+gdb valgrind && \
+apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Install cmake 3.22.1
+RUN git clone https://gitlab.kitware.com/cmake/cmake.git && \
+cd cmake && git checkout tags/v3.22.1 && ./bootstrap --parallel=8 && make -j8 && make install && \
+cd .. && rm -rf cmake
+
+# Install tmux 3.2
+RUN apt-get update && apt-get install -y automake autoconf pkg-config libevent-dev libncurses5-dev bison && \
+apt-get clean && rm -rf /var/lib/apt/lists/*
+RUN git clone https://github.com/tmux/tmux.git && \
+cd tmux && git checkout tags/3.2 && ls -la && sh autogen.sh && ./configure && make -j8 && make install
+
+# Install new paramiko (solves ssh issues)
+RUN apt-add-repository universe
+RUN apt-get update && apt-get install -y python-pip python build-essential && apt-get clean && rm -rf /var/lib/apt/lists/*
+RUN /usr/bin/yes | pip install --upgrade "pip < 21.0"
+RUN /usr/bin/yes | pip install --upgrade virtualenv
+RUN /usr/bin/yes | pip install --upgrade paramiko
+RUN /usr/bin/yes | pip install --ignore-installed --upgrade numpy protobuf
+
+# Locale
+RUN locale-gen en_US.UTF-8  
+ENV LANG en_US.UTF-8  
+ENV LANGUAGE en_US:en  
+ENV LC_ALL en_US.UTF-8
+ENV ROS_DISTRO melodic
+
+# Terminator Config
+RUN mkdir -p /root/.config/terminator/
+COPY docker/ros-melodic/terminator_config /root/.config/terminator/config 
+COPY docker/ros-melodic/terminator_background.png /root/.config/terminator/background.png
+
+# Install ROS
+RUN sh -c 'echo "deb http://packages.ros.org/ros/ubuntu $(lsb_release -sc) main" > /etc/apt/sources.list.d/ros-latest.list'
+RUN apt-key adv --keyserver 'hkp://keyserver.ubuntu.com:80' --recv-key C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654
+
+RUN apt-get update && apt-get install -y --allow-downgrades --allow-remove-essential --allow-change-held-packages \
+libpcap-dev \
+libopenblas-dev \
+gstreamer1.0-tools libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev libgstreamer-plugins-good1.0-dev \
+ros-melodic-desktop-full python-rosinstall python-rosinstall-generator python-wstool build-essential python-rosdep \
+ros-melodic-socketcan-bridge \
+ros-melodic-geodesy && \
+apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Configure ROS
+RUN rosdep init && rosdep update
+RUN echo "source /opt/ros/melodic/setup.bash" >> /root/.bashrc
+RUN echo "export ROSLAUNCH_SSH_UNKNOWN=1" >> /root/.bashrc
+
+# Entry script - This will also run terminator
+COPY docker/ros-melodic/entrypoint_setup.sh /
+ENTRYPOINT ["/entrypoint_setup.sh"]
+
+# ---
+CMD ["terminator"]

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/ros-melodic/README.md

@@ -0,0 +1,10 @@
+## ros-melodic image
+
+This image provide a ROS iamge with GUI sipport. We referenced this [repository](https://github.com/turlucode/ros-docker-gui), but still made a lot of improvements to better match our application scenarios.
+
+You can use the Dockerfile build the ros-melodic image:
+
+```
+docker build . -t jike5/ros-melodic:cpu -f docker/ros-melodic/Dockerfile 
+```
+

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/ros-melodic/entrypoint_setup.sh

@@ -0,0 +1,6 @@
+#!/bin/bash
+set -e
+
+# setup ros environment
+source "/opt/ros/$ROS_DISTRO/setup.bash"
+exec "$@"

sig-robotics/examples/Implement-VSLAM-algorithm-based-on-KubeEdge/build/ros-melodic/terminator_config

@@ -0,0 +1,23 @@
+[global_config]
+  title_receive_bg_color = "#64d8ff"
+  enabled_plugins = TerminalShot, LaunchpadCodeURLHandler, APTURLHandler, Logger, LaunchpadBugURLHandler
+  title_transmit_bg_color = "#15b0dc"
+[keybindings]
+[profiles]
+  [[default]]
+    use_system_font = False
+    background_darkness = 0.69
+    background_type = transparent
+    background_image = /root/.config/terminator/background.png
+    scroll_background = False
+    font = Monospace 12
+    scrollback_infinite = True
+[layouts]
+  [[default]]
+    [[[child1]]]
+      type = Terminal
+      parent = window0
+    [[[window0]]]
+      type = Window
+      parent = ""
+[plugins]