Explore the vast range of titles available.

Effective trajectory planning is critical for achieving precise autonomous navigation and safe landing of parafoil delivery systems. However, current parafoil trajectory planning still faces challenges in ensuring consistency between actual system behavior and algorithmic real-time performance. Due to the strong fluid–structure interaction (FSI) between the flexible canopy and airflow, traditional dynamic models based on point mass and rigid-body assumptions often lack aerodynamic accuracy. These models produce planned trajectories in simulation environments that are inconsistent with the actual system’s behavior and cannot directly provide an effective reference for airdrop experiments. Additionally, traditional planning methods require a significant amount of time to calculate complex dynamic models and generate fixed trajectories in advance. These methods not only fail to provide usable results in a short period of time, but also cannot prevent the accumulation of tracking errors by adjusting the target trajectory in real time. To address these issues, this paper proposes a flexible 8-degree-of-freedom (8-DOF) dynamic model based on the FSI method, utilizing the actual aerodynamic parameters of the canopy to achieve improved consistency with the behavior of the actual system. The Soft Actor–Critic (SAC) algorithm is then employed to achieve real-time trajectory planning for parafoil airdrop systems, addressing the real-time planning performance limitations of traditional algorithms. The airdrop experiments validate that the simulation trajectories generated using this model demonstrate higher consistency with actual flight trajectories, providing more accurate references for pre-flight trajectory optimization. Moreover, the proposed method enables real-time trajectory planning and dynamically adjusts target trajectories based on the current position and attitude of the parafoil, effectively mitigating the accumulation of errors.

A parafoil delivery system has usually been used in the fields of military and civilian airdrop supply and aircraft recovery in recent years. However, since the altitude of the unpowered parafoil is monotonically decreasing, it is limited by the initial flight altitude. Thus, combining the multiple constraints, such as the ground obstacle avoidance and flight time, it puts forward a more stringent standard for the real-time performance of trajectory planning of the parafoil delivery system. Thus, to enhance the real-time performance, we propose a new parafoil trajectory planning method based on an improved twin delayed deep deterministic policy gradient. In this method, by pre-evaluating the value of the action, a scale of noise will be dynamically selected for improving the globality and randomness, especially for the actions with a low value. Furthermore, not like the traditional numerical computation algorithm, by building the planning model in advance, the deep reinforcement learning method does not recalculate the optimal flight trajectory of the system when the parafoil delivery system is launched at different initial positions. In this condition, the trajectory planning method of deep reinforcement learning has greatly improved in real-time performance. Finally, several groups of simulation data show that the trajectory planning theory in this paper is feasible and correct. Compared with the traditional twin delayed deep deterministic policy gradient and deep deterministic policy gradient, the landing accuracy and success rate of the proposed method are improved greatly.

Amid intensifying competition in airdrop equipment development, there is a growing demand for large-load, high-precision, maneuverable, and low-cost airdrop systems. However, Precision Aerial Delivery Systems (PADS) exhibit structural complexity and immature dynamics theory for flexible-body parachute/parafoil systems. Flight testing proves prohibitively expensive, while random environmental interference hinders data consistency. To address these challenges, this paper integrates navigation control systems and actuators with dynamics models through a Hardware-in-the-Loop (HIL) simulation system for comprehensive performance evaluation.