Interactive Trajectory Prediction of Surrounding Road Users for Autonomous Driving Using Structural-LSTM Network

For autonomous driving, accurate trajectory prediction is paramount, necessitating effective harnessing of spatiotemporal data. This study proposes an innovative Spatiotemporal Transformer-based model, enhancing trajectory prediction precision by leveraging a multi-head self-attention mechanism. This mechanism intricately captures both inter-vehicular interactions and temporal trajectory dependencies. The model, structured around an LSTM-based encoder-decoder framework, innovatively considers spatial interactions among observed and future trajectories during the decoding process. Evaluated on the NGSIM dataset, our model demonstrates superior predictive accuracy over existing methodologies, underscoring the efficacy of Spatiotemporal Transformers in complex dynamic environments.

Perceptions of autonomous vehicles: Relationships with road users, risk, gender and age

Numerical analysis of oxygen transport in magnetic Czochralski growth of silicon

Brain serotonin (5-HT) is known to be involved in the control of a wide range of physiological functions as well as of different kinds of behavior. Such polyfunctionality of 5-HT is mediated by numerous 5-HT receptors. Currently, 14 different 5-HT receptor subtypes expressed in the mammals have been identified. The 5-HT1А receptor is one of the most extensively characterised members of the serotonin receptor family. Increased interest to the 5-HT1А receptor is based on (1) a key role in the autoregulation of the brain serotonergic system due to the postsynaptic and presynaptic localization, (2) a great body of data demonstrating implication of 5-HT1А receptor in the control of various physiological functions (3) involvement of 5-HT1А receptors in the mechanisms of depression, anxiety and suicide. The review describes literature and original data on factors affecting the expression and functional activity of 5-HT1А receptors and the involvement of 5-HT1А receptors in the regulation of normal and pathological behavior. The structure of the 5-HT1А receptor gene is described and new data on the posttranslational regulation of 5-HT1А receptor functional activity are provided. A special focus was given to the interaction between 5-HT1А and 5-HT7 receptors followed by heterodimer formation and the role of heterodimerization in the functional inactivation of the 5-HT1А receptor. The implication of 5-HT1А receptors in the regulation of aggressive behavior, catalepsy, anxiety, depression and hibernation was shown. Special attention is focused on the involvement of 5-HT1А receptors in the regulation of 1) fear-induced aggression towards man – the basis of domestication, 2) intermale aggression underling asocial behavior in men, 3) depression and in the mechanism of antidepressant action. The described data extend the idea on the 5-HT1А receptor as a key player in the brain 5-HT system.

Vehicle trajectory information is a crucial part of improving the efficiency and the safety of the ITS. Data missing or irregular sampling in the real-world road traffic makes it hard to obtain accurate real-time vehicle trajectories. In this paper, we focus on trajectory imputation and prediction tasks with small data (magnitude set as 10 1 and 10 2 ). Limited by insufficient data, the simulation results of the existing data-driven algorithms are unsatisfactory. With car-following models integrated as prior physical information to constrain the training process, we design the car-following-informed neural network (CFINN). A multi-head self-attention layer is attached to the fully connected network layer to extract vehicle features. Different from the structure of most neural networks in regression analysis, an extra physics-based dataset is constructed in the CFINN. The loss function consists of two parts including the given trajectory's fitting error and the generated trajectory's residual error. We embed the gated recurrent unit-based encoder–decoder layer to the CFINN framework for trajectory predictions. The rationality and the superiority of our model are validated on the NGSIM dataset and the HighD dataset. Compared with baseline models in both single-vehicle and queue-typed trajectory imputation experiments, lower error can be achieved via the CFINN and coefficients of car-following models can be calibrated. According to driving regimes derived from CFINN-based trajectory prediction experiments, we discuss the impact of cut-in behaviours on the target vehicle and carry out the kinetic analysis. The novel neural network model driven by both data and physical knowledge provides technical support in vehicle status assessments and trajectory predictions.

