About rcruzoliver

In this How-to I present a demo example in which MoveIt! ROS2 library, running on ctrlX CORE X7, plans a robot trajectory.  1. Concept Let's first understand how MoveIt2! works. The MoveIt2! package main functionalities are included in the "move_group" ROS2 node. This node receives inputs from the user and is able to plan low-level robotic actions thanks to its knowledge of the robot configuration (loaded in the ROS Param Server). Moreover, it communicates with the robot hardware, i.e. sensors and controllers, in order to get environment information or deploy actions respectively.  The user interface to the "move_group" can be done via C++, Python or directly from Rviz. However, this way of interacting with the planner is really low-level. For this reason, MoveIt has developed "MoveIt Task Constructor" (MTC). MTC is a higher-level layer of abstraction that allows the user to easily plan complex tasks, in which several several sub-tasks are involved.  With this ideas in mind, the system architecture would be described as presented in the following picture. Concept diagram 2. Demo description This demo is based on the example available in the MTC documentation. It uses a Franka Emika Panda robot to execute a pick and place task. This robotic platform is widely used in the research world and loved for its well ROS2 compatibility. In our case, we will run the robot emulation using functionalities from ros2_control and ros2_controllers, the move_group node and a "mtc_demo" node that used the MTC functionalities in ctrlX OS.  Such components running in ctrlX OS will communicate information thanks to the ROS2 DDS middleware to a Desktop Computer running Ubuntu22, in which an Rviz visualization will be hosted. System overview 3. Deployment in ctrlX OS For this demo you need to install two snaps: ros2-base-humble and ros2-moveit-mtc. Sadly we cannot share the the instructions to compile them in this How-To. Please feel free to reach us if you are interested by leaving a comment in this post. ros2-base-humble includes the basic runtime for ROS2 humble. Its resources are used by other ROS2 snaps. ros2-moveit-mtc includes three main components: the Robot HW controllers, the move_group node and a specific node that uses MTC to plan a complex task. We would need a launcher to start the Rviz simulation on your Desktop Computer.  4. Run the example This example assumes you have SSH connection rights with your X7. 4.1. Connect the devices Connect your ctrlX CORE X7 and the Ubuntu Desktop Computer in the same local network. 4.2. Run the robot HW and the move_group Open a SSH connection and run: sudo ros2-moveit-mtc.initALL 4.3. Start the visualization Open a terminal in your Desktop computer and run the bash that starts the simulation. It sources the ROS2 workspace that was created during the snap compilation, so it will only work if you have compiled the snap, which runs colcon build in the process, and hence the install directory is present. cd <your_path>/moveit-mtc/ ./visualization_demo.sh You should see the robot in this specific pose. Robot in ready poseIf it's not exactly like this, it means the initALL routine failed in the ctrlX CORE. In this case, shut it down with ctrl^C and call the command again.  4.4. Start the MTC node Open a SSH connection and run: sudo ros2-moveit-mtc.task If everything is correct, you should see the robot moving. Robot executing task Related Links How to activate SSH communication in a ctrlX CORE  ROS2 on ctrlX AUTOMATION    Have fun with your robots! ... View more

PostgreSQL is a powerful, open-source relational database management system (RDBMS) known for its extensibility, robustness, and adherence to SQL standards. It provides advanced features such as support for complex data types, indexing, transactions, and concurrency control. PostgreSQL is designed to handle large of data and is often chosen for mission-critical applications, offering a high level of data integrity and reliability. It supports various programming languages and has a thriving community that contributes to its ongoing development and enhancement. Additionally, PostgreSQL is known for its extensibility, allowing users to define custom data types, operators, and functions. For these reasons, its use in the ctrlX Ecosystem can really power up our applications. In the following picture you can see what the system structure looks like: System overviewThis project targets the current generation ctrlX CORE virtual , ctrlX CORE X5 and ctrlX CORE X7 running ctrlX OS 2.xx (base apps running an ubuntu core 22). This project is currently supported for amd64, you are free to explore the possibilities for arm64. The functionalities showed in this How-To have been tested in ctrlX CORE virtual  and ctrlX CORE X7. This guide is simply a demo example, feel free to keep playing with it! 1. Installation  In your Engineering PC you need to download PgAdmin4, you can do it from the official PgAdmin website. Choose your operating system. The functionality showed in this How-To has been tested in Windows 10 and Ubuntu 22.04 as host system in the Engineering PC. Install the PostgreSQL runtime in your ctrlX CORE. You need to compile it yourself following the instructions that you can find in the attached zip.file (ctrlx_postgresql-main.zip). 