For long-endurance missions such as environmental monitoring and disaster response, UAVs remain fundamentally constrained by limited onboard energy because continuous lift generation is required during hovering. Perching on natural or artificial structures, such as branches or street lamps, has therefore been regarded as an effective means of extending mission duration by replacing sustained hovering with structurally supported resting. Although existing bioinspired perching systems have demonstrated attachment using robotic grippers, they typically require precise coordination between flight dynamics and grasping actuation, and many still depend on continuous or intermittent power input to maintain post-perching stability. In recent years, bistable grippers have emerged as a promising route toward reduced control complexity and energy consumption through passive engagement; however, in conventional bistable designs, the energy barrier is generally fixed once fabricated, making it difficult to reconcile compliant triggering with robust grasp retention. "While prior studies have explored tunable barriers using shape memory alloys or active actuators, such approaches still rely on external control and additional energy supply." said the author Lulu Han, a researcher at Sun Yat-Sen University, "Therefore, the development of a perching gripper capable of passively achieving adaptive energy-barrier modulation is of considerable significance for improving the environmental adaptability, grasping stability, and energy efficiency of UAV perching systems, and holds promising application prospects."
In this study, the authors developed a magnetic tensegrity-enabled bistable robotic gripper (MTRG), composed of two finger-like rigid frames hinged to a base, nonelastic cables, sliding supports, and neodymium magnets. The gripper maintains its initial stable state through the balance between magnetic attraction and cable tension, and undergoes rapid transition to a closed grasping state upon external triggering. To enable repeatable operation, an inflatable airbag and an elastic restoring element were further integrated into the base to actively reset the gripper after grasping. Methodologically, the authors first established a theoretical model based on the geometric configuration of the structure, and analyzed the effects of key geometric parameters, magnet spacing, magnetic force, and state-transition energy in order to determine a design region that balances triggering sensitivity and grasping stability. The system was then characterized through high-speed imaging and quasi-static mechanical testing to quantify closing dynamics, triggering force, and failure force. Finally, the gripper was integrated into a multirotor UAV platform together with an onboard pump, PWM control, sensing, and communication modules, followed by hovering-versus-perching energy tests, outdoor perching demonstrations, and UWB-based positioning experiments to evaluate its feasibility and performance in UAV perching applications.
The results demonstrate that the proposed MTRG establishes a strongly asymmetric energy barrier through nonlinear magnetic interaction: the transition from the initial state to the grasping state requires only about 0.58 J, whereas the reverse transition requires approximately 48.88 J, yielding an almost 85-fold difference. This enables rapid closure within about 42 ms while simultaneously combining low triggering force with high holding capability. Experimentally, the maximum triggering force was only about 0.15 N and remained stable over 1,000 cycles, whereas the failure force reached 25.38 N—approximately 200 times the triggering force. The holding performance was further enhanced by about 30% on rough surfaces, and the gripper was able to stably support payloads ranging from 0.45 to 2.07 kg. Although grasp stability decreased with increasing vibration frequency under dynamic disturbances, the system still showed meaningful environmental adaptability. In addition, the integrated airbag-based reset mechanism reliably recovered the bistable state at 28 kPa and, after optimization, enabled reset within about 15 s and deflation within about 20 s, substantially improving repeatable operation. When integrated into a UAV, the system further demonstrated reliable perching performance with negligible influence on flight behavior and attitude stability, thereby supporting its feasibility for low-energy perching and long-endurance aerial operations.
In summary, this study shows that a magnetic tensegrity-enabled bistable gripper can passively realize adaptive energy-barrier modulation, thereby combining compliant triggering and robust grasping within a single structural design, while enabling repeatable operation through an integrated airbag-based resetting module. When deployed on a UAV, the gripper demonstrates reliable perching across diverse scenarios, indicating that passive, physically intelligent grasping mechanisms may provide an effective route toward low-power, adaptive, and stable aerial perching. "More broadly, the work offers a new solution to the long-standing trade-off between sensitive triggering and secure retention in bistable grippers, and highlights the potential of magnetically actuated tensegrity mechanisms for long-duration deployment, high-altitude operation, and aerial robotic tasks in complex environments." said Lulu Han.
Authors of the paper include Lulu Han, Hao Yang, Luobin Wang, Yuquan Zheng, Jingrui Yang, Yuxuan Fu, Jieliang Zhao, Zhong Wan, Zhigang Wu, Jie Zhang, and Jianing Wu.
This work was supported by the National Natural Science Foundation of China (grant nos. 62388101, T2422031, and 52275298), the China Postdoctoral Science Foundation (2025M781273), the Natural Science Foundation of Liaoning Province Program (2025080026-JH3/101), the Postdoctoral Fellowship Program of CPSF (GZC20240192), and the Fundamental Research Funds for the Central Universities (DUT25Z3209).
The paper, "Magnetic Tensegrity-Enabled Robotic Gripper with Adaptive Energy Barrier for UAV Perching" was published in the journal Cyborg and Bionic Systems on Mar 9, 2026, at https://doi.org/10.34133/cbsystems.0535.
