No abstract

No abstract

This paper presents a surround vehicle motion prediction algorithm for multi-lane turn intersections using a Long Short-Term Memory (LSTM)-based Recurrent Neural Network (RNN). The motion predictor is trained using the states of subject and surrounding vehicles, which are collected by sensors mounted on an autonomous vehicle. Data on 484 vehicle trajectories were collected from real traffic situations at multi-lane turn intersections. 11,662 and 4,998 samples acquired from the vehicle trajectories were used to train and evaluate the networks, respectively. A motion planner based on Model Predictive Control (MPC) is designed to determine the longitudinal acceleration command based on the predicted states of surrounding vehicles. The future states of the subject vehicle derived by MPC is used as an input feature to reflect the interaction of subject and target vehicles in LSTM-RNN based motion predictor. The proposed algorithm was evaluated in terms of its accuracy and its effects on the motion planning algorithm based on the driving data sets. The improved prediction accuracy substantially increased safety by bounding the prediction error within the safety margin. The application results of the proposed predictor demonstrate the improved recognition timing of the preceding vehicle and the similarity of longitudinal acceleration with drivers. INDEX TERMS Autonomous vehicle, intersection driving data, motion prediction, machine learning, recurrent neural network, long short-term memory, model predictive control.

The effects of LMGS (large molecule guest substance) amount on the thermodynamics of natural gas hydrates, as well as structural characteristics of mixed hydrates of LMGS and natural gas, have been studied. The addition of 1.7 wt % neohexane (NH) to water induced inhibition of natural gas hydrates, and this inhibition effect increased with increased addition of NH up to 7.8 wt %. However, the hydrate equilibrium condition changed slightly when the concentration of NH further increased from 7.8 to 14.5 wt %. Investigations on structural characteristics were carried out by analyzing (13)C NMR spectra of mixed hydrates formed from the mixture of natural gas and NH. They indicate that two hydrate structures of II and H coexist simultaneously, and the ratio of structure H to II decreased from 0.97 to 0.43 when the NH concentration decreased from 14.5 to 7.8 wt %. In addition, it was confirmed that ethane, propane, and iso-butane gas molecules do not participate in the formation of structure H and only enclathrated in large cages of structure II. These results indicate the existence of multiple hydrate structures, which must be considered in many industrial applications when mixed hydrates are formed from multicomponent gas mixtures and liquid hydrocarbons.

No abstract