aiqbench analysis
The paper applies a parameter-shared MAPPO policy, with estimators and feedforward terms, to decentralized control of cooperative aerial manipulation. Its reported gains are measured against a matched arm baseline, a cable-suspended variant, and two model-based controllers, with errors of 7.68 cm versus 10.95 cm. For ML readers, the useful points are the centralized-training, decentralized-execution setup and whether the learned policy holds up under unseen degradation. Those results are simulation-and-held-out-trajectory evidence, so real-world transfer remains an open question.
CALM: Decentralized Control of Cooperative Aerial Manipulation with Multi-Agent Reinforcement Learning
arXiv:2610.18425v1 Announce Type: new Abstract: Cooperative aerial manipulation is difficult under wind, payload motion, and noisy goal broadcasts because multiple Unmanned Aerial Vehicles (UAVs) are coupled through a shared payload. We present CALM (Cooperative Aerial Learned Manipulation), a decentralized controller for aerial manipulators carrying a basket while tracking a moving ground robot. CALM combines a parameter-shared Multi-Agent Reinforcement Learning (MARL) policy, trained using Multi-Agent Proximal Policy Optimization (MAPPO), with goal estimation, per-UAV Extended State Observers (ESOs), basket-mass estimation, and anticipatory formation-yaw feedforward. The anticipatory formation-yaw feedforward uses estimated goal motion to rotate the formation before position error develops, reducing basket-yaw lag without adding net commanded force. A disturbance-conditioned critic supports centralized training while execution remains decentralized. Across held-out trajectories, CALM achieves lower mean position and orientation errors than all evaluated baselines, including the matched arm baseline (7.68 versus 10.95 cm and 3.3 versus 4.7$^\circ$), the cable-suspended realization, and two model-based controllers. Under unseen full-information-path degradation, CALM remains statistically comparable or superior to the arm baseline. Videos are available at https://petreric.github.io/CALM/.
Abstract:Cooperative aerial manipulation is difficult under wind, payload motion, and noisy goal broadcasts because multiple Unmanned Aerial Vehicles (UAVs) are coupled through a shared payload. We present CALM (Cooperative Aerial Learned Manipulation), a decentralized controller for aerial manipulators carrying a basket while tracking a moving ground robot. CALM combines a parameter-shared Multi-Agent Reinforcement Learning (MARL) policy, trained using Multi-Agent Proximal Policy Optimization (MAPPO), with goal estimation, per-UAV Extended State Observers (ESOs), basket-mass estimation, and anticipatory formation-yaw feedforward. The anticipatory formation-yaw feedforward uses estimated goal motion to rotate the formation before position error develops, reducing basket-yaw lag without adding net commanded force. A disturbance-conditioned critic supports centralized training while execution remains decentralized. Across held-out trajectories, CALM achieves lower mean position and orientation errors than all evaluated baselines, including the matched arm baseline (7.68 versus 10.95 cm and 3.3 versus 4.7$^\circ$), the cable-suspended realization, and two model-based controllers. Under unseen full-information-path degradation, CALM remains statistically comparable or superior to the arm baseline. Videos are available at this https URL.
Robotics (cs.RO); Multiagent Systems (cs.MA)
arXiv:2610.18425 [cs.RO]
(or arXiv:2610.18425v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.18425 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 263 words
Related coverage
SynAgent: Generalizable Cooperative Humanoid Manipulation via Solo-to-Cooperative Agent Synergy
arXiv · Aug 4, 2026
Learning Multi-Humanoid Pickup and Transport via Decentralized Object-Centric Control
arXiv · Sep 17, 2026
SMaRT-Tug: Structured Multi-Agent Reinforcement Learning for Physics-Based Tugboat-Barge Collaborative Manipulation
arXiv · Sep 9, 2026
Decentralized End-to-End Multi-AAV Pursuit Using Predictive Spatio-Temporal Observation via Deep Reinforcement Learning
arXiv · Jun 9, 2026