Robot Configuration

Within the "Robot" tab of the "Create Your Own" section, users can select a robot type and configure its specific model and starting parameters for the simulation. This customization is pivotal as it determines the robot's operational range, adaptability, and the fidelity of the data collection within the simulation.

Main Interface Overview

Selecting a Robot Type

Users are presented with a choice between aerial and ground robots. Currently, the available options are:

  • Multirotor: This category encompasses various models of aerial drones, each suited to different simulation needs. Users can choose from the following multirotor types:

    • Default: A standard, small-sized drone.

    • Industrial Drone: A larger, industrial-grade drone modeled after the DJI Matrice 300 RTK.

    • eVTOL (Electric Vertical Takeoff and Landing): Aerial vehicle that can do both vertical takeoff and efficient forward flight, ideal for simulations envisioning urban air mobility and other innovative applications.

  • Car: Ground vehicles that can be used for various simulation scenarios, such as autonomous driving, warehouses, offroad scenarios. Users can choose from the following car/wheeled types, each of which offers a different vehicle dynamics.

    • Delivery Robot: A small wheeled vehicle modeled after a typical delivery robot form factor that can run on sidewalks.

    • Forklift: A forklift for warehouses and other industrial applications.

    • Offroad: A heavy-duty offroad vehicle.

    • Rover: A space rover that can handle complex terrain.

    • SUV: A typical car.

    • Tractor: A tractor for agricultural robotics scenarios.

  • Virtual Agent: GRID simulations can also be setup without any explicit vehicle type, which is called the Virtual Agent mode. In this mode the agent is only a camera in space with no dynamics associated with it, and can be controlled with the keyboard. This is useful for computer vision scenarios.

Initial Position and Geo-Location Configuration

After selecting a multirotor model, users can define the initial starting parameters:

  • Position:

    Each simulation environment has a pre-defined starting position for the robot. The X, Y, and Z values are offsets from the start position.

  • Geo Location:

    This setting is mainly useful for geospatial environments (Google Maps or Bing Maps) where the robot's initial position is defined by the latitude, longitude, and altitude in decimal format.

Note

The selected robot model will have predefined capabilities and restrictions. It is essential to choose a model that aligns with the objectives of your simulation.

Warning

Be mindful of the different flight characteristics and operational constraints associated with each multirotor model. For instance, the Matrice model would offer different flight dynamics and payload capacities compared to the Default drone.

Once the robot type and its initial parameters are configured, users should proceed to the 'Cameras' tab to outfit the robot with visual sensors and configure their properties.