The paper presents the method for multicriteria design of a synchronous generator voltage regulator. The results of the voltage regulator polyoptimisation are compromise sets for a classic controller of type PI and fuzzy logic controller of type Takagi-Sugeno-Kang. A genetic algorithm is used to solve the polyoptimisation problem.

The purpose of this paper is to show possibility and advantages of initial control plane reproduction for an adaptive fuzzy controller. Usually the fuzzy control is used when the object is not very well known. Yet the truth is, however, that some, at least general information about the object, is available. Usually, in such a case, optimization algorithms are used to tune the control structure. The purpose of this article is to show how to find a starting point that is closer to optimum than a statistically random point, and this way to obtain better results in a shorter time.

This paper proposes an autonomous obstacle avoidance method combining improved A-star (A*) and improved artificial potential field (APF) to solve the planning and tracking problems of autonomous vehicles in a road environment. The A*APF algorithm to perform path planning tasks, and based on the longitudinal braking distance model, a dynamically changing obstacle influence range is designed. When there is no obstacle affecting the controlled vehicle, the improved A* algorithm with angle constraint combined with steering cost can quickly generate the optimal route and reduce turning points. If the controlled vehicle enters the influence domain of obstacle, the improved artificial potential field algorithm will generate lane changing paths and optimize the local optimal locations based on simulated annealing. Pondering the influence of surrounding participants, the four-mode obstacle avoidance process is established, and the corresponding safe distance condition is analyzed. A particular index is introduced to comprehensively evaluate speed, risk warning, and safe distance factors, so the proposed method is designed based on the fuzzy control theory. In the tracking task, a model predictive controller in the light of the kinematics model is devised to make the longitudinal and lateral process of lane changing meet comfort requirements, generating a feasible autonomous lane-change path. Finally, the simulation was performed in the Matlab/Simulink and Carsim combined environment. The proposed fusion path generation algorithm can overcome the shortcomings of the traditional single method and better adapt to the dynamic environment. The feasibility of the obstacle avoidance algorithm is verified in the three-lane simulation scenario to meet safety and comfort requirements.

When the traditional multi-motor speed synchronous control strategy is applied to the vacuum pump system, it is prone to the drawbacks of large synchronization error. In this paper, a simplified mathematical model of the motor for a vacuum pump is established and the transfer function is introduced, which weakens the multivariable, strong coupling and nonlinear characteristics of the motor system. According to the basic principle of the relative coupling control strategy, the neural network Proportion Integration Differentiation (PID) is introduced as a speed compensator in this system. It effectively improves the synchronization and anti-interference ability of the multi motor.