AirCrab: A Hybrid Aerial-Ground Manipulator with An Active Wheel

Muqing Cao, Jiayan Zhao, Xinhang Xu, Lihua Xie

Abstract

Inspired by the behavior of birds, we present AirCrab, a hybrid aerial ground manipulator (HAGM) with a single active wheel and a 3-degree of freedom (3-DoF) manipulator. AirCrab leverages a single point of contact with the ground to reduce position drift and improve manipulation accuracy. The single active wheel enables locomotion on narrow surfaces without adding significant weight to the robot. To realize accurate attitude maintenance using propellers on the ground, we design a control allocation method for AirCrab that prioritizes attitude control and dynamically adjusts the thrust input to reduce energy consumption. Experiments verify the effectiveness of the proposed control method and the gain in manipulation accuracy with ground contact. A series of operations to complete the letters 'NTU' demonstrates the capability of the robot to perform challenging hybrid aerial-ground manipulation missions.

Introduction

Aerial manipulation—using drones to interact with objects—is notoriously difficult because multi-rotor systems are inherently unstable. Position drift, wind disturbances, and the mechanical coupling between a drone's attitude and its position make high-precision tasks like fine assembly a major challenge. While researchers have tried adding wheels to create Hybrid Aerial-Ground Manipulators (HAGMs), existing designs often become too heavy, compromising flight time and agility.

AirCrab seeks a "middle ground" by mimicking the behavior of an egret. Just as a bird stands on one leg and uses its wings for balance to keep its head steady, AirCrab uses a single active wheel as a pivot and its rotors as "balancing wings." This design effectively eliminates the horizontal position drift that plagues free-flying drones without the weight penalty of a full drivetrain.

System Design and Methodology

The AirCrab platform consists of a quadrotor base equipped with a 3-DoF manipulator and a single, servo-driven active wheel located directly below the center of gravity.

AirCrab hardware components and biological inspiration from an egret.

The core of the software is a Partial Control Allocation (PCA) algorithm. Standard flight controllers prioritize maintaining altitude, which can lead to instability at low thrust levels. In contrast, AirCrab’s PCA prioritizes attitude (roll and pitch) to ensure the manipulator has a stable base. On the ground, the wheel handles translational movement while the propellers maintain a level attitude. The algorithm dynamically calculates the minimum thrust required to keep the motors in an effective operating range, reducing energy consumption while maximizing stability against ground-contact disturbances.

Experimental Validation

The researchers validated the system through physical experiments using Vicon motion capture for millimeter-accurate ground truth. In ground mode, the end-effector achieved a tracking accuracy of 1.02 cm RMSE. This is nearly as precise as a fixed-base robot (0.90 cm RMSE) and twice as accurate as performing the same task while hovering (2.00 cm RMSE).

Comparison of power consumption and pitch stability between the proposed PCA method and standard Ardupilot.

Comparative benchmarking against industry-standard Ardupilot firmware showed that the PCA controller reduced pitch error by up to 180% and consumed up to 47% less power in specific hover-throttle comparisons. The robot's capability was further demonstrated through the "NTU" task, where it successfully flew over obstacles, landed, and precisely placed wooden blocks to spell out the letters.

Conclusion and Future Work

AirCrab demonstrates that a single-point ground contact can significantly improve manipulation precision while maintaining the agility of an aerial platform. Future work will focus on enhancing terrain sensitivity to allow operation on sloped or rough surfaces and implementing active disturbance rejection to smooth out oscillations during ground locomotion. The ultimate goal is a fully autonomous system capable of complex navigation and manipulation without human intervention.

Summary updated

Abstract

Inspired by the behavior of birds, we present AirCrab, a hybrid aerial ground manipulator (HAGM) with a single active wheel and a 3-degree of freedom (3-DoF) manipulator. AirCrab leverages a single point of contact with the ground to reduce position drift and improve manipulation accuracy. The single active wheel enables locomotion on narrow surfaces without adding significant weight to the robot. To realize accurate attitude maintenance using propellers on the ground, we design a control allocation method for AirCrab that prioritizes attitude control and dynamically adjusts the thrust input to reduce energy consumption. Experiments verify the effectiveness of the proposed control method and the gain in manipulation accuracy with ground contact. A series of operations to complete the letters 'NTU' demonstrates the capability of the robot to perform challenging hybrid aerial-ground manipulation missions.
BibTeX
@inproceedings{cao2024aircrab,
  author = {Muqing Cao and Jiayan Zhao and Xinhang Xu and Lihua Xie},
  title = {{AirCrab: A Hybrid Aerial-Ground Manipulator with An Active Wheel}},
  booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems},
  year = {2024},
  doi = {10.1109/iros58592.2024.10802472},
}
Figure