Visual Studio Code Workspace for ProSeCo Planning

This Visual Studio Code workspace is convenient for working with the ProSeCo Planning project. However, if you want to use a different IDE it is still advisable to use this workspace as a top-level folder of the ROS project.

Setup

git clone --recurse-submodules --shallow-submodules https://github.com/ProSeCo-Planning/proseco_workspace.git

Once the project is cloned, the workspace can be opened with VS Code:

code ./proseco_workspace/proseco.code-workspace

Docker Setup (recommended)

Install Docker

Install Docker for Windows, Mac, or Linux (https://docs.docker.com/get-docker/)

Start the Docker Container

docker run -dit --restart unless-stopped --name proseco -p 8050:8050 --mount type=bind,source="$(pwd)"/proseco_workspace,target=/proseco_workspace karlkurzer/proseco:devel

Attach to the Docker

Either using docker attach proseco or Visual Studio Code (https://code.visualstudio.com/docs/remote/attach-container).

Local Setup

Requirements

• Ubuntu 20.04 (Focal)
• ROS Noetic Ninjemys
• Python 3.8
• python3-pip
• libeigen3
• libcairo2-dev
• swig4.0

Commands

Build

Navigate to the proseco_workspace directory and run the following command to build the project.

. /opt/ros/noetic/setup.sh && catkin_make_isolated --cmake-args -DCMAKE_BUILD_TYPE=RELEASE -DPROSECO_RUN_TEST=ON -DPYTHON_EXECUTABLE:FILEPATH=/usr/bin/python3

Install Python Package

Navigate to the proseco_workspace/src/ros_proseco_planning/python directory and run the following command to install the ProSeCo Python package.

pip3 install -e .

Source

Navigate to the proseco_workspace directory and run the following command to source the project.

. ./devel_isolated/setup.bash

Evaluate

Navigate to the proseco_workspace/src/ros_proseco_planning/python directory and run the following command to evaluate the ProSeCo planner on a specific config. Note a roscore needs to be running roscore

python3 proseco/evaluator/evaluator.py -c config.json -y

Visualize

Navigate to the proseco_workspace/src/ros_proseco_planning/python directory and run the following command to launch a dashboard (http://localhost:8050) to inspect the results from the evaluation of the ProSeCo planner on a specific config.

python3 proseco/dashboard/index.py

Configured Tasks (VS Code)

The most important tasks are configured at .vscode/tasks.json.

Configured Debugging (VS Code)

The debugging tasks are configured at .vscode/launch.json.