On-Board Implementation of the Differential Thrust-Based Lateral-Directional Model Predictive Control in the Distributed Electric Propulsion Aircraft

Distributed electric propulsion (DEP) technology has demonstrated significant potential due to its advantages in reducing carbon emissions, enhancing aerodynamic efficiency, and improving takeoff and landing performance. Moreover, DEP provides a hardware-level opportunity to integrate multiple propulsors into flight control, particularly for lateral-directional control. However, achieving this capability requires a multivariable optimal control approach to effectively coordinate both the propulsors and aerodynamic control surfaces. To address this challenge, this article proposes a model predictive control (MPC) scheme for lateral-directional control. Based on a linearized DEP flight dynamics model, the proposed MPC scheme enables dynamic optimal lateral-directional control for DEP aircraft. For the first time, the scheme has been implemented in both software and hardware using Jetson Orin and unmanned aerial vehicle (CUAV) autopilot running PX4 firmware. A DEP flight platform with eight electric propulsors, a wingspan of 1.8 m, and a cruise speed of 14 m/s is developed to integrate and validate the proposed control system. Flight tests demonstrate that the proposed scheme not only achieves dynamically optimal control but also satisfies real-time requirements.