multi_agent_celar is designed as a modular AI platform enabling emergent-language communication among multiple intelligent agents in simulated environments. Users can define agent behaviors via policy files, configure environment parameters, and launch coordinated training sessions where agents evolve their own communication protocols to solve cooperative tasks. The framework includes evaluation scripts, visualization tools, and support for scalable experiments, making it ideal for research on multi-agent collaboration, emergent language, and decision-making processes.

The Multi-Agent-based Search and Rescue System in ROS is a robotics framework that leverages ROS for deploying multiple autonomous agents to perform coordinated search and rescue operations. Each agent uses onboard sensors and ROS topics for real-time mapping, obstacle avoidance, and target detection. A central coordinator assigns tasks dynamically based on agent status and environment feedback. The system can be run in Gazebo or on actual robots, enabling researchers and developers to test and refine multi-robot cooperation, communication protocols, and adaptive mission planning under realistic conditions.

This open-source project delivers a complete multi-agent reinforcement learning framework built on PyTorch and Unity ML-Agents. It offers decentralized DDPG algorithms, environment wrappers, and training scripts. Users can configure agent policies, critic networks, replay buffers, and parallel training workers. Logging hooks allow TensorBoard monitoring, while modular code supports custom reward functions and environment parameters. The repository includes sample Unity scenes demonstrating collaborative navigation tasks, making it ideal for extending and benchmarking multi-agent scenarios in simulation.

SeeAct is designed to empower vision-language agents with a two-stage pipeline: a planning module powered by large language models generates subgoals based on observed scenes, and an execution module translates subgoals into environment-specific actions. A perception backbone extracts object and scene features from images or simulations. The modular architecture allows easy replacement of planners or perception networks and supports evaluation on AI2-THOR, Habitat, and custom environments. SeeAct accelerates research on interactive embodied AI by providing end-to-end task decomposition, grounding, and execution.