Single image super-resolution (SISR) has been a very attractive research topic in recent years. Breakthroughs in SISR have been achieved due to deep learning and generative adversarial networks (GANs). However, the generated image still suffers from undesired artifacts. In this paper, we propose a new method named GMGAN for SISR tasks. In this method, to generate images more in line with human vision system (HVS), we design a quality loss by integrating an image quality assessment (IQA) metric named gradient magnitude similarity deviation (GMSD). To our knowledge, it is the first time to truly integrate an IQA metric into SISR. Moreover, to overcome the instability of the original GAN, we use a variant of GANs named improved training of Wasserstein GANs (WGAN-GP). Besides GMGAN, we highlight the importance of training datasets. Experiments show that GMGAN with quality loss and WGAN-GP can generate visually appealing results and set a new state of the art. In addition, large quantity of high-quality training images with rich textures can benefit the results.

Rule-based traditional motion planning methods usually perform well with prior knowledge of the macro-scale environments but encounter challenges in unknown and uncertain environments. Deep reinforcement learning (DRL) is a solution that can effectively deal with micro-scale unknown and uncertain environments. Nevertheless, DRL is unstable and lacks interpretability. Therefore, it raises a new challenge: how to combine the effectiveness and overcome the drawbacks of the two methods while guaranteeing stability in uncertain environments. In this study, a multi-constraint and multi-scale motion planning method is proposed for automated driving with the use of constrained reinforcement learning (RL), named RLTT, and comprising RL, a topological reachability analysis used for vehicle path space (TPS), and a trajectory lane model (TLM). First, a dynamic model of vehicles is formulated; then, TLM is developed on the basis of the dynamic model, thus constraining RL action and state space. Second, macro-scale path planning is achieved through TPS, and in the micro-scale range, discrete routing points are achieved via RLTT. Third, the proposed motion planning method is designed by combining sophisticated rules, and a theoretical analysis is provided to guarantee the efficiency of our method. Finally, related experiments are conducted to evaluate the effectiveness of the proposed method; our method can reduce 19.9% of the distance cost in the experiments as compared to the traditional method. Experimental results indicate that the proposed method can help mitigate the gap between data-driven and traditional methods, provide better performance for automated driving, and facilitate the use of RL methods in more fields.

The brake system of a passenger car is a complex system composed of an assembly of mechanical parts, such as springs and rubber parts, and structures filled with gas and other liquids. In order to predict the brake pedal feel of a passenger car, a dynamic model of a brake system during application of the brake pedal was proposed; this considered and modelled the physical structures and the working processes of the key components. The model takes into account not only the spring preloads and the frictional forces in the booster and the master cylinder but also the reaction disc features and the variation in the bulk modulus of the brake fluid. When the pedal travel is regarded as the input, the boundary conditions for the dynamic equations of the brake system components are strictly defined. Also, bench experiments for the booster and the master cylinders were designed for parameter identification; the effectiveness of the component models was validated under conditions including with and without vacuum assist and with and without brake fluid. On the basis of these experiments, the static and dynamic characteristics of the brake pedal feel were simulated using the static model and the dynamic model of the brake system. Then, vehicle road experiments were carried out. With the assessment index of a three-quadrant diagram covering the pedal travel, the pedal force and the master cylinder pressure, as well as a four-quadrant diagram covering the pedal travel, the pedal force, the master cylinder pressure and the longitudinal deceleration of the vehicle, the results from the simulations and the experiment were compared and analysed. We conclude that this novel model of a brake system can accurately predict the brake pedal feel characteristics of passenger cars and can be used for brake system design and adjustment.

Aiming at the electromagnetic vibration and noise problem of an 8-pole 48-slot permanent magnet synchronous motor for a vehicle, the multi-physics coupling simulation model of the motor is introduced to optimize the rotor structure of the motor to reduce the vibration and noise of the permanent magnet synchronous motor. The effectiveness of the research method is verified by the bench test in the anechoic chamber. Finite element software was used to establish the stator core and system model considering the anisotropy of materials, and the simulation model was verified by modal experiment. For the 2in1 electric drive system, the electromagnetic-structure-acoustic multi-physics coupling noise prediction model is established. Based on the three-dimensional distributed electromagnetic force excitation, the electromagnetic radiation noise of the motor under full load acceleration is calculated, and the characteristics of electromagnetic noise are analyzed. The accuracy of the electromagnetic-structure-acoustic multi-physics coupling model of permanent magnet synchronous motor is verified by the bench test results of the anechoic chamber. By changing the angle and shape of the motor rotor, the cogging torque ripple between the stator and the rotor is reduced, and the 48th order harmonic amplitude is reduced. Finally, the optimized sample is tested on the vehicle, and the 48th order electromagnetic noise can be reduced by 5–15 dB(A). The accuracy of the electromagnetic-structure-acoustic multi-physics coupling model of permanent magnet synchronous motor is verified by the bench test results of the anechoic chamber. Therefore, the research results can be further used for the design and development of a vehicle permanent magnet synchronous motor and the research on the mechanism of electromagnetic vibration and noise.