Highlights

Installation with pip: The software can be easily installed using Python's pip package manager, ensuring a smooth and straightforward setup process. For more details, please visit our official documentation.
Documentation: Detailed modules and tutorials are now provided in the documentation.
Cross-platform compatibility: The framework works seamlessly across multiple operating systems, including Windows, macOS, and Linux, enabling a broad range of users to adopt the platform.
Unit tests: Robust unit testing is implemented to ensure the reliability and stability of the codebase.
Static and dynamic environments: Supports both static (unchanging during simulation) and dynamic (interactive with robots) simulation environments, catering to a wide variety of use cases and experiments.
Sensors: Includes support for various virtual sensors such as cameras, LiDAR, and point clouds, enabling realistic simulation scenarios for robotics and autonomous systems.
Reinforcement learning example: Provides out-of-the-box examples for reinforcement learning (RL), helping users quickly train agents in diverse simulated environments.
Imitation learning example: Includes example workflows for imitation learning, allowing agents to learn from human demonstrations or other expert behaviors.

Known issues

Rendering without monitor: Rendering on a headless machine (e.g., servers or clusters without a connected monitor) might encounter issues. A workaround using virtual display tools like Xvfb or EGL is often required.
Segfault: A segmentation fault can occur when attempting to spawn a large number of pedestrians (e.g., more than 200) in a single environment. This issue may be related to memory constraints or threading limitations. Users are advised to reduce the number of pedestrians or split the scenario into smaller environments to avoid crashes.
Inconsistent simulation speeds: When running simulations across different hardware or systems, users may observe inconsistent simulation speeds due to varying GPU or CPU capabilities. To ensure reproducibility, limit frame rates or enforce deterministic settings.