jackzeng-robotics/decentralized_aerial_manipulation_marl

These models are associated with the paper "Decentralized Aerial Manipulation of a Cable-Suspended Load using Multi-Agent Reinforcement Learning". The models are agents as .pt files and can be directly loaded using the SKRL library. You can also extract the weights and settings (e.g. running mean scaler parameters to normalize observations) and implement them into your own pipeline.