Early Recognition of Driving Intention for Lane Change Based on Recurrent Hidden Semi-Markov Model

Liu, Qingxiao; Xu, Shaohang; Lü, Chao; Yao, Hui; Chen, Huiyan

doi:10.1109/tvt.2020.3011672

Cited by 29 publications

(11 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HMM can be used to predict the probability of whether a vehicle changes lanes [ 27 – 29 ]. The vehicle and surrounding vehicles' driving state determines to a large extent whether the vehicle has the conditions for changing lanes.…”

Section: Methodsmentioning

confidence: 99%

A Novel Intelligent Approach to Lane-Change Behavior Prediction for Intelligent and Connected Vehicles

Chen

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The prediction of lane-change behavior is a challenging issue in intelligent and connected vehicles (ICVs), which can help vehicles predict in advance and change lanes safely. In this paper, a novel intelligent approach, which considering both the driving style-based lane-change environment and the driving trajectory-related parameters of the ICV and surrounding vehicles, is proposed to predict the lane-change behaviors for ICVs. By analyzing the characteristics of the lane-change behavior of the vehicle, a modified dataset for the prediction of lane-change behavior was established based on the Next-Generation Simulation (NGSIM) dataset, which is made up of real vehicle trajectories collected by camera. In the proposed approach, the hidden Markov model (HMM)-based model is designed to judge whether the current environment is suitable for lane change according to the driving environment parameters around the vehicle; then according to the driving state of the vehicle, a learning-based prediction-then-judgment model is proposed and designed to realize the prediction of the ICV’s lane-change behavior. Experiments are implemented by using the modified dataset. From the experimental results, the lane-change probability value on the target lane in the truth of the lane-change behavior calculated by the designed HMM-based model is basically above 0.5, indicating that the model can make a more accurate judgment on the surrounding lane-change environment. The proposed learning-based prediction-then-judgment model has an accuracy of 99.32% for the prediction of lane-change behavior, and the accuracy of the lane-change detection algorithm in the model is 99.56%. The experimental results show that the proposed approach has a good performance in the prediction of lane-change behavior, which could effectively assist ICVs to change lanes safely.

show abstract

Section: Methodsmentioning

confidence: 99%

A Novel Intelligent Approach to Lane-Change Behavior Prediction for Intelligent and Connected Vehicles

Chen

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Zhang 15 proposed a lane change intention estimation framework based on Gauss-Hidden Markov Mixture Model (GMM + HMM). Liu 16 proposed a semi-Markov model based on nonlinear polynomial regression and recursive hidden model (R-HSMM), which can identify driver intentions earlier than common methods and better adapt to long-term continuous state. Considering the influence of interaction between vehicles on behavior prediction, Zhang 10 proposed an interactive prediction and recognition based on game theory and GMM + HMM model to predict the intention of other vehicles and identify their behaviors.…”

Section: Related Workmentioning

confidence: 99%

Research on quantum cognition in autonomous driving

Song

Wang

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Autonomous vehicles for the intention of human behavior of the estimated traffic participants and their interaction is the main problem in automatic driving system. Classical cognitive theory assumes that the behavior of human traffic participants is completely reasonable when studying estimation of intention and interaction. However, according to the quantum cognition and decision theory as well as practical traffic cases, human behavior including traffic behavior is often unreasonable, which violates classical cognition and decision theory. Based on the quantum cognitive theory, this paper studies the cognitive problem of pedestrian crossing. Through the case analysis, it is proved that the Quantum-like Bayesian (QLB) model can consider the reasonability of pedestrians when crossing the street compared with the classical probability model, being more consistent with the actual situation. The experiment of trajectory prediction proves that the QLB model can cover the edge events in interactive scenes compared with the data-driven Social-LSTM model, being more consistent with the real trajectory. This paper provides a new reference for the research on the cognitive problem of intention on bounded rational behavior of human traffic participants in autonomous driving.

show abstract

“…As for implementing lane changing actions, driver motivation, such as extracting steering wheel angle, angular velocity, angular acceleration, is also used as a feature input to identify the driver's intention to change lanes in the early stage [12]. For instance, driver behavior recognition based camera information is used in manifold learning to predict lane-change behavior [13].…”

Section: Related Workmentioning

confidence: 99%

“…where d t is the current time step gradient, Vd t is the one-order moments estimation of the gradient at time step t, Sd t is the two-order moments estimation of the gradient at time step t, ε is added to maintain numerical stability, mostly 1 × 10 −8 , β 1 is the exponential decay rate of the one-order moments, mostly 0.9, and β 2 is the exponential decay rate of the two-order moments, mostly 0.999. L2 regularization controls model complexity, reduces overfitting, adds penalty items to the original loss function, and punishes models with high complexity, as shown in Equation (12).…”

Section: Model Optimizationmentioning

confidence: 99%

Vehicle Interaction Behavior Prediction with Self-Attention

Xin

Lian

et al. 2022

Sensors

View full text Add to dashboard Cite

The structured road is a scene with high interaction between vehicles, but due to the high uncertainty of behavior, the prediction of vehicle interaction behavior is still a challenge. This prediction is significant for controlling the ego-vehicle. We propose an interaction behavior prediction model based on vehicle cluster (VC) by self-attention (VC-Attention) to improve the prediction performance. Firstly, a five-vehicle based cluster structure is designed to extract the interactive features between ego-vehicle and target vehicle, such as Deceleration Rate to Avoid a Crash (DRAC) and the lane gap. In addition, the proposed model utilizes the sliding window algorithm to extract VC behavior information. Then the temporal characteristics of the three interactive features mentioned above will be caught by two layers of self-attention encoder with six heads respectively. Finally, target vehicle’s future behavior will be predicted by a sub-network consists of a fully connected layer and SoftMax module. The experimental results show that this method has achieved accuracy, precision, recall, and F1 score of more than 92% and time to event of 2.9 s on a Next Generation Simulation (NGSIM) dataset. It accurately predicts the interactive behaviors in class-imbalance prediction and adapts to various driving scenarios.

show abstract

Early Recognition of Driving Intention for Lane Change Based on Recurrent Hidden Semi-Markov Model

Cited by 29 publications

References 30 publications

A Novel Intelligent Approach to Lane-Change Behavior Prediction for Intelligent and Connected Vehicles

A Novel Intelligent Approach to Lane-Change Behavior Prediction for Intelligent and Connected Vehicles

Research on quantum cognition in autonomous driving

Vehicle Interaction Behavior Prediction with Self-Attention

Contact Info

Product

Resources

About