Tethered UAV system demonstrates autonomous knotting for heavy load aerial transport

by Riko Seibo
Tokyo, Japan (SPX) Jan 07, 2026

"Cable-driven systems excel at heavy-load transport but are limited by fixed anchoring points in unstructured environments," explained study corresponding author Lihua Xie from Nanyang Technological University. The research team has demonstrated a tethered unmanned aerial vehicle system that can autonomously form knots on existing structures and then lift heavy payloads, providing anchor points where none are pre-installed.

The approach uses a cable-driven transport configuration in which a UAV carries a winch and tether and performs an enclosing and binding maneuver around a target structure. A human-in-the-loop knot planner interprets user sketches, extracts an enclosing plane around the object, searches feasible paths via frontier clusters, and generates a knotting trajectory that the UAV executes. The design integrates UAV mobility for positioning and a winch for load bearing, so the aircraft can secure the tether and then reel the load upward.

The planner is guided by three optimization metrics: enclosing planarity, tether visibility, and tether clearance. Enclosing planarity helps keep the tether from slipping off the structure, tether visibility maintains visual tracking of the cable, and tether clearance reduces the chance of collision with obstacles. "The three metrics work synergistically-enclosing planarity prevents tether slippage, tether visibility maintains real-time monitoring, and tether clearance avoids collisions," said co-first author Rui Jin.

The UAV platform carries LiDAR and RGB-D cameras linked to an onboard computer that manages mapping, perception, and tether detection in real time. With this sensing stack, the aircraft can first enclose the object, then search for and locate the tether segment, and finally execute the binding step that completes the knot.

The system was validated through outdoor experiments and simulations. In an urban outdoor test, the UAV autonomously completed knotting on the roof of a linkway and then lifted a 15.3 kg payload to a height of 3.5 m in 42.1 s. In simulation, the method achieved shape-agnostic knotting, with success rates above 90% across 30 trials each on four structure types: pipeline, archway, billboard, and bridge.

Ablation experiments examined the role of each optimization metric. Removing enclosing planarity dropped the success rate to 8%, while omitting tether visibility or tether clearance reduced the robustness of tether binding and degraded overall task performance. These tests underscored that all three metrics are required to maintain reliable operation during the knotting process.

"While the system shows strong performance, it faces limitations: reliance on clear visual access to the tether, sensitivity to environmental disturbances, and the need for mechanical optimization of the winch-tether mechanism," said co-first author Xinhang Xu. Future research directions include control algorithms that compensate for tether-induced disturbances, multimodal sensing for more robust tether detection, and explicit modeling of how the tether interacts with surrounding structures.

Overall, the autonomous knotting system offers a method for rapid deployment of heavy-load transport in unstructured environments, where pre-installed anchors are not available. The authors suggest that such cable-driven aerial systems could expand the range of robotic logistics tasks that can be carried out in complex field settings.

Authors of the paper include Rui Jin, Xinhang Xu, Yizhuo Yang, Jianping Li, Muqing Cao, and Lihua Xie. The work received partial support from Singapore's Ministry of Education under AcRF TIER 1 Grant RG64/23 and from the National Research Foundation Medium-Sized Centre for Advanced Robotics Technology Innovation.

Research Report: Tethered UAV Autonomous Knotting on Environmental Structures for Transport