This paper proposes an improved TD3 (Twin Delayed Deep Deterministic Policy Gradient) algorithm to address the flaws of low success rate and slow training speed, when using the original TD3 algorithm in mobile robot path planning in dynamic environment. Firstly, prioritized experience replay and transfer learning are introduced to enhance the learning efficiency, where the probability of beneficial experiences being sampled in the experience pool is increased, and the pre-trained model is applied in an obstacle-free environment as the initial model for training in a dynamic environment. Secondly, dynamic delay update strategy is devised and OU noise is added to improve the success rate of path planning, where the probability of missing high-quality value estimate is reduced through changing the delay update interval dynamically, and the correlated exploration of the mobile robot inertial navigation system in the dynamic environment is temporally improved. The algorithm is tested by simulation where the Turtlebot3 robot model as a training object, the ROS melodic operating system and Gazebo simulation software as an experimental environment. Meanwhile, the result shows that the improved TD3 algorithm has a 16.6 % increase in success rate and a 23.5 % reduction in algorithm training time. A generalization experiment was designed finally, and it indicates that superior generation performance has been acquired in mobile robot path planning with continuous action spaces through the improved TD3 algorithm.

The motion of unmanned surface vehicles (USVs) is frequently disturbed by ocean wind, waves, and currents. A poorly designed controller will cause failures and safety problems during actual navigation. To obtain a satisfactory motion control performance for the USVs, a model predictive control (MPC) method based on an improved Nonlinear Disturbance Observer (NDO) is proposed. First, the USV model is approximately linearized and MPC is designed for the multivariable system with constraints. To compensate for the influence of disturbances, an improved NDO is designed where the calculation time for MPC is reduced. Finally, comparison simulations are conducted between MPC with the original NDO and MPC with an improved NDO, and the results show that they have similar performances to the USVs. However, the proposed method has fewer parameters that need to be tuned and is much more time-saving compared to MPC with a traditional NDO.

This paper presents an extensive analysis of the properties of different control horizon sets in an Extended Prediction Self-Adaptive Control (EPSAC) model predictive control framework. Analysis is performed on the linear multivariable model of the steam/water loop in large-scale watercraft/ships. The results indicate that larger control horizon values lead to better loop performance, at the cost of computational complexity. Hence, it is necessary to find a good trade-off between the performance of the system and allocated or available computational complexity. In this original work, this problem is explicitly treated as an optimization task, leading to the optimal control horizon sets for the steam/water loop example. Based on simulation results, it is concluded that specific tuning of control horizons outperforms the case when only a single valued control horizon is used for all the loops.

Fast and strong interacting systems are hard to control from both performance and control effort points of view. Moreover, multiple objective functions or objectives with various identifiers of varying weights can hold unfeasible solutions at times. A novel cost objective function is proposed here to overcome both feasibility set limitations and computational burdens. An application example is used to illustrate its added value, which is a fast and strong interacting multivariable system: a landscape office lighting regulatory problem. New lighting technology and an intelligent control system have been produced to improve control accuracy and reduce power consumption. While optimizing the hardware of the lighting system, the energy consumption can be further reduced by applying advanced control strategy in the lighting system. This paper designed a fractional order distributed model predictive control (FOMPC) scheme to realize the reference tracking and stability control of multiple illuminations at the same time. In order to test the efficiency of the control strategy, an experiment was carried out on the lighting setup based on the dSPACE control system. The FOMPC scheme was analyzed through simulation and lighting experiments based on the dSPACE control system. Through a comparison with the mode predictive control (MPC) scheme, the superiority of the FOMPC scheme for the dynamic behavior and control performance of multiple lighting systems was verified. The research results provide a basis for multiple lighting control and its application.