Tendon-driven Grasper Design for Aerial Robot Perching on Tree Branches

💡Abstract

Protecting and restoring forest ecosystems has become an important conservation issue. Although various robots have been used for field data collection to protect forest ecosystems, the complex terrain and dense canopy make the data collection less efficient. In order to solve the above data-collection problem, we propose a bio-inspired mechanism: like many birds, it actively selects habitats to perch on and consequently saves as much energy as possible while gathering eco-information. The mechanism is driven by a multi-jointed tendon system that combines rigid and flexible materials, establishes multiple prospective frictional contacts for versatile perching, and incorporates a visual recognition algorithm to select target points. Experimental results show that it can perch on branches ranging from 3 cm to 11 cm in diameter. The real-world tests validated the system's ability to select and adapt to target points, and it is expected to be useful in complex forest ecosystems.

Anticipated images of aerial robotic platforms inhabiting real natural environments.

✨Tendon-driven Grasper Design

Design and operation of the perching mechanism. (a) The close and open state is achieved through tendon actuation and a servo motor. (b) Claw design. (c) Claw pad for additional friction. (d) Friction pad with fingerprint pattern.

👆Interactive CAD design of the tendon-driven grasper.

🔧Mechanism Design and Fabrication

Interaction profile between the branch and the perching mechanism depends on the branch diameter. The red circles show the contact points from an axial view. (a), (d) Small branches. Contact forces are shown in red arrows and the corresponding friction forces are shown in green. (b), (e) Medium branches. Internal joint forces are also shown. (c), (f) Large branches.

Static analysis of the perching mechanism in contact with a target. (a) Predicted payload capacity. (b) Simplified static interaction model.

✨Real-world Perching Experiments

Illustration of the perching point selection. (a) shows the result of the algorithm in the simulated environment and (b) shows the result of the algorithm in the real world.

RMSE and standard deviation of trunk and branch prediction errors at different distances in the Gazebo environment. On a box plot, the center mark indicates the median and the red ’*’ symbol indicates the mean. The bottom and top edges of the box illustrate the 25th and 75th percentiles, respectively.

Illustration of the real-world perching experiment. The aerial robotic system starts with takeoff, then tracks the trajectory, and finally executes the perching maneuver.

Experiments Video: take-off, trajectory planning, trajectory tracking, perching and resume.

📌Conclusion

In this paper, a mechanical design has been presented for an aerial robotic system that can perch on suitable branches in the air based on visual segmentation. An adaptive gripping mechanism controlled by a servo and tendons is first designed. The mechanism is proved suitable for perching on branches with a range of diameter from 30 mm to 80 mm, without having to consider shape, condition or surface texture of the branch. Then, to find a perching position, we design a visual segmentation algorithm for outputting 3D coordinate points in a suitable position. Finally, the trajectory of the aerial robot was planned and controlled to accomplish perching manoeuvres on aerial branches.

Initial experimental evaluations show that our aerial robotic platform is promising in acquiring the centre coordinates of the target to be perched and can perch well on tree trunks of different diameters in the tolerance of the proposed perching mechanism. Future work would also need to quantify the success rates in multiple starting points with multiple sets of experiments to release the bottlenecks of the proposed framework systematically.

🌻Acknowledgements

We gratefully acknowledge the Bristol Robotics Laboratory for providing access to the flight arena. This research was also supported by seedcorn funds by Civil, Aerospace and Design Engineering, Isambard AI and Bristol Digital Futures Institute at the University of Bristol. We also extend our appreciation to technician Patrick Brinson for his valuable assistance in conducting the experiments. Furthermore, we acknowledge Chuanbeibei Shi and Halim Atli for their initial explorations with this project’s concept, and Alex Dunnett for his contributions to photo editing and proofreading of this work.

🎈BibTeX

If you find this work helpful, please cite us.


@misc{li2025tendondrivengrasperdesignaerial,
      title={Tendon-driven Grasper Design for Aerial Robot Perching on Tree Branches}, 
      author={Haichuan Li and Ziang Zhao and Ziniu Wu and Parth Potdar and Long Tran and Ali Tahir Karasahin and Shane Windsor and Stephen G. Burrow and Basaran Bahadir Kocer},
      year={2025},
      eprint={2503.00214},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2503.00214}, 
}