Evolution of pedestrian evacuation considering different human behaviors

Occupant behavioral patterns, once extracted, could reveal cues about activities and space usage that could effectively get used for building systems to achieve energy savings. The ability to accurately predict the trajectories of occupants inside a room branched into different zones has many notable and compelling applications. For example - efficient space utilization and floor plans, intelligent building operations, crowd management, comfortable indoor environment, security, and evacuation or managing personnel. This paper proposes future occupant trajectory prediction using state-of-the-art time series prediction methods, i.e., Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models. These models are being implemented and compared to forecast occupant trajectories at a given time and location in a non-intrusive and reliable manner. The considered test-space for the collection of the dataset is a multi-utility area in an instrumented public building. The deployed 3D Stereo Vision Cameras capture the spatial location coordinates (x- and y- coordinates) from a bird's view angle without eliciting any other information that could reveal confidential data or uniquely identify a person. Our results showed that the GRU model forecasts were considerably more accurate than the LSTM model for the trajectory prediction. GRU prediction model achieved a Mean Squared Error (MSE) of 30.72 cm between actual and predicted location coordinates, and LSTM achieved an MSE of 47.13 cm, respectively, for multiple occupant trajectories within the monitored area. Another evaluation metric Mean Absolute Error (MAE) is used, and the GRU prediction model achieved an MAE of 3.14 cm, and the LSTM model achieved an MAE of 4.07 cm. The GRU model guarantees a high-fidelity occupant trajectory prediction for any given case with higher accuracy when compared to the baseline LSTM model.

An internal stochastic car-following model: Stochasticity analysis of mixed traffic environment

Vehicle trajectory prediction and driving intention recognition are essential technologies for improving safety and optimizing traffic efficiency in autonomous driving systems. Although traditional LSTM models are effective in trajectory prediction, their complexity and computational requirements impede practical implementation efficiency. To overcome this challenge, this study proposes a hybrid model, MTF-GRU, which integrates multi-feature fusion. Initially, the datasets are preprocessed in this study through denoising, feature extraction, and timing extraction to capture vehicle information from single and fused multi-features. Subsequently, a GRU encoding-decoding model is developed. The encoder processes the feature data to generate context vectors, while the decoder employs a combination of recursive and teaching-driven input modes. Furthermore, a teaching rate control mechanism is integrated to dynamically convert context vectors into future trajectories. The proposed model is validated using the NGSIM datasets, demonstrating superior prediction performance with multi-feature inputs outperforming single features by reducing the average endpoint displacement error by 20.5%. Our model also achieves improved accuracy rates, particularly excelling in long-term predictions with an endpoint displacement error of only 2.31 meters at 5 seconds. Moreover, the overall accuracy rate for lane change intention recognition reaches 91.3%. The model's computational efficiency supports practical deployment in real-time autonomous systems, while future efforts will integrate multi-modal sensor data to enhance adaptability in complex urban scenarios and extreme conditions. Further validation will extend to diverse traffic environments and edge computing platforms to optimize real-world robustness.

Traffic trajectory prediction plays a crucial role in ensuring the safety and efficiency of autonomous driving systems. With increasing traffic complexity, accurate and real-time prediction of surrounding vehicle trajectories has become a major challenge. Existing methods often define trajectory prediction as a static task, predicting future trajectories at once based solely on historical states. However, this approach leads to significant long-term prediction errors and struggles to capture potential hazards in complex, dynamic traffic conditions. In this paper, we propose a lightweight end-to-end trajectory prediction model integrating a long-term guidance module and a short-term feedback mechanism. The long-term guidance module predicts the final driving goal that remains unchanged over time, guiding the vehicle’s eventual position. The short-term feedback mechanism captures the dynamic driving environment, continuously updating the model with immediate changes in vehicle interactions and road conditions to avoid unexpected dangers. At each time step, our approach leverages accumulated historical information and incorporates both short-term and long-term future information to iteratively predict multi-modal trajectories, effectively reducing long-term prediction errors. Additionally, our method introduces the Mamba module, which enhances real-time prediction performance and reduces historical information loss during long-term prediction. Experiments on the Argoverse 1 and Argoverse 2 datasets show that our method, with fewer parameters and higher inference speed, achieves results comparable to or surpassing previous state-of-the-art methods.

In a mixed traffic environment of human and autonomous driving, it is crucial for an autonomous vehicle to predict the lane change intentions and trajectories of vehicles that pose a risk to it. However, due to the uncertainty of human intentions, accurately predicting lane change intentions and trajectories is a great challenge. Therefore, this paper aims to establish the connection between intentions and trajectories and propose a dual Transformer model for the target vehicle. The dual Transformer model contains a lane change intention prediction model and a trajectory prediction model. The lane change intention prediction model is able to extract social correlations in terms of vehicle states and outputs an intention probability vector. The trajectory prediction model fuses the intention probability vector, which enables it to obtain prior knowledge. For the intention prediction model, the accuracy can be improved by designing the multi-head attention. For the trajectory prediction model, the performance can be optimized by incorporating intention probability vectors and adding the LSTM. Verified on NGSIM and highD datasets, the experimental results show that this model has encouraging accuracy. Compared with the model without intention probability vectors, the impact of the model on NGSIM dataset and highD dataset in RMSE is improved by 57.27% and 58.70% respectively. Compared with two existed models, evaluation metrics of the intention prediction can be improved by 7.40-10.09% on NGSIM dataset and 2.17-2.69% on highD dataset within advanced prediction time 1s. This method provides the insights for designing advanced perceptual systems for autonomous vehicles.

