Search Results

The fusion of multi-modal perception in autonomous driving plays a pivotal role in vehicle behavior decision-making. However, much of the previous research has predominantly focused on the fusion of Lidar and cameras. Although Lidar offers an ample supply of point cloud data, its high cost and the substantial volume of point cloud data can lead to computational delays. Consequently, investigating perception fusion under the context of 4D millimeter-wave radar is of paramount importance for cost reduction and enhanced safety. Nevertheless, 4D millimeter-wave radar faces challenges including sparse point clouds, limited information content, and a lack of fusion strategies. In this paper, we introduce, for the first time, an approach that leverages Graph Neural Networks to assist in expressing features from 4D millimeter-wave radar point clouds. This approach effectively extracts unstructured point cloud features, addressing the loss of object detection due to sparsity.

Study and modeling of diesel combustion during transient operations is an important scientific objective. This is partially due to the fact that emissions under transient operations have aroused increasing attention by control groups during recent decades. The objective of this paper is to develop a combustion model to predict the peculiarities of transient combustion for developing and testing control strategies. To by-pass the complicated principles of transient combustion, the Neural Networks are applied to link the coefficients in an empirical combustion model with engine operating parameters. Finally, the Neural Networks combustion model would not only reflect the influence of turbocharge lag on combustion process during transient event, which cannot be predicted by its interpolation alternative, but also shown great potential for analyzing combustion characteristics during load increase transient event or other transient operations.

The diesel (engine)-generator is a vital part in an electric transmission system, on which an all-electric combat vehicle (AECV) is based. How to make the modeling for an engine and control it are important issues. In the paper, firstly the diesel characteristic is identified by a neural network of radial basis function(RBF) for its modeling; secondly, it is controlled by proportional integral(PI) algorithm along two kinds of charts, namely the ones in the normal and emergent statuses respectively; finally the results analyzed are shown by simulation. The control modes and conclusions are significant in the electric transmission system in the future.

How to control and analyze the turning process of a tracked vehicle with electric transmission system is an important issue. In the paper two turning control methods are presented according to its study. The balance relations of its tractive effort and power versus radius are obtained by the calculation with using the mathematical model of basic turning dynamics and constraint conditions. The model of continuously variable radius turning is implemented by a RBF neural network which is of the better identifying ability, and the more turning results can be given from it. These turning analyses are significant for the electric transmission system.