To enhance the ride comfort and control performance of the semi-active pneumatic suspension system (PSS) of automobiles on the different road surfaces, a machine learning method (MLM) developed on the optimal control rules of the fuzzy logic control is proposed for the semi-active PSS. A nonlinear dynamic model of the automobile with eight degrees of freedom (DOF) is established to compute the results. The root mean square (RMS) accelerations of the vertical driver’s seat and the pitching angle and rolling angle of the automobile are selected to evaluate the ride comfort of the automobile on the rigid road and off-road terrain surfaces. The research results show that the off-road terrain surfaces remarkably affect the ride comfort of the automobile, especially at a high moving speed range of the automobile over 17.5 m/s. The performance of the MLM in improving the ride comfort of the automobile is better than the fuzzy logic control under various simulation conditions. Particularly, the RMS accelerations of the vertical driver’s seat and the pitching angle and rolling angle of the automobile with the MLM are smaller than that of the fuzzy logic control by 14.6%, 9.6%, and 5.3% on the rigid road surfaces and reduced by 14.9%, 8.7%, and 9.8% on the soil terrain of off-road terrain surfaces, respectively. However, the research results also indicate that the performance of the MLM significantly depends on the data map of the learning process. Thus, to further enhance the performance of the MLM, the data map for the machine learning process should be expanded under different operating conditions of the automobile.
Based on the background of the Internet of Things, this paper proposes a communication node information transmission protocol for wireless network sensors. This method adopts the strategy of alternating listening and sleeping. It can also reflect the overall characteristics of the monitoring range while meeting the real-time requirements. The protocol method proposed in this paper has the characteristics of low energy consumption and no synchronization. The algorithm reserves multi-hop timeslots at one time in the data forwarding direction at the frontier of data transmission. It avoids the delay caused by intermittent transmission of single-packet data on the forwarding path. The simulation results show that the energy efficiency and delay performance are much better than SMAC.
As one of the indispensable basic branches of computer vision, visual object tracking has very important research value. Therefore, a deep learning based on robot vision tracking is evaluated. Based on the basic principles of target tracking and search principle, a deep learning algorithm for visual tracking is constructed, and finally, evaluated, and simulated. The results showed that the accuracy rate increased from 90.9% to 90.13% after the addition of channel attention mechanism module. Variance was reduced from 3.78% to 1.27%, with better stability. The EAO, accuracy, and robustness of the algorithm are better than those without significant region weighting strategy. The strategy of using the improved residual network SE-ResNet network to extract multiresolution features from the correlation filtering framework is effective and helpful to improve the tracking performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.