# AM-Bench: A Modular Simulation Suite and Benchmark for Aerial Manipulation Policy Learning

arXiv: https://arxiv.org/abs/2609.00641 · HTML: https://mlsanity.com/paper/2609.00641/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Sep 2026.

## Built on by (most cited first, 2 of 2)

- [AeroManip-VLA: Scalable Vision-Language-Action Learning for Aerial Manipulation with RL-Generated Demonstrations](https://mlsanity.com/paper/2609.36915/) (2026)
- [Fly-by-Code: Embodied Coding Agents for Aerial Manipulation with Active Visual and Physical Feedback](https://mlsanity.com/paper/2610.12549/) (2026)

## Abstract (arXiv)

Standardized benchmarks have played a central role in advancing robot manipulation learning, yet most focus on ground-supported manipulation systems, which limits their applicability to dynamics-critical domains such as aerial manipulation (AM). AM presents distinct system-level challenges, including environmental disturbances, coupled dynamics between the manipulator and floating base, and constrained degrees of freedom. Consequently, task performance depends jointly on robot embodiment, low-level control, and high-level policy design. We introduce AM-Bench, a modular simulation suite and benchmark for multirotor-based AM policy learning. AM-Bench includes representative embodiments spanning underactuated, fully actuated, and overactuated systems, 12 tasks across contact, transport, and constrained interaction, configurable aerodynamic disturbances and actuator saturation, standard low-level controllers, and baseline policy-learning algorithms. Unlike prior manipulation benchmarks that primarily emphasize end-to-end policy performance, AM-Bench enables system-level evaluation of how embodiment, control, disturbances, and policy choices interact. We demonstrate its diagnostic value through three simulation studies spanning high-level policies, policy--control interfaces, and embodiments, together with real-world validation of modeled effects and a hardware test of the learning pipeline.