This paper proposes a novel incremental training mode to address the problem of Deep Reinforcement Learning (DRL) based path planning for a mobile robot. Firstly, we evaluate the related graphic search algorithms and Reinforcement Learning (RL) algorithms in a lightweight 2D environment. Then, we design the algorithm based on DRL, including observation states, reward function, network structure as well as parameters optimization, in a 2D environment to circumvent the time-consuming works for a 3D environment. We transfer the designed algorithm to a simple 3D environment for retraining to obtain the converged network parameters, including the weights and biases of deep neural network (DNN), etc. Using these parameters as initial values, we continue to train the model in a complex 3D environment. To improve the generalization of the model in different scenes, we propose to combine the DRL algorithm Twin Delayed Deep Deterministic policy gradients (TD3) with the traditional global path planning algorithm Probabilistic Roadmap (PRM) as a novel path planner (PRM+TD3). Experimental results show that the incremental training mode can notably improve the development efficiency. Moreover, the PRM+TD3 path planner can effectively improve the generalization of the model.

This paper presents the disturbance observers‐based distributed formation control for multiple quadrotor aircrafts with external disturbances and uncertain parameters using multi‐agent theory and finite‐time control method. Firstly, the finite‐time disturbance observers are proposed to handle the external disturbances on the position‐loop. Similarly, when there are both the uncertain parameters and external disturbances on the attitude‐loop, the finite‐time disturbance observers are designed to estimate the total lump disturbances. By skillfully using homogeneous system theory, Lyapunov theory, and multi‐agent theory, the distributed formation control algorithms are developed. Finally, through simulations, the efficiency of the proposed method (including the convergence rate and disturbance rejection) is verified.

This paper mainly studies the formation control problem of multiple quadrotor aircraft via fixed‐time control theory. First, based on the bilimit homogeneous theory and the framework of multiagent theory, for multiaircraft, a fixed‐time formation control strategy is proposed. Considering the external disturbance existing on the attitude loop of the aircraft, the corresponding fixed‐time disturbance observer is designed with the observer technology. Then, a fixed‐time attitude controller is designed based on the accurate observation and fast compensation from the disturbance observer. Finally, some simulations are performed to verify the effectiveness of the proposed theoretical method.

This article investigates the challenges associated with variable speed pumped storage (VSPS) units, particularly focusing on the rapidity and stability of their integration into the power grid under no-load conditions. Given the frequent connection and disconnection of VSPS units to the grid, it is essential to develop a control algorithm that enhances the efficiency of the transition process during these connections. To achieve this, first introduce the modeling of the ac-excited asynchronous generator under no-load, grid-connected conditions. Subsequently, integrating the extended Kalman observer (EKO), three types of closed-loop control strategies based on the amplitude, phase, and frequency of the generator’s stator voltage are proposed. The simulations imply that these EKO-based strategies can improve the response speed and the performance of the active disturbance rejection control (ADRC). Finally, the effectiveness of the proposed EKO-ADRC control strategy has been validated through deployment on the OPAL-OP5707XG hardware-in-the-loop (HIL) platform, which is a high-end real-time power electronics simulation platform to verify the control performance of VSPS unit control systems. The results demonstrate that the Grid synchronization under no-load speed of EKO-ADRC is faster than the proportional-integral-derivative (PID), ADRC and sliding mode control (SMC) methods.