Holodeck is a high-fidelity simulator for reinforcement learning and robotics research, offering a rich environment for training agents. It targets RL researchers and roboticists, providing a powerful tool for developing and testing intelligent agents in complex, simulated scenarios.

How It Works

Built on Unreal Engine 4, Holodeck leverages its advanced rendering and physics capabilities to create realistic simulation environments. It offers a Python interface similar to OpenAI Gym, abstracting away the complexities of the game engine. This approach allows for high-performance simulation speeds, up to 2x real-time, with a performance penalty only for features actively used by the agent.

Quick Start & Requirements

• Install via pip: pip install holodeck (requires Python >= 3.5).
• Docker installation is also supported.
• See Installation for complete instructions.
• Official Documentation and Quickstart are available.

Highlighted Details

• Supports multi-agent training and control.
• Offers 7+ rich worlds and numerous scenarios.
• Can run headless with GPU-accelerated rendering.
• Achieves simulation speeds up to 2x real-time.

Maintenance & Community

• Project of BYU's Perception, Cognition and Control Lab (PCCL).
• A Discord server is available for questions.

Licensing & Compatibility

• The README does not explicitly state a license.

Limitations & Caveats

• The project is described as a simulator for research, implying potential for ongoing development and potential breaking changes.