No abstract

An adaptive-learning regeneration control strategy to enhance the regeneration quality for electric vehicles (EV) is proposed. In recent years, several kinds of EV are equipped with regeneration function. For example, i-MiEV, the EV of Mitsubishi motors, whose energy regeneration ratio is adjusted via the gear shift for standard using, increasing energy regeneration ratio and decreasing energy regeneration ratio. In Taiwan, the TOBE W' car and Luxgen MPV EV, whose energy regeneration ratios are adjusted by a knob and a shaft, respectively. However, the abovementioned methods are not adaptively to be adjusted to adapt the various customs of drivers. There are some drawbacks, such as manually adjusting energy regeneration ratio and constant energy regeneration ratio, etc. Therefore, an adaptive-learning regeneration control strategy is proposed to account for the above-mentioned drawbacks. The function of the proposed strategy consists of driving mode judge unit, analyzing unit and regeneration judge unit. The driving mode judge unit determines a driving mode according to an accelerator signal, a brake signal and a speed signal from a vehicle. It outputs a coasting duration and coasting information associated with the driving mode to analyzing unit for obtaining acceleration information. The regeneration judge unit obtains target regeneration data containing target vehicle speeds that varying with time based upon the acceleration information and the regeneration reference data. Hence, the adaptive-learning regeneration control strategy can provide an adaptive energy regeneration ratio to the various customs of drivers. Finally, the simulation results show the feasibility of the proposed regeneration control strategy.

This paper develops a pedestrian potentially dangerous behaviour prediction method based on attention-long-short-term memory (Attention-LSTM) architecture to predict pedestrian trajectory and intention for the unexpected pedestrian crossing accident avoidance.To extract the road scene information for short periods of time, and improve the accuracy of subsequent intention inference and trajectory prediction, the panoramic segmentation is used to extrapolate pedestrian instances and segment areas of the environment. Next, an encoder-decoder framework based on Attention-LSTM model is proposed to infer a pedestrian's intention to run or walk out into oncoming traffic straight and to predict the future trajectory. The proposed network involves two parts: temporal feature encoder and multi-task decoder. The temporal feature encoder is mainly used to selectively emphasize the temporal features using attention mechanism, and then LSTM is employed for its encoding. In the multi-task decoder, a multi-head self-attention mechanism and LSTM are used to forecast the pedestrians' intention and future trajectory, respectively. Extensive experiments on pedestrian intention estimation (PIE) datasets demonstrate that the authors' proposed approach surpasses prior studies in terms of prediction accuracy in trajectory and intention. This study can not only effectively avoid serious accidents caused by illegal road crossing but also achieve early warning for collision avoidance.

In recent years, the research of Battery Electric Vehicle (BEV) focuses on energy storage systems, drive system and control strategy. However, one of the important subjects to enhance the performance of BEV is to research the parameters of transmission until the power battery and other technologies are mellow. For the BEV drive system parameters design, such as motor power and output torque as well as transmission gear ratio and the matching have significant impacts on each other for driving range and performance of BEV. Hence, in BEV research and developmental phase, in order to meet the requirements of vehicle performance access to the best structure, generally it must base on different needs to optimize the parameters of the drive system and to consider the combination of vehicle operation among the various components. For this reason, it will be conducive to the rapid development and be widely used for BEV. Therefore, in this paper, the discussion of the design, analysis, matching and system integration of BEV component parameters is proposed, which can be used as a reference for the future development of BEV.

No abstract