2. Configuration The PostgreSQL runtime uses the files found in the attached zip.file under ctrlx_postgresql/shcripts/data :  configuration/ : it contains all the neccesary configuration files for PostgreSQL to work. data_postgresql/ : it contains the information data. 3. Connect PgAdmin4 with ctrlX OS Connect the Engineering PC and the CORE in the same network. If you are using ctrlX CORE virtual , do not forget to fordward the port in which the runtime is accesible (5432 by default). Open PgAdmin4 and add a new server. Add new serverA window will pop up. Fill in the fields as indicated in the image. The server is running in ctrlX CORE virtual , hence localhost (127.0.0.1) must be indicated, if it is a real CORE, simply indicate the corresponding IP address. The user postgres is by default granted to access without password. This can be later edited. Setting up a server Setting up a serverIf the connection does not work, in some cases the connection to that IP is protected. You can try with SSH connection with password, you must fill in as follows with your SSH credentials. SSH connectionAnd then you can click save, and you will be logged in. Server connected 4. Test to publish something to PostgreSQL using Node-RED There exists a Node-RED palette to communicate with PostgreSQL. You can download it as seen in the following images. The CORE must be connected to the internet. Add new palettes PostgreSQL palette We have prepared some examples to publish to the data base from Node-RED. Find them attached to this How-to (flows_PostgreSQL.zip). To import it, open the "Flow Editor" from Node-RED and proceed as shown in the following pictures: Import example Select the .json file Example json loaded You must import both flows (.json) in that .zip: Products: Working in the background, you do not need to interactuate with this. It simulates a product object that would be later publish to the database. Product object generationPostgreSQL: It allows you create a table, and then publish cyclicly a product object. Likewise, we have setup a functionality to read objects from the database. Node-RED exampleOnce you have published something, you can go to PgAdmin4 and visualize what has been published to the Data Base. After creating the table from Node-RED, you need to refresh the table list as shown in the following picture.  Right click on tables and refresh the list After that you can visualize the rows in such table as follows: Right click "products" and visualize all rows To refresh the visualization you can query again by clicking on the play button on the top. You can also export the data in a csv.file by clicking in the download arrow on top of the table. Table visualization Have fun! ... View more

MongoDB is an open-source NoSQL database system designed for handling large volumes of data. It stores data in flexible, JSON-like documents using a format called BSON. MongoDB is known for its scalability, high performance and support for dynamic data models. It is widely used in applications that require efficient storage, retrieval, and manipulation of data, such as content management systems and real-time analytics. The database supports ad hoc queries, indexing, and has a powerful aggregation framework for data analysis. For these reasons, its use in the ctrlX Ecosystem can really power up your applications. In the following picture you can see what the system structure looks like: System overview This project targets the current generation ctrlX CORE virtual , ctrlX CORE X5, ctrlX CORE X7, and ctrlX OS on a IPC when running ctrlX OS 2.xx (base apps running an ubuntu core 22). This application is (and will be) only supported for amd64 architectures as explained in the attached .zip. Therefore, Mongo DB 7.0 cannot be ported to the current generation of ctrlX CORE X3 with a 4 Bit Quad-Core ARM CPU, in particular with arm64:v8.0. The functionalities shown in this How-to have been tested in ctrlX CORE virtual  and in ctrlX CORE X7. This guide is simply a demo example, feel free to keep playing with it ! 1. Installation  In your Engineering PC you need to download Mongo DB Compass, you can do it from the official Mongo DB website. Choose your operating system. The functionality shown in this How-to has been tested in Windows 10 and Ubuntu 22.04 as host system in the Engineering PC. Install the MongoDB runtime in your ctrlX CORE. You need to compile it youself following the instructions that you can find in the attached .zip. (ctrlx_mongodb-main.zip) 2. Configuration 2.1. General settings If the MongoDB runtime was successfully installed in your CORE. You will find a folder called "Mongo DB" in the persistent data, i.e. "Manage app data" section.  