Modeling human-like longitudinal driver model for intelligent vehicles based on reinforcement learning

Xie, Ju; Xu, Xing; Wang, Rui; Jiang, Haobin

doi:10.1177/0954407020983579

Cited by 12 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the stochastic nature and the variability of different drivers, modelling driver behaviour is quite challenging. 9 Some research has been done about this. A car-following driver model was modelled with a statistical method of Gaussian mixture model (GMM) that represent the distributions of raw pedal operation signals or spectral features extracted through spectral analysis of the raw pedal operation signals.…”

Section: Introductionmentioning

confidence: 99%

RETRACTED: The human-like trajectory planning for autonomous vehicles based on optimal control in a test track environment

Wang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part D:

Self Cite

View full text Add to dashboard Cite

A human-like trajectory could give a safe and comfortable feeling for the occupants in an autonomous vehicle especially in corners. The research of this paper focuses on planning a human-like trajectory along a section road on a test track using optimal control method that could reflect natural driving behaviour considering the sense of natural and comfortable for the passengers, which could improve the acceptability of driverless vehicles in the future. A mass point vehicle dynamic model is modelled in the curvilinear coordinate system, then an optimal trajectory is generated by using an optimal control method. The optimal control problem is formulated and then solved by using the Matlab tool GPOPS-II. Trials are carried out on a test track, and the tested data are collected and processed, then the trajectory data in different corners are obtained. Different TLCs calculations are derived and applied to different track sections. After that, the human driver’s trajectories and the optimal line are compared to see the correlation using TLC methods. The results show that the optimal trajectory shows a similar trend with human’s trajectories to some extent when driving through a corner although it is not so perfectly aligned with the tested trajectories, which could conform with people’s driving intuition and improve the occupants’ comfort when driving in a corner. This could improve the acceptability of AVs in the automotive market in the future. The driver tends to move to the outside of the lane gradually after passing the apex when driving in corners on the road with hard-lines on both sides.

show abstract

Section: Introductionmentioning

confidence: 99%

RETRACTED: The human-like trajectory planning for autonomous vehicles based on optimal control in a test track environment

Wang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part D:

Self Cite

View full text Add to dashboard Cite

show abstract

“…Interest on driver modeling has grown in the recent years, thus many types of drivers can be found in the literature, such as path-following drivers to assess vehicle handling and stability, [17][18][19] or more recently driver models for autonomous vehicle control. [20][21][22][23] For the current EiL application, the driver has a more basic, yet crucial function. For a given speed cycle, its main role is longitudinal speed control, that is, to follow the speed demand by acting on the engine accelerator pedal as well as the brake model.…”

Section: Contextmentioning

confidence: 99%

Design and off-line tuning of a longitudinal driver model for EiL applications

Gilormini

Chessé

Tauzia

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

Tightnening emission regulations and increasing powertrain complexity lead car manufacturers to rely on novel testing methods in order to frontload development. Among these, Engine-in-the-Loop, that is, the coupling of a physical internal combustion engine (ICE) on a testbed with a virtual environment, shows great promise for emission- and consumption-related tasks. In particular, this study focuses on the driver model, a simple yet crucial component of the virtual environment. A longitudinal driver model is developed in Simulink based on the PI-regulation structure and augmented with anti-windup, cycle preview, and takeoff strategy. While the PI approach is generally chosen in the literature, this study details the implementation of the added functions, and proposes a method for the gains of the model to be tuned in simulation by considering engine dynamics, and using several performance indicators. The virtual driver is then tested in a complete EiL setup simulating an electric hybrid driveline and shows satisfactory cycle-following and overall behavior on a WLTC. Robustness of the tuning method is also studied by varying vehicle parameters on the EiL bench.

show abstract

“…This method has been widely used in human-like driving and decision making [20][21][22], especially the deep neural network approach, usually combined with reinforcement and imitation learning, has been used to model various driving tasks in different environments [23]. A human-like longitudinal driving model for AVs is established, by using reinforcement learning (RL) [24]. Deep reinforcement learning is also used to design human-like car-following models [25].…”

Section: Introductionmentioning

confidence: 99%

Human-like Decision Making for Autonomous Vehicles at the Intersection Using Inverse Reinforcement Learning

Zheng

Yang

et al. 2022

Sensors

View full text Add to dashboard Cite

With the rapid development of autonomous driving technology, both self-driven and human-driven vehicles will share roads in the future and complex information exchange among vehicles will be required. Therefore, autonomous vehicles need to behave as similar to human drivers as possible, to ensure that their behavior can be effectively understood by the drivers of other vehicles and be more in line with the cognition of humans on driving behavior. Therefore, this paper studies the evaluation function of human drivers, using the method of inverse reinforcement learning, aiming for the learned behavior to better imitate the behavior of human drivers. At the same time, this paper proposes a semi-Markov model, to extract the intentions of surrounding related vehicles and divides them into defensive and cooperative, leading the vehicle to adopt a reasonable response to different types of driving scenarios.

show abstract

Modeling human-like longitudinal driver model for intelligent vehicles based on reinforcement learning

Cited by 12 publications

References 37 publications

RETRACTED: The human-like trajectory planning for autonomous vehicles based on optimal control in a test track environment

RETRACTED: The human-like trajectory planning for autonomous vehicles based on optimal control in a test track environment

Design and off-line tuning of a longitudinal driver model for EiL applications

Human-like Decision Making for Autonomous Vehicles at the Intersection Using Inverse Reinforcement Learning

Contact Info

Product

Resources

About