… group user trajectoryprediction. Specifically, we introduce a mobile user trajectoryprediction … Term Memory networks (LSTM) with ReinforcementLearning (RL) to automate the model …
Y Huang, J Du, Z Yang, Z Zhou… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
… In summary, reinforcementlearning-based trajectory prediction methods for AVs can be classified into Table V. Such methods use MDP to maximize the expected cumulative reward …
D Choi, K Min, J Choi - IET Computer Vision, 2020 - Wiley Online Library
… method for a trajectoryprediction model based on the encoder–decoder architecture, which has the following advantages over the GMN. First, instead of predicting the probability …
B Yang, Y Lu, R Wan, H Hu, C Yang, R Ni - Expert Systems with …, 2024 - Elsevier
… are essential for accurate trajectoryprediction. Afterward, we provide a brief review of inverse reinforcementlearning, which is used to explore the trajectory–scene associations in this …
… trajectorypredictor based on ReinforcementLearning (RL) to automatically realize a high-performing LSTM predictor for a given learning … , we benefit from Transfer Learning (TL). The …
D Choi, TH An, K Ahn, J Choi - … on Machine Learning and …, 2018 - ieeexplore.ieee.org
… In this paper, we presented a future trajectoryprediction framework based on RNN and maximum margin IRL. We trained the RNN and the reward function simultaneously by minimizing …
M Geng, Z Cai, Y Zhu, X Chen… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
… trajectoryprediction framework that combines a multimodal trajectory generation network with inverse reinforcement … The current research on vehicular trajectoryprediction mainly …
… , we address an important gap in trajectoryprediction. We propose SafeCritic, a model that … multiple “real” trajectories with reinforcementlearning to generate “safe” trajectories. The …
S Wang, Z Wang, R Jiang, R Yan… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
… in the front queue, we propose a trajectoryprediction strategy that consists of two steps. The first is, at each time step, simulating the trajectories of the vehicle queue in front of CAV. The …