In this paper, an optimization algorithm of energy recovery efficiency is proposed for parallel hydraulic hybrid systems (PHHS) using dynamic programming (DP). Global optimal solution of pump displacement and transmission ratio under the known urban drive cycles is obtained by using the DP approach, where the total amount of energy recovery is defined as the cost function, and the pump displacement and the transmission ratio of the torque coupler are defined as the deciding variables. Two major steps are involved in verifying the proposed approach. Firstly, a PHHS Simulink model is accurately obtained by repeated comparison with the bench test. Subsequently, we derive a parallel hydraulic hybrid vehicle (PHHV) from adding a hydraulic hybrid system to an electric vehicle in ADVISOR (advanced vehicle simulator). This vehicle is used to validate the effectiveness of the proposed method in energy recovery efficiency.

Cell-phone use while driving results in potentially severe safety hazards. In this paper, a scheme for detecting cell-phone use that is based on deep learning is proposed, which can eliminate the potential risk by detecting the driver behavior and issuing an early warning. The proposed scheme consists of two stages: model training and practical testing. In the former, a multi-angle arrangement of cameras is first designed. Then, based on self-established data set, two independent convolutional neural networks (CNNs) are trained by optimizing the size and number of the convolution kernels, which can efficiently recognize cell-phones and hands in real time. In the testing stage, dynamic region extraction and skin color detection are employed as preprocessing to improve the accuracy of target recognition. Then, with the trained CNNs, the detection of cell-phone and hand targets is carried out, and the corresponding early warning is issued based on the distance of the interaction between the cell-phone and the hand. Numerous experiments are conducted and the results demonstrate that the proposed scheme can accurately detect cell-phone use during driving in real time, with a running time of 144 fps and an accuracy of 95.7%.INDEX TERMS Cell-phone use, deep learning, dynamic region extraction. I. INTRODUCTION

Driver requires a high level of concentration when driving at high speed. Cell-phone use while driving can cause serious problems. To solve this issue, this paper proposes a proposal scheme about driver cell-phone use detection based on CornerNet-Lite Network. The scheme can eliminate the traffic accident risk caused by cell-phone use through detecting driver cell-phone use. The scheme includes two stages: model training and simulation test. Firstly, the data set of cell-phone use was established. Secondly, the data set was preprocessed by the preset processing method. Finally, the CornerNet-Lite network architecture was optimized to reduce the training time and improve the detection accuracy and real-time detection. Through a large number of experiments, the results showed that the scheme had a good detection effect, with the accuracy of 86.2% and with 30fps. Under the noise interference of simulated cab environment, it still had a high robustness.

No abstract