Dynamic Agents

The Parallel Domain Platform supports three classes of dynamic agents, objects with some level of logic that enable control or navigation within our Worlds. For more information about them, see our Ontology section. To view our dynamic agents in action, check out our Web Visualizer .

Vehicles & Cyclists

We support a large variety of vehicles ranging from construction and emergency vehicles to sedans, SUVs and 2-wheelers (bicycles and motorcycles). Our vehicles emulate the diversity in the real world with support for accessories (roof racks, decals etc..), turn signals and more. Please see our Vehicles section for more.


Parallel Domain provides a wide selection of human agents, including pedestrians, drivers, medical workers, and more. These agents feature a variety of characteristics, such as age, race, gender, and body shape, to help identify and reduce bias in computer vision models.

They appear in various scenarios, from jogging to jaywalking, and can be adjusted to move at different speeds. Additionally, keypoints annotations are provided for each pedestrian. Check out our Pedestrians section for more information.


Animals pose unique challenges for deploying autonomous vehicles. Our support for animals includes dogs, cats, rabbits, deer, elk, racoons, rats, and more. The animals can be embedded into different scenarios to create edge cases that are difficult to capture in the real world. Our animal library is always expanding! To make a request for new animals or to learn more about our plans for the future, contact our sales team.