Adaptive Hierarchical Energy Management Design for a Plug-In Hybrid Electric Vehicle

Li, Teng; Tang, Xiaolin; Wang, Hong; Yu, Huilong; Hu, Xiaosong

doi:10.1109/tvt.2019.2926733

Cited by 144 publications

(62 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of the agent is to search an optimal sequence of control actions based on feedback from the environment. Owing to its characteristics of self-evaluation and self-promotion, RL is widely used in many research fields [28][29][30][31][32].…”

Section: A Rl Conceptmentioning

confidence: 99%

Decision-Making Strategy on Highway for Autonomous Vehicles Using Deep Reinforcement Learning

Liao

Tang

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Autonomous driving is a promising technology to reduce traffic accidents and improve driving efficiency. In this work, a deep reinforcement learning (DRL)-enabled decision-making policy is constructed for autonomous vehicles to address the overtaking behaviors on the highway. First, a highway driving environment is founded, wherein the ego vehicle aims to pass through the surrounding vehicles with an efficient and safe maneuver. A hierarchical control framework is presented to control these vehicles, which indicates the upper-level manages the driving decisions, and the lower-level cares about the supervision of vehicle speed and acceleration. Then, the particular DRL method named dueling deep Qnetwork (DDQN) algorithm is applied to derive the highway decision-making strategy. The exhaustive calculative procedures of deep Q-network and DDQN algorithms are discussed and compared. Finally, a series of estimation simulation experiments are conducted to evaluate the effectiveness of the proposed highway decision-making policy. The advantages of the proposed framework in convergence rate and control performance are illuminated. Simulation results reveal that the DDQN-based overtaking policy could accomplish highway driving tasks efficiently and safely. INDEX TERMS Autonomous driving, decision-making, deep reinforcement learning, dueling deep Qnetwork, deep Q-learning, overtaking policy.

show abstract

Section: A Rl Conceptmentioning

confidence: 99%

Decision-Making Strategy on Highway for Autonomous Vehicles Using Deep Reinforcement Learning

Liao

Tang

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…This kind of SOC estimation often has certain limitations. For example, the mean error in the method of [18] can be as high as 1.64%, the maximum error in a recent method of [19] is at least 0.46%, and these errors could exceed 2% under certain driving cycles [18][19][20]. For this reason, it is advised to take the average value of the above SOC range, i.e., 37% as the best SOC value.…”

Section: Case Studymentioning

confidence: 99%

Optimal State-of-Charge Value for Charge-Sustaining Mode of Plug-In Hybrid Electric Vehicles

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

For plug-in hybrid electric vehicles or extended range electric vehicles, battery stored energy often cannot fully meet the travel needs, and the battery needs to stay in a charge-sustaining mode to allow for backup sources such as gasoline or diesel to power the vehicle. It is crucial to identify the optimal stateof-charge (SOC) value for the battery to maintain this charge-sustaining mode since this SOC has significant impact to battery degradation. In existing studies, Just-in-Time control proposes that this SOC should be maintained at 55% but without theoretical justification. With the help of a battery degradation model based on solid electrolyte interphase growth, this paper develops a method to decide the optimal SOC value for charge-sustaining mode. Following the principle of superposition, degradation during the battery discharging process is divided as a fixed degradation caused by the drop of SOC from maximum to minimum values during charge-depleting mode, and a dynamic degradation caused by the oscillation at the charge-sustaining SOC value. Then this oscillation-caused degradation is further modeled and minimized through the investigation of the side reaction current density. The optimal SOC value obtained will be the SOC at which the side reaction current density has the slowest changing rate. This SOC value indeed relies on battery parameters, charging/discharging current and ambient temperature, and a case study shows that the best range of SOC value is 36%~38% for a 1.8Ah SONY 18650 cell. An average SOC of 37% is therefore recommended for charge-sustaining mode considering the possible errors in SOC estimation. INDEX TERMS Plug-in hybrid electric vehicle, state-of-charge, charge-sustaining, solid electrolyte interphase

show abstract

“…In this section, a 7-DOF bus dynamics model is established based on previous literatures [34][35][36], which includes the longitudinal motion, lateral motion, yaw motion and the rotational dynamics of four wheels. According to Newton's theorem, the dynamic equations of the bus longitudinal, lateral and yaw motions can be established as follows [37]:…”

Section: Bus Dynamics Modelmentioning

confidence: 99%

An adaptive nonsingular fast terminal sliding mode control for yaw stability control of bus based on STI tire model

Sun

Wang

Cai

et al. 2020

Preprint

View full text Add to dashboard Cite

In this paper, a novel adaptive nonsingular fast terminal sliding mode (ANFTSM) control scheme for yaw stability control (YSC) is proposed to improve the bus curve driving stability and safety on slippery roads. There are three major contributions in the design process of the bus YSC system. The first contribution is that the STI (Systems Technologies Inc.) tire model, which can effectively reflect the coupling relationship between the tire longitudinal force and lateral force, is established based on experimental data and firstly adopted in the bus YSC system design. The second contribution is a novel YSC strategy based on ANFTSM, which has the merits of fast transient response, finite time convergence and high robustness against uncertainties and external disturbances. The third contribution is that the robust least-squares allocation method is used to solve the optimal allocation problem of the tire forces, whose objective is to achieve the desired direct yaw moment through the effective distribution of the brake force of each tire. To verify the feasibility, effectiveness and practicality of the proposed bus YSC approach, the TruckSim-Simulink co-simulation results are finally provided.

show abstract

Adaptive Hierarchical Energy Management Design for a Plug-In Hybrid Electric Vehicle

Cited by 144 publications

References 35 publications

Decision-Making Strategy on Highway for Autonomous Vehicles Using Deep Reinforcement Learning

Decision-Making Strategy on Highway for Autonomous Vehicles Using Deep Reinforcement Learning

Optimal State-of-Charge Value for Charge-Sustaining Mode of Plug-In Hybrid Electric Vehicles

An adaptive nonsingular fast terminal sliding mode control for yaw stability control of bus based on STI tire model

Contact Info

Product

Resources

About