# Learning End-to-End Control for Omnidirectional Aerial Motion on Overactuated Tilt-rotor Quadrotors

arXiv: https://arxiv.org/abs/2602.21583 · HTML: https://mlsanity.com/paper/2602.21583/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods.
- Builds on 23 of its 30 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot navigation and UAV control](https://mlsanity.com/deeprl/rl-for-robot-navigation-and-uav-control/)

## Builds on (most cited first, 23 of 23)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [Sim-to-Real: Learning Agile Locomotion For Quadruped Robots](https://mlsanity.com/paper/1804.10332/) (2018)
- [Asymmetric Actor Critic for Image-Based Robot Learning](https://mlsanity.com/paper/1710.06542/) (2017)
- [Reaching the Limit in Autonomous Racing: Optimal Control versus Reinforcement Learning](https://mlsanity.com/paper/2310.10943/) (2023)
- [Control of a Quadrotor with Reinforcement Learning](https://mlsanity.com/paper/1707.05110/) (2017)
- [Partially Observable Markov Decision Processes in Robotics: A Survey](https://mlsanity.com/paper/2209.10342/) (2022)
- [Neural-Fly Enables Rapid Learning for Agile Flight in Strong Winds](https://mlsanity.com/paper/2205.06908/) (2022)
- [RSL-RL: A Learning Library for Robotics Research](https://mlsanity.com/paper/2509.10771/) (2025)
- [A Benchmark Comparison of Learned Control Policies for Agile Quadrotor Flight](https://mlsanity.com/paper/2202.10796/) (2022)
- [Gallant: Voxel Grid-based Humanoid Locomotion and Local-navigation across 3D Constrained Terrains](https://mlsanity.com/paper/2511.14625/) (2025)
- [UMI-on-Air: Embodiment-Aware Guidance for Embodiment-Agnostic Visuomotor Policies](https://mlsanity.com/paper/2510.02614/) (2025)
- [RAPTOR: A Foundation Policy for Quadrotor Control](https://mlsanity.com/paper/2509.11481/) (2025)
- [Leveling the Playing Field: Carefully Comparing Classical and Learned Controllers for Quadrotor Trajectory Tracking](https://mlsanity.com/paper/2506.17832/) (2025)
- [Design and optimal control of a tiltrotor micro aerial vehicle for efficient omnidirectional flight](https://mlsanity.com/paper/2003.09512/) (2020)
- [Low-level Pose Control of Tilting Multirotor for Wall Perching Tasks Using Reinforcement Learning](https://mlsanity.com/paper/2108.05457/) (2021)
- [Autonomous Aerial Manipulation at Arbitrary Pose in SE(3) with Robust Control and Whole-body Planning](https://mlsanity.com/paper/2508.19608/) (2025)
- [MPC with Learned Residual Dynamics with Application on Omnidirectional MAVs](https://mlsanity.com/paper/2207.01451/) (2022)
- [Learning to Fly Omnidirectional Micro Aerial Vehicles with an End-To-End Control Network](https://mlsanity.com/paper/2312.05125/) (2023)
- [BEATLE -- Self-Reconfigurable Aerial Robot: Design, Control and Experimental Validation](https://mlsanity.com/paper/2404.09153/) (2024)
- [Servo Integrated Nonlinear Model Predictive Control for Overactuated Tiltable-Quadrotors](https://mlsanity.com/paper/2405.09871/) (2024)
- [Allocation for Omnidirectional Aerial Robots: Incorporating Power Dynamics](https://mlsanity.com/paper/2412.16107/) (2024)
- [Six-DoF Hand-Based Teleoperation for Omnidirectional Aerial Robots](https://mlsanity.com/paper/2506.15009/) (2025)

## Abstract (arXiv)

While reinforcement learning (RL) has been successfully applied to conventional quadrotors for agile and robust flight, whether actuator-level RL can be reliably deployed on tilt-rotor aerial robots remains an open question, as the hybrid actuation coupling brushless rotors with rotational joints introduces a substantially harder sim-to-real gap. In this work, we propose an end-to-end RL framework for omnidirectional motion control on overactuated tilt-rotor quadrotors, directly mapping target poses to joint and rotor commands. The learning framework combines actuator-level simulation with an asymmetric actor-critic architecture for 6D pose-reaching. For reliable sim-to-real transfer on the hybrid actuation, we integrate system identification with minimal yet physically grounded domain randomization. The trained policy is deployed zero-shot on real hardware and evaluated across waypoint hovering, external disturbances, payload variation and trajectory tracking, together with simulated traversal of allocation-singular configurations. The policy is compared with a state-of-the-art NMPC baseline: NMPC attains lower steady-state position error, whereas the RL policy offers a more uniform orientation error across evaluations, transitions between poses faster, and requires less onboard computation.