Search papers, labs, and topics across Lattice.
This paper presents a collision-free trajectory generation and robust nonlinear distributed model predictive control (NLDMPC) strategy for tethered multi-rotor UAVs transporting heavy loads. The key innovation lies in reformulating non-differentiable collision avoidance constraints using strong duality to create a smooth optimization problem for trajectory planning. A constraint-tightening-based robust NLDMPC is then developed to handle composite disturbances, ensuring robust feasibility and stability without terminal constraints.
Smooth optimization unlocks collision-free trajectory planning for tethered multi-rotor UAVs, enabling robust heavy-load transport in complex environments.
This article investigates the collaborative transportation planning and control of tethered multi-rotor uncrewed aerial vehicles within intelligent transportation systems. These uncrewed aerial vehicles handle heavy-load delivery, including emergency airdrop and aerial assembly of structural components. To ensure algorithm generality, obstacles, loads, and uncrewed aerial vehicles are modeled as unions of convex sets. Collision avoidance constraints, originally nondifferentiable due to convex set distances, are exactly reformulated into differentiable forms via strong duality. This leads to a smooth, optimization-based trajectory planning framework with obstacle avoidance. Considering composite disturbances, a robust nonlinear distributed model predictive control strategy based on constraint tightening is developed, ensuring robust feasibility and stability without terminal constraints. Numerical simulations in cluttered environments validate the method’s effectiveness and applicability to next-generation aerial logistics and emergency response in complex terrains. Note to Practitioners—In aerial logistics and emergency response, practitioners often face the challenge of transporting heavy loads in regions that are inaccessible to ground vehicles. Conventional uncrewed aerial vehicles solutions are limited by payload capacity and difficulties in maintaining stability under disturbances and cluttered environments. This work investigates tethered multi-rotor uncrewed aerial vehicles systems for collaborative load transportation, providing a practical alternative for emergency airdrop, structural assembly, and delivery tasks. By reformulating obstacle avoidance constraints into smooth, differentiable forms, the proposed trajectory planning framework enables efficient optimization in complex environments. Furthermore, a robust distributed model predictive control strategy ensures safe and stable flight under uncertainties without requiring terminal constraints, reducing computational overhead and enhancing reliability. Simulation results in cluttered scenarios demonstrate that the method effectively guarantees collision-free load transportation and stable cooperation among uncrewed aerial vehicles, highlighting its potential for real-world deployment in time-critical logistics and rescue missions.
