PaperlayerView exact revision on arXiv

Source-linked AI summary

Shared-Control Teleoperation of Omnidirectional Aerial Robots for Continuous Valve Rotation

Jinjie Li, Yicheng Chen, Haokun Liu, Moju Zhao

arXiv:2610.14478v1cs.RO

TL;DR

Limited forearm motion prevented the earlier hand-based aerial teleoperation system from continuously rotating a valve. This work uses threshold-triggered shared control to transfer rotational authority temporarily to the robot while retaining operator control of position. A real-world experiment achieved continuous rotation, while observations point to alignment and calibration concerns.

  • Problem

    Limited controller and forearm motion prevented the earlier aerial teleoperation system from continuously rotating a valve.

    Source: §I · p. 1

  • Method

    The system introduces shared control that temporarily shifts control authority to the robot for threshold-triggered continuous valve turning.

    Source: §I · p. 1, §III · p. 2

  • Results

    The robot continuously rotated a vertically mounted valve in a real-world experiment.

    Source: §IV · p. 2

  • Takeaways & Limitations

    The authors suggest extending shared control toward a general framework for human intent recognition and human–robot collaboration.

    Source: §V · p. 2

    Valve engagement produced a position offset indicating possible internal forces, and end-effector slip may have resulted from uneven rotation due to model error.

    Source: §IV · p. 2

Abstract

from arXiv · show

Omnidirectional aerial robots offer full 6-DoF independent control over position and orientation, making them popular for aerial manipulation. Although there have been advancements in robotic autonomy, human operation remains essential in complex aerial environments. Previously, we proposed an aerial teleoperation system that brings the rotational flexibility of human hands into the unbounded aerial workspace, which was evaluated through a vertically mounted valve-turning task. However, due to the limited range of human forearm's motion, the robot could not continuously rotate the valve, hindering its application in industry. In this work, we propose an approach to initiate continuous rotation when the forearm angle exceeds a threshold. This switch of control authority exemplifies the concept shared control from human-robot interaction, marking a further step in aerial teleoperation.

I. INTRODUCTION

Aerial manipulation still needs human operators in complex environments, while continuous valve rotation exceeded the limits of the earlier hand-based teleoperation system. This work introduces shared control to temporarily transfer control authority to the robot for continuous turning.

I. INTRODUCTION

  • Complex aerial environments make keeping a human operator in the loop necessary, despite progress toward fully autonomous manipulation.Source: §I · p. 1
  • The earlier six-degree-of-freedom system used four interaction modes, including Operation Mode for precise manipulation, and demonstrated wrist-driven rotation on an omnidirectional robot.Source: §I · p. 1
  • Limited controller and forearm motion prevented the earlier system from continuously rotating the valve; this work introduces shared control to shift authority temporarily to the robot.Source: §I · p. 1

II. OMNIDIRECTIONAL AERIAL TELEOPERATION

The teleoperation system combines a human operator, hand tracking and gesture sensing, and an omnidirectional aerial robot. Its motion model and interaction framework use robot and hand poses to determine the robot’s target pose.

II. OMNIDIRECTIONAL AERIAL TELEOPERATION

  • The system comprises a human operator, motion-tracking marker sets, a data glove, and an omnidirectional aerial robot.Source: §II · p. 1
  • The robot is an omnidirectional upgrade of a tiltable quadrotor, controlled with actuator-level NMPC upgraded to handle singularities during continuous rotation.Source: §II · p. 1
  • The first-order robot model and the operator’s tracked hand position and orientation provide inputs for calculating the robot’s target pose.Source: §II · p. 1

III. HAND-BASED INTERACTION

In Operation Mode, the system maps the operator’s hand pose to the robot, with a threshold-triggered algorithm initiating continuous vertical rotation. Safety measures preserve operator control of position and return control when robot and hand orientations approach one another or a time condition is met.

III. HAND-BASED INTERACTION

  • The algorithm initiates rotation when hand rotation exceeds a trigger threshold, then commands a continuous 360° rotation with a specified period and direction.Source: §III · p. 2
  • During rotation, the robot’s position remains operator-controlled, and rotation stops when hand and robot orientations become sufficiently close or a time condition is met after a guard period.Source: §III · p. 2
  • The stop orientation is chosen close to the hand’s orientation to avoid a startling jump when control returns to the hand-mapped pose.Source: §III · p. 2
  • During valve turning, the reported RMSEs were px: 0.0690 m, py: 0.0622 m, pz: 0.0468 m; roll: 7.8917°, pitch: 9.0527°, yaw: 4.1610°.Source: Figure 2 · p. 2

IV. EXPERIMENTS

The real-world experiment demonstrated continuous rotation of a vertically mounted valve. The observations also identified possible internal-force and engagement concerns that motivate compliance and more accurate model calibration.

IV. EXPERIMENTS

  • The robot continuously rotated the vertically mounted valve in the real-world experiment.Source: §IV · p. 2
  • A position offset during valve engagement indicated possible internal forces that could damage the end-effector, motivating future compliance.Source: §IV · p. 2
  • The end-effector slipped out near the end of attachment, possibly because model error caused uneven rotation; more accurate calibration may improve engagement.Source: §IV · p. 2

V. CONCLUSION

The work adds shared control to the prior system and achieves continuous valve rotation in a real-world experiment. The authors suggest extending this approach toward human-intent recognition and human–robot collaboration.

V. CONCLUSION

  • The system achieves continuous valve rotation in the real world using shared control.The shared-control concept is introduced into the authors’ previous system.Source: §V · p. 2
  • The authors propose extending this shared-control philosophy into a framework for human-intent recognition and human–robot collaboration.Source: §V · p. 2
Loading 2610.14478v1…