A singularity-free online neural network-based sliding mode control of the fixed-wing unmanned aerial vehicle optimal perching maneuver

The perching behavior of birds is very common in nature, but it is extraordinary for fixed-wing unmanned aerial vehicles (UAVs), because of post-stall maneuver. The main problems of the perching maneuver are weak control authority and strong model nonlinearity. This paper proposed a control method to realize the fixed-wing perch maneuver. For the problem of weak control authority, an optimal trajectory is designed. And a feasible reference trajectory is given under the conditions of model dynamic characteristics and actuator constraints. For the problem of strong nonlinearity, a singularity-free online neural network-based sliding mode controller is designed. The simulation test shows that the proposed controller can ensure a reliable perching maneuver under the initial state deviation.