Driver Intention Recognition Method Using Continuous Hidden Markov Model

Hou, Huilian; Jin, Lisheng; Niu, Qingning; Sun, Yansong; Li, Meng

doi:10.2991/ijcis.2011.4.3.13

Cited by 15 publications

(13 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the limitations to be emphasised is the sequential nature of the decisions of the original model, which causes a decision to invalidate multiple branches in the lane change decision tree, causing discretionary lane changes hardly to be selected. This limitation has been addressed with traditional techniques, generally based on approximate reasoning, such as probabilistic trees [ 15 ] or Hidden Markov Models [ 16 ]. However, the problem remains in the fact that, to improve the predictive capacity of these models it is necessary to extend the number of input variables, with the consequent increase in the relationships between them and, therefore, the complexity of the models.…”

Section: Literature Reviewmentioning

confidence: 99%

Inferring the Driver’s Lane Change Intention through LiDAR-Based Environment Analysis Using Convolutional Neural Networks

Díaz-Álvarez

Clavijo²,

Jiménez³

et al. 2021

Sensors

View full text Add to dashboard Cite

Most of the tactic manoeuvres during driving require a certain understanding of the surrounding environment from which to devise our future behaviour. In this paper, a Convolutional Neural Network (CNN) approach is used to model the lane change behaviour to identify when a driver is going to perform this manoeuvre. To that end, a slightly modified CNN architecture adapted to both spatial (i.e., surrounding environment) and non-spatial (i.e., rest of variables such as relative speed to the front vehicle) input variables. Anticipating a driver’s lane change intention means it is possible to use this information as a new source of data in wide range of different scenarios. One example of such scenarios might be the decision making process support for human drivers through Advanced Driver Assistance Systems (ADAS) fed with the data of the surrounding cars in an inter-vehicular network. Another example might even be its use in autonomous vehicles by using the data of a specific driver profile to make automated driving more human-like. Several CNN architectures have been tested on a simulation environment to assess their performance. Results show that the selected architecture provides a higher degree of accuracy than random guessing (i.e., assigning a class randomly for each observation in the data set), and it can capture subtle differences in behaviour between different driving profiles.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Inferring the Driver’s Lane Change Intention through LiDAR-Based Environment Analysis Using Convolutional Neural Networks

Díaz-Álvarez

Clavijo²,

Jiménez³

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Weidl et al 15 optimized a Bayesian recognition network to distinguish a manoeuvre in order to reduce the average computation time. Three different identification models based on a continuous HMM were developed to identify the left-lane-change intentions and right-lanechange intentions from the normal lane-keeping intention by Hou et al 16 Berndt and Dietmayer 17 also investigated a method based on an HMM to infer a driver's intention to leave the lane-following state or the carfollowing state and to start a specific manoeuvre from the easily accessible vehicle onboard sensors.…”

Section: Introductionmentioning

confidence: 99%

Identification of a driver’s starting intention based on an artificial neural network for vehicles equipped with an automated manual transmission

Zhang

Zhu

Wang

et al. 2016

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

The driver's starting intention, which coordinates the engine output torque and the engagement speed of clutch for a vehicle equipped with an automated manual transmission, may be the key state for automated manual transmission clutch control. Fast and accurate identification of the starting intention can ensure a smooth clutch engagement and a smooth start of a vehicle. In this paper, a novel method based on an artificial error back-propagation neural network is proposed to identify the driver's starting intention. By analysis of the experimental data, the driver's starting intention can be defined strictly and divided into three modes: a slow start, a medium start and a fast start. The statistical regularity of the acceleration pedal opening is obtained on the basis of a novel method for processing the experimental data. Because in the first period of time in a starting process, the time proportion of the acceleration pedal opening over a certain value is closely related to the driver's starting intention, therefore, this statistical regularity of the acceleration pedal opening is regarded as the input of the neural network, and the Broyden-Fletcher-Goldfarb-Shanno algorithm is applied to train the neural network. The real-vehicle test results with different drivers show that the identification accuracy of the driver's starting intention is greater than 95% during the first 600 ms with the proposed artificial error backpropagation neural network. This can provide a reasonable quantization method of the driver's starting intention for smooth automated manual transmission clutch control.

show abstract

“…Dou et al [24] suggested that combining an SVM and a neural network by using weight allocations could improve the recognition rate for lane changes; vehicle data were extracted, including horizontal and vertical coordinates, speed, and type, from the next-generation simulation (NGSIM) database. Haijing et al [25] developed an identification model for lane change intention that integrated a hybrid Gaussian HMM with a SVM and used the standard deviation of the horizontal angle of the driver's head, the number of times the driver gazed at the rear-view mirror, the average scanning range, and the steering wheel angle entropy as input parameters. Semantic segmentation and target detection based on deep learning techniques have been used to detect the motion state of the target using data derived from stereo cameras and LIDAR [26]- [29].…”

Section: Introductionmentioning

confidence: 99%

Cognitive Competence Improvement for Autonomous Vehicles: A Lane Change Identification Model for Distant Preceding Vehicles

Wang

Sun

et al. 2019

IEEE Access

View full text Add to dashboard Cite

The motion status recognition of the preceding vehicle in a long-distance region is a requirement for autonomous vehicles to make appropriate decisions and increase their comprehension of the environment. At present, the lane change behavior of the leading vehicle at a short distance is detected using stereo cameras and LiDAR. However, the short detection distance (about 100 m) does not meet the requirements of high-speed driving of autonomous vehicles on expressways; this is a fundamental problem limiting the development of autonomous vehicles exhibiting human-like behavior. In this paper, a comprehensive model consisting of a back-propagation (BP) neural network model optimized by a particle swarm optimization (PSO) algorithm, and a continuous identification model is developed based on the results of naturalistic on-road experiments using millimeter-wave radar data. By considering different time-to-lane crossings (TLCs), the PSO-BP neural network model is trained using real vehicle lane change data and implemented when the TLC of the leading vehicle is longer than 1.0 s. In contrast, when the TLC is less than 1 s, the continuous recognition model of the TLC is used. By comparison with the BP neural network model, the recognition accuracy rate of the proposed model is increased from 80% to 87% after the PSO optimization for a time window of 1.0 s; these results meet the recognition requirements of the autonomous driving systems for distant targets. The findings of this paper improve the cognitive competence and safety of autonomous driving systems.INDEX TERMS Autonomous driving perception, cognitive competence, lane change behavior, BP neural network.

show abstract

Driver Intention Recognition Method Using Continuous Hidden Markov Model

Cited by 15 publications

References 9 publications

Inferring the Driver’s Lane Change Intention through LiDAR-Based Environment Analysis Using Convolutional Neural Networks

Inferring the Driver’s Lane Change Intention through LiDAR-Based Environment Analysis Using Convolutional Neural Networks

Identification of a driver’s starting intention based on an artificial neural network for vehicles equipped with an automated manual transmission

Cognitive Competence Improvement for Autonomous Vehicles: A Lane Change Identification Model for Distant Preceding Vehicles

Contact Info

Product

Resources

About