An intelligent train control approach based on the monte carlo reinforcement learning algorithm

Li, Wentao; Tang, Tao; Su, Shuai; Cao, Yuan; Bao, Feng; Gao, Jianguo

doi:10.1109/itsc.2018.8569399

Cited by 10 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remark 1. Generally, the matrix C can be obtained using the solution to (17). However, if the system measured state is included in the lifting function as shown in ( 10), the matrix C can be directly defined by oneself.…”

Section: B Koopman Model Of High-speed Trainmentioning

confidence: 99%

“…In [16], deep reinforcement learning was adopted to achieve an optimal control policy in the communication-based train control system, where the speed tracking and the energy consumption were optimized. The Monte Carlo reinforcement learning method was proposed in [17] to select the optimal solution for the train control decision. Based on integral reinforcement learning and parameter identification methods, an adaptive control scheme was proposed in [18] to achieve the tracking control of high-speed trains.…”

Section: Introductionmentioning

confidence: 99%

“…Using the Gauss basis function defined in(36), the corresponding X lift and X + lift of size 12 × 2 • 10 6 can be obtained by taking the time 0.16 s averagely. For such a big data set, it only takes an average of 0.07 s to solve(17) and obtains the matrices A, B and C of the Koopman model. Considering the short time to build the Koopman model, it is feasible to update the Koopman model online.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Data-Driven Koopman Model Predictive Control for Optimal Operation of High-Speed Trains

et al. 2021

View full text Add to dashboard Cite

Automatic train operation systems of high-speed trains are critical to guarantee operational safety, comfort, and parking accuracy. However, implementing optimal automatic operation control is challenging due to the train's uncertain dynamics and actuator saturation. To address this issue, this paper develops a data-driven Koopman model based predictive control method for automatic train operation systems. The proposed control scheme is designed within a data-driven framework. First, using operational data of trains and the Koopman operator, an explicit linear Koopman model is built to characterize the train dynamics. Then, a model predictive controller is designed based on the Koopman model under comfort and actuator constraints. Furthermore, an online update mechanism for the Koopman model is developed to cope with the changing dynamic characteristics of trains, which reduces the accumulation errors and improves control performance. Stability analysis of the closed-loop control system is provided. Comparative simulation results validate the effectiveness of the proposed control approach.

show abstract

Section: B Koopman Model Of High-speed Trainmentioning

confidence: 99%