Real-Time Coverage Path Planning for Fixed-Wing Aerial Robots Using Partial Gradient-Based MPC and Augmented Dubins Trajectories

Coverage Path Planning (CPP) for fixed-wing aerial robots remains challenging in dynamic and partially unknown environments due to the need to simultaneously satisfy coverage completeness, kinematic feasibility, and real-time adaptability. Conventional approaches typically rely on pre-defined geometric patterns or simplified motion models, which limit their effectiveness when encountering environmental disturbances or unforeseen obstacles. This paper presents a hybrid real-time CPP framework that integrates an offline coverage strategy with an online optimisation and control scheme. In the offline phase, a back-and-forth coverage pattern is generated based on the geometric properties of the region and sensor characteristics, ensuring full nominal coverage. During execution, this trajectory is adaptively refined using a Model Predictive Control (MPC) formulation augmented by a policy gradient-based update mechanism and an augmented Dubins path smoothing strategy. The MPC framework explicitly accounts for vehicle dynamics, actuator limitations, and obstacle avoidance constraints, while the policy gradient component improves the responsiveness of the optimisation process under rapidly changing conditions. The augmented Dubins formulation enables smooth and dynamically feasible transitions, allowing the vehicle to deviate from and reliably return to the nominal coverage path after disturbance or avoidance manoeuvres. Simulation results in cluttered environments with static and dynamic obstacles demonstrate that the proposed approach achieves improved tracking consistency and smoother trajectories, while maintaining real-time feasibility. Moreover, the proposed approach reduces the maximum computational burden by approximately 0.35 s compared to the conventionally utilized optimization algorithm. These results highlight the potential of the framework for practical deployment in fixed-wing aerial coverage missions operating in uncertain environments.