Tactical decision-making and strategic motion planning for autonomous highway driving are challenging due to predicting other road users' behaviors, diversity of environments, and complexity of the traffic interactions. This paper presents a novel end-to-end continuous deep reinforcement learning approach towards autonomous cars' decision-making and motion planning. For the first time, we define both states and action spaces on the Frenet space to make the driving behavior less variant to the road curvatures than the surrounding actors' dynamics and traffic interactions. The agent receives time-series data of past trajectories of the surrounding vehicles and applies convolutional neural networks along the time channels to extract features in the backbone. The algorithm generates continuous spatiotemporal trajectories on the Frenet frame for the feedback controller to track. Extensive high-fidelity highway simulations on CARLA show the superiority of the presented approach compared with commonly used baselines and discrete reinforcement learning on various traffic scenarios. Furthermore, the proposed method's advantage is confirmed with a more comprehensive performance evaluation against 1000 randomly generated test scenarios. Code: https://github.com/MajidMoghadam2006/RL-frenet-trajectory-planning-in-CARLA

Connected and Automated Vehicles (CAV) have been rapidly developed, which, inevitably, renders that human-driven and autonomous vehicles share the road. Thus, trajectory prediction is an important research topic, which helps each CAV to efficiently follow a Human-Driven Vehicle (HV). In a wider scope, trajectory prediction, also, helps to improve the throughput of traffic flow and enhance its stability. To realize the trajectory prediction of Connected and Automated Vehicles to Human-Driven Vehicles, a car-following model, which is based on trajectory data, was established. Adding deep neural networks and an Attention mechanism, this paper established a data-driven car-following model, based on CNN-BiLSTM-Attention for CAV, to predict trajectory, by referring to the modeling idea of the traditional car-following model. The trajectory data in the next-generation-simulation (NGSIM) datasets that match the car-following characteristics were selected. In addition, noise-reduction pre-processing of the trajectory data was performed, to make it match the actual car-following situation. Experiments, for selecting the optimal structure of the model and the method of trajectory prediction, were carried out. The data-driven car-following models, such as LSTM, GRU, and CNN-BiLSTM, were selected for comparative analysis of trajectory prediction. The results show that the CNN-BiLSTM-Attention model has the smallest MAE and MSE as well as the largest R2. The CNN-BiLSTM-Attention model has the highest accuracy in vehicle-trajectory prediction. The model can, effectively, realize vehicle-trajectory prediction and provide a theoretical basis for vehicle-trajectory-based velocity guidance of Human-Driven Vehicles. In the future, the model can, also, provide the theoretical basis for Connected and Automated Vehicles, to make car-following decisions in mixed traffic flow.

Safe and comfortable path planning in a dynamic urban environment is essential to an autonomous vehicle. This requires the future trajectories of all other road users in the environment of the vehicle. These trajectories are predicted through modeling the motion and behaviour of these road users. In this work we state that for efficient trajectory prediction only motion indicators are not sufficient. Therefore, we propose using a curvilinear coordinate system with curvature as road infrastructure indicators to improve motion modeling and trajectory prediction. With experiments, we show that the curvilinear coordinate system with curvature sufficiently incorporates the road structure. Furthermore, we show that a sequence-to-sequence RNN model is suitable to incorporate road curvature indicators directly into the modeling and prediction.

Accurately predicting trajectory of vehicles is a key capability for autonomous vehicles during real-world driving scenarios, which requires the autonomous vehicles to perceive the surrounding environment and analyze the track histories to judge the driving intentions of the focused surrounding vehicles in order to make decisions for the next trajectory route. In this paper, we propose a hybrid attentional trajectory prediction model incorporating both temporal attentional and spatial attentional mechanisms, which has a temporal attentional model that can be used to capture important temporal information affecting trajectory prediction and a spatial attentional mechanism that can better introduce spatial dependencies and potential driving intention information of surrounding and target vehicles. The models are evaluated for performance on publicly available highD and NGSIM datasets. By comparing the results with other state-of-the-art models for trajectory prediction, it is found that the model achieves an improvement in prediction performance, indicating that this model can more accurately learn driving information from the spatio-temporal perspective of the target driving vehicle’s attention. Further, by qualitatively analyzing the distribution of attention weights from the spatio-temporal perspective, we assessed the interpretability of the model on the maneuverability judgment of the vehicle.