Parafoil systems have gained widespread attention for their ability to achieve high-precision landings, making them ideal for military airdrops and aerospace recovery missions. However, the strong nonlinearities, coupling effects, and environmental disturbances involved bring significant challenges to controller design. To address these issues, this paper proposes an adaptive fuzzy control method enhanced by the Deep Deterministic Policy Gradient (DDPG) algorithm. A fuzzy controller with dual inputs and dual independent outputs is designed, using heading error and altitude as input variables. The membership function parameters of the fuzzy controller are continuously optimized through reinforcement learning, allowing the control system to adapt to dynamic environments while maintaining interpretability. An eight-degree-of-freedom model of the parafoil-payload system is employed to simulate the full-body dynamics, and a hierarchical control strategy is introduced to handle different flight stages. Simulation results show that the proposed method effectively improves both landing accuracy and adaptability.