Files in persistent data Inside you will find three items: mongodb: the actual data server where the data is stored. In this example the data is stored locally in the CORE, however this can be easily changed as we later will see. mongodb.conf: config file for the Mongo DB runtime. You can download the app data (appdata.zip), modify this file and upload it again to the CORE. By doing this, one can change the configuration of Mongo DB. mongodb.log: log messages about the state of the Mongo DB runtime. Now, let's have a closer look at mongodb.conf. Most of the options are left commented, if you want to know what they do, please refer to the official documentation of Mongo DB. I want to highlight a couple of things that you may want to edit: Storage paths dbpath: specify where the data server will be hosted. For example, if you have an storage media mounted in your CORE, you could specify this path. See section 2.2 where more details about this are given. logpath: where the log file will be generated bind_ip = 0.0.0.0.0, it means the daemon will explore all the ip address in the local network port: specify the port from which the runtime can be externaly accessed (important for Compass) I recommend you to keep the configurations like this the first time, and you try to modify them later. It is possible to add user and password to access to the data base.  2.2. Configure an external storage media If you want to store a great amount of data, it is advisable to do so in a external storage media and not in the interal memory of the ctrlX CORE. In the following, I will present how to configure a external media storage device in ctrlX CORE X7. (Speaking about external media does not make sense in ctrlX CORE virtual ) ctrlX CORE X7 supports both USB-Flash Drives on the XF10A/B/C ports and micro-SD card on CF01 port. First, make your external media a single GPT partition, formated with FAT32. You can learn how to do it in the following How-To. Plug the device in the ctrlX CORE and navigate to the storage page (Settings>Storage). You should see the external media device together with the local "System Data". Storage in ctrlX OS The click on the triangule on the right upper corner of the external media to mount the filesystem. You will prompted with the following. Mounting the filesystem Select "For data exchange" if you do not want to encript the data, i.e. it could be eventually read from outside (e.g. another ctrlX CORE), otherwise, if you want this ctrlX CORE to be the only one able to read the data, select "To extend system data". After this step, the external media is ready to be accesed by the different snaps. Actually, for a snap to have access to the external media device, it must include the "removable-media" plug (see the snapcraft.yalm file for more details). Although this plug is not generally connected in a snap, in the case of ctrlX OS, it is connected, so no further steps are neccesary. Now you can modify the "mongodb.conf" file and state that the database should be stored in the external media. External media path example NOTE: if you want to store in a sub-folder in the media device, you must manually create it (i.e. include a mkdir command in the hook that runs at the app booting, or use a ssh connection to access to the system memory) 3. Connect Compass with ctrlX OS Connect the Engineering PC and the CORE in the same network. If you are using ctrlX CORE virtual , do not forget to fordward port in which the runtime is accesible (27017). Port fowarding for ctrlX COREvirtual Open Compass and check your connection string. Connection string in Compass In our case the ctrlX CORE virtual, the IP address is localhost (127.0.0.1), and the port is the default one. If you are using a real ctrlX CORE, the IP address is 192.168.1.1. by default. Adapt it depending on your setup. There is no user/password to access the data base by default. If the connection does not work, in some cases the connection to that IP is protected. You can try with SSH connection with password (open advanced Options) if you have that feature enabled in your device (in virtual it is enabled by default). Probably you will not need SSH again, only the first time. New updates of this How-to will address this issue. SSH connection details If you are able to connect, you are able to visualize your data. It should look like this: Compass initial window 4. Test to publish something to Mongo DB using Node-RED There exists a Node-RED palette to communicate with Mongo DB. You can download it as seen in the following images. The CORE must be connected to the internet. Add new palettes  MongoDB palette We have prepared some examples to publish to the data base from Node-RED. Find them attached to this How-to (flows_mongoDB.zip). To import it, open the "Flow Editor" from Node-RED and proceed as shown in the following pictures: Import example Select the .json file Example json loaded You must import both flows (.json) in that .zip: Products: Working in the background, you do not need to interactuate with this. It simulates a product object that would be later publish to the database. Product object generation mongoDB: It allows you to publish cyclicly a product object. Likewise, we have setup a functionality to read objects from the database. Node RED example Once you have published something, you can go to Compass, refresh it, and see they data you published.  You can also export the data in a csv or json object, following what is shown in the next picture: Export data from Compass Have fun! ... View more