AI Model Configuration

The AI Model Configuration page is designed to equip your simulation with a suite of advanced AI capabilities. Users can tailor their simulation by selecting from a variety of AI models, each serving different aspects of robotic perception and navigation.

Main Interface Overview

Model Categories and Selection

Below are the available categories of AI models. Within each category, select the models that will be incorporated into your simulation:

Perception

The perception models enhance the robot's ability to detect and understand its surroundings. These include:

  • Object Detection: Models that identify and locate objects within the simulation environment.

  • Image Segmentation: Models that partition an image into multiple segments to simplify and change the representation of an image into something more meaningful.

  • Monocular Depth: Models that estimate the depth from a single camera feed, providing 3D perception from 2D images.

  • Visual Question Answering: Models capable of interpreting the visual scene to answer qualitative and quantitative questions.

  • Visual Odometry: Models that estimate the robot's path and orientation based on image data.

Safety

Safety models are designed to predict and avoid hazardous scenarios within the simulation:

  • Time to Collision: Models that estimate the time before a potential collision could occur.

Configuring Your Selection

To configure your AI model selection:

  1. Check the box next to each AI model you wish to include from the listed categories.

  2. Follow the link to the specific model's documentation for detailed configuration instructions and use-case examples.

Note

Selecting the appropriate AI models is crucial for achieving the desired outcomes in your simulation. Ensure that you understand the function and application of each category to make informed choices.

Tip

You can combine multiple models from different categories to create sophisticated and layered AI behaviors within your simulation.

Warning

Be mindful of the simulation's computational resources. Utilizing multiple complex AI models simultaneously may require more powerful hardware specifications.

Once your AI models are selected and configured, proceed to finalize your simulation setup and initiate the session to observe the models in action.