To resolve the local path generating problem for unmanned ground vehicles (UGV) in unstructured environments, a method combining a basic path subdivision method for topological maps of local environments and a Support Vector Machine (SVM) is proposed in this paper. Based on the basic path subdivision method, topological maps of local environments can be extracted with little expanded nodes, without the constraints of obstacle representation, so meeting the need for autonomous navigation in unstructured environments. Next, to optimize the candidate routes in topological maps and generate a smoother path, an SVM is introduced. The candidate routes boundary points are defined as positive and negative samples, and SVMs are employed to train the separating surface. An original SVM is extended to satisfy extra constraints such as vehicle position and heading constraints. Experimental results show the effectiveness and advantages of the proposed method.

The longitudinal altitude and airspeed coordinated control is critical for unmanned fixed-wing aircrafts, especially during the taxiing landing process However, due to the strongly coupled relation of kinetic energy and gravitational potential energy, the precision longitudinal control is difficult. The traditional longitudinal control method based on PID (Proportional Integral-Differential) usually adopts different control parameters at different flight stages. But the regulation of the control structure is complex, and some fluctuation may exist during the transition process, which will affect the control performance. The TECS (Total Energy Control System) method adopts the decoupling strategy that the throttle controls the total energy and the pitch angle controls the energy balance between kinetic and potential. However, the parameters-adjusting process is relatively complex and the adaptability to model nonlinearity is imperfect. Based on the INDI (Incremental Nonlinear Dynamic Inversion) method and the trim relation of the aircrafts, a longitudinal altitude/airspeed decoupling control method is proposed in this paper. Firstly, the process of control law design is systematically described. The core points of states balance and transition are described for the design of the method. Secondly, the feasibility and basic control effects of the control method are verified by simulations. During the simulation, a variable coefficient PI pseudo is proposed, to accommodate the requirements of accurate gliding and smooth grounding. Finally, to further verify the effectiveness and superiority of the proposed method, comparisons between the proposed method and the TECS method is carried out. From the simulation results, the in the sense of precise altitude control during auto landing process.

The tail-sitter unmanned aerial vehicles have the advantages of multi-rotors and fixed-wing aircrafts, such as vertical takeoff and landing, long endurance and high-speed cruise. These make the tail-sitter unmanned aerial vehicle capable for special tasks in complex environments. In this article, we present the modeling and the control system design for a quadrotor tail-sitter unmanned aerial vehicle whose main structure consists of a traditional quadrotor with four wings fixed on the four rotor arms. The key point of the control system is the transition process between hover flight mode and level flight mode. However, the normal Euler angle representation cannot tackle both of the hover and level flight modes because of the singularity when pitch angle tends to $\pm {90}^{°}$ . The dual-Euler method using two Euler-angle representations in two body-fixed coordinate frames is presented to couple with this problem, which gives continuous attitude representation throughout the whole flight envelope. The control system is divided into hover and level controllers to adapt to the two different flight modes. The nonlinear dynamic inverse method is employed to realize fuselage rotation and attitude stabilization. In guidance control, the vector field method is used in level flight guidance logic, and the quadrotor guidance method is used in hover flight mode. The framework of the whole system is established by MATLAB and Simulink, and the effectiveness of the guidance and control algorithms are verified by simulation. Finally, the flight test of the prototype shows the feasibility of the whole system.

To enhance the robustness of the nonlinear dynamic inverse (NDI) technique in the presence of model uncertainties, this study introduces a control scheme that integrates a reinforcement learning (RL) agent with the NDI approach. Initially, a fixed-wing unmanned aerial vehicle (UAV) is selected as the primary subject of investigation, and its attitude dynamics model is established. A control system employing a nonlinear disturbance observer within a dynamic inverse framework has been developed based on this model. Subsequently, the stability of the resulting control system is verified through Lyapunov analysis. Following this, a twin delayed deep deterministic policy gradient (TD3) agent is introduced, with the closed-loop system serving as the training environment. Through continuous interaction with its surroundings, the agent learns to dynamically adjust control parameters in response to control errors. Ultimately, the trained RL agent is utilized to optimize the control parameters for the dynamic system, and a flight simulation of the fixed-wing UAV’s attitude control is conducted. The simulation results demonstrate that the control parameters can be adaptively adjusted using the TD3-NDI method, which mitigates overshoot and suppresses oscillations during the control process. These findings confirm the effectiveness and robustness of the proposed control strategy.