Skip to content


Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time
occupancy_grids @ ecfae53

Remote Monitoring

A web interface for remotely monitoring robots and their operation. The interface is developed in Python 3 using Flask (there are no plans for backwards compatibility with Python 2).

This interface is developed as part of the ROPOD project and thus has multiple direct and indirect dependencies on other components developed in the project. In particular:


The web application provides six main functionalities:

  1. Monitoring the status of robot components: Information on various robot components (both hardware and software) is provided so that robots can be monitored in real time Robot status
  2. Retrieving the configuration of the component monitors: The status of the robot components is obtained by component monitors, whose configuration can be retrieved for the purpose of transparency Robot status
  3. Retrieving and displaying historical black box data associated with a particular robot: Data stored on the black box associated with a robot can be queried and displayed Black box historical data interface
  4. Displaying the latest black box data associated with a particular robot: Important variables (such as measurements from the sensor wheels and velocity commands) are shown in real time Real-time monitoring
  5. Performing remote experiments for simplified remote debugging: Requests for various predefined experiments can be sent to robots; during the execution of an experiment, continuous progress feedback and live data from the robot's black box are displayed Remote experiment interface
  6. Sending task requests: Simple transportation task requests can be sent to the fleet management library Task scheduling


  • A robot and its associated black box have the same numeric ID, such that if a robot has an ID robot_<xyz>, the black box will have an ID black_box_<xyz> (the underscore separating the numeric ID from the string ID is assumed to be there as a delimiter)
  • The Zyre-based communication between the interface and the robots/black boxes uses JSON messages whose schemas are defined in

Design principles

The application is exposed as a remote_monitoring Python package and is developed in a modular manner using Flask blueprints, such that each major functionality has its own blueprint, namely:

  • robot_status: Handlers and helper methods for providing the status of robots
  • black_box: Handlers for retrieving black box data
  • experiments: Handlers for sending experiment requests and monitoring the status of ongoing experiments
  • task_scheduling: Handlers for sending task requests to the fleet management system
  • central_operator_console:
  • component_management:
  • component_monitoring_config:
  • real_time_monitoring:

Each of these blueprints is a separate package in the remote_monitoring.blueprints subpackage.

The most important configuration parameters for the application (robot information, list of experiments) are stored in a remote_monitoring_config MongoDB database.

Functionalities common to the application are defined directly in the remote_monitoring package:

  • Defines a Config class for interacting with the configuration database as well as various variables reused throughout the application (in particular, message definitions and a socketio object)
  • Defines a BBUtils class with various static methods for simplified black box interaction
  • Defines a ZyreWebCommunicator class that takes care of all Zyre communication in the application
  • Populates the remote_monitoring_config database with all necessary parameters


Native installation

If all application dependencies are installed natively, running the application requires:

  1. installing the remote_monitoring Python package: sudo python3 install (or sudo python3 develop if installation is not desired)
  2. starting the script, namely running the command python3; the application is then accessible at port 5000


  1. When cloning for the first time, clone recursively using: git clone -b develop --recursive <url>
  2. The configuration database should be populated before starting the application for the first time (the script can be used for this purpose).


To run the application without installing any dependencies, we additionally provide a Docker image.

The Docker image can be created by running docker-compose build in the root of this repository. docker-compose up will then start two Docker containers - the web application and a MongoDB container that runs on port 27019 and is thus separate from the default MongoDB instance running on port 27017 (this is particularly useful on shared machines since it does not pollute the default instance with potentially unwanted databases). To populate the configuration database for the application when using the Docker image and interact with the correct database, the value of the db_port field in the Config class in should be updated accordingly.

If the default MongoDB instance should be used instead, simply commenting out the mongo section in the docker-compose.yml file before starting the containers will prevent the MongoDB container from running.


A Flask-based web application for remote robot operation monitoring







No releases published


No packages published