VMAS is a comprehensive toolkit for building and training multi-agent systems using deep reinforcement learning. It supports GPU-based parallel simulation of hundreds of environment instances, enabling high-throughput data collection and scalable training. VMAS includes implementations of popular MARL algorithms like PPO, MADDPG, QMIX, and COMA, along with modular policy and environment interfaces for rapid prototyping. The framework facilitates centralized training with decentralized execution (CTDE), offers customizable reward shaping, observation spaces, and callback hooks for logging and visualization. With its modular design, VMAS seamlessly integrates with PyTorch models and external environments, making it ideal for research in cooperative, competitive, and mixed-motive tasks across robotics, traffic control, resource allocation, and game AI scenarios.

Aurora Innovation specializes in creating cutting-edge AI technology for self-driving vehicles. Their systems leverage deep learning and robotics to enhance perception, planning, and control, enabling cars to navigate safely and efficiently in various conditions. Aurora’s software integrates with existing vehicle platforms, offering manufacturers a reliable route to autonomy while focusing on real-world testing and safety.

Embedded BDI provides a full BDI lifecycle engine: it models an agent’s beliefs about its environment, manages evolving desires or goals, selects intentions from a library of plans, and executes behaviors in real time. The framework includes modules for belief base storage, plan library definition, event triggering, and concurrency control tailored for memory-limited microcontrollers. With a simple API, developers annotate beliefs, specify desires, and implement plans in code. Its scheduler handles priority-based intention execution and integrates with hardware interfaces for sensors, actuators, and network communication, making it ideal for autonomous IoT devices, mobile robots, and industrial controllers.