ori-drs/ROS-MSCL

Description

Interface (driver) software, including ROS node, for inertial sensors compatible with the Microstrain Communication Library (MSCL).

MSCL is developed by LORD Sensing - Microstrain in Williston, VT.

Build Instructions

MSCL

MSCL is now installed in the CMakeLists.txt. The version installed can be changed by passing the flag -DMSCL_VERSION="62.0.0"

We do our best to keep ROS-MSCL up-to-date with the latest MSCL changes, but sometimes there is a delay. The currently supported version of MSCL is v62.0.0

Building from source

Install ROS and create a workspace: Installing and Configuring Your ROS Environment
Move the entire ROS-MSCL folder (ros_mscl and mscl_msgs for just source) to the your_workspace/src directory.
Locate and register the ros_mscl package: rospack find ros_mscl
Build your workspace:
```
 cd ~/your_workspace
 catkin_make
 source ~/your_workspace/devel/setup.bash
```
The source command may need to be run in each terminal prior to launching a ROS node.

Launch the node and publish data

The following command will launch the driver. Keep in mind each instance needs to be run in a separate terminal.

    roslaunch ros_mscl microstrain.launch

Optional launch parameters:

name: namespace the node will publish messages to, default: gx5
port: serial port name to connect to the device over, default: /dev/ttyACM0
baudrate: baud rate to open the connection with, default: 115200
imu_rate: sample rate for IMU data (hz), default: 100
debug: output debug info? default: false
diagnostics: output diagnostic info? default: true

To check published topics:

rostopic list

Example: Connect to and publish data from two devices simultaneously
In two different terminals:

roslaunch ros_mscl microstrain.launch name:=sensor1234

roslaunch ros_mscl microstrain.launch name:=bestSensor port:=/dev/ttyACM1

This will launch two nodes that publish data to different namespaces:

sensor1234, connected over port: /dev/ttyACM0
bestSensor, connected over port: /dev/ttyACM1

An example subscriber node can be found here: ROS-MSCL Examples

Docker Integration

VSCode

The easiest way to use docker while still using an IDE is to use VSCode as an IDE. Follow the steps below to develop on this repo in a docker container

Install the following dependencies:
1. VSCode
2. Docker
Open VSCode and install the following plugins:
1. VSCode Docker plugin
2. VSCode Remote Containers plugin
Open this directory in a container by following this guide
1. Due to a bug in the remote container plugin, you will need to refresh the window once it comes up. To do this, type Ctrl+Shift+p and type Reload Window and hit enter. Note that this will have to be repeated every time the container is rebuilt
Once the folder is open in VSCode, you can build the project by running Ctrl+Shift+B to trigger a build, or Ctrl+p to open quick open, then type task build and hit enter
You can run the project by following this guide

Make

If you are comfortable working from the command line, or want to produce runtime images, the Makefile in the .devcontainer directory can be used to build docker images, run a shell inside the docker images and produce a runtime image. Follow the steps below to setup your environment to use the Makefile

Install the following dependencies:
1. Make
2. Docker
3. qemu-user-static (for multiarch builds)
  1. Run the following command to register the qemu binaries with docker: docker run --rm --privileged multiarch/qemu-user-static:register

The Makefile exposes the following tasks. They can all be run from the .devcontainer directory:

make build-shell - Builds the docker image and starts a shell session in the image allowing the user to develop and build the ROS project using common commands such as catkin_make
make image - Builds the runtime image that contains only the required dependencies and the ROS node. The resulting image is names ros-mscl
make clean - Cleans up after the above two tasks

License

ROS-MSCL is released under the MIT License - see the LICENSE file in the source distribution.

ori-drs / ROS-MSCL