张拉整体跨域机器人的设计与控制

Translated title of the contribution: Design and Control of Tensegrity Based Cross-domain Robot

Zichen Tao, Songyuan Liu, Yun Gui, Siyuan Hao, Hao Fang, Qingkai Yang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

To achieve deep integration of cross-domain capabilities, an innovative air-ground cross-domain robot design is proposed, combining the deformable 6-bar tensegrity structure with the 4-axis 8-rotor UAV (unmanned aerial vehicle). This design utilizes the flexible characteristics of tensegrity structure, which not only endows the robot strong environmental adaptability, but also effectively reduces the risk of collision damage during flight, thereby significantly enhancing impact resistance and structural stability. Compared to traditional cross-domain robots that simply splice ground and flight modules together, this paper achieves deep integration between modules, allowing the robot to flexibly switch between ground movement and aerial flight. The core innovation of this design lies in the application of a deformable tensegrity shell. By controlling the tension of the tensioned ropes, the shell provides protection for the drone in flight mode, while adjusts its shape to enable rolling in ground mode. In addition, a rule-based rolling controller is designed, and PID (proportional-integral-derivative) control is applied to flight mode, to ensure the multimodal mobility of robot. Overall, this study provides a new approach for cross-domain robot design, and experimentally validates its multimodal cross-domain capabilities and stability.

Translated title of the contributionDesign and Control of Tensegrity Based Cross-domain Robot
Original languageChinese (Traditional)
Pages (from-to)338-347 and 360
JournalJiqiren/Robot
Volume47
Issue number3
DOIs
Publication statusPublished - May 2025
Externally publishedYes

Fingerprint

Dive into the research topics of 'Design and Control of Tensegrity Based Cross-domain Robot'. Together they form a unique fingerprint.

Cite this