ABSTRACT
In train marshalling, the mass of the freight train will dynamically change in a wide range, which is the main difficulty in realizing its automatic driving. This paper proposes a deep reinforcement learning method that combines domain knowledge and mass estimation network (MEN). The domain knowledge of excellent drivers is utilized to accelerate the convergence speed of the algorithm and improve the driving performance. Furthermore, the MEN is introduced for estimating the mass of the entire train during driving. Finally, the deep reinforcement learning algorithm selects the output gear based on the estimated mass. The simulation results show that the proposed method has significant effects on performance optimization such as reducing parking error, improving marshalling efficiency, optimizing coupler force and reducing jerk.
- Yin, J., Tang, T., Yang, L., Xun, J., Huang, Y. and Gao, Z. Research and development of automatic train operation for railway transportation systems: A survey. Transportation Research Part C: Emerging Technologies, 85 (2017), 548-572.Google ScholarCross Ref
- Larranaga, M., Anselmi, J., Ayesta, U., Jacko, P. and Romo, A. Optimization techniques applied to railway systems (2013).Google Scholar
- Tuyttens, D., Fei, H., Mezmaz, M. and Jalwan, J. Simulation-based genetic algorithm towards an energy-efficient railway traffic control. Mathematical Problems in Engineering, 2013 (2013).Google Scholar
- Miyatake, M. and Ko, H. Optimization of train speed profile for minimum energy consumption. IEEJ Transactions on Electrical and Electronic Engineering, 5, 3 (2010), 263-269.Google ScholarCross Ref
- Yin, J., Chen, D. and Li, L. Intelligent Train Operation Algorithms for Subway by Expert System and Reinforcement Learning. IEEE Transactions on Intelligent Transportation Systems, 15, 6 (2014), 2561-2571.Google ScholarCross Ref
- Xi Wang and Tao Tang Optimal control of heavy haul train on steep downward slope (2016).Google Scholar
- Cui, Y., Zhang, G., Dong, W., Sun, X. and Yang, W. Knowledge-based Deep Reinforcement Learning for Train Automatic Stop Control of High-Speed Railway. City, 2020.Google Scholar
- Yang, X., Li, X., Ning, B. and Tang, T. A survey on energy-efficient train operation for urban rail transit. IEEE Transactions on Intelligent Transportation Systems, 17, 1 (2015), 2-13.Google Scholar
- Liu, R. R. and Golovitcher, I. M. Energy-efficient operation of rail vehicles. Transportation Research Part A, 37, 10 (2003), 917-932.Google ScholarCross Ref
- Li, L., Dong, W., Ji, Y., Zhang, Z. and Tong, L. Minimal-energy driving strategy for high-speed electric train with hybrid system model. IEEE Transactions on Intelligent Transportation Systems, 14, 4 (2013), 1642-1653.Google ScholarDigital Library
- Chen, D., Chen, R., Li, Y. and Tang, T. Online learning algorithms for train automatic stop control using precise location data of balises. IEEE Transactions on Intelligent Transportation Systems, 14, 3 (2013), 1526-1535.Google ScholarDigital Library
- Dong, H., Ning, B., Cai, B. and Hou, Z. Automatic train control system development and simulation for high-speed railways. IEEE circuits and systems magazine, 10, 2 (2010), 6-18.Google ScholarDigital Library
- Howlett, P. Optimal strategies for the control of a train. Automatica, 32, 4 (1996), 519-532.Google ScholarDigital Library
- Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K. and Ostrovski, G. Human-level control through deep reinforcement learning. Nature, 518, 7540 (2015), 529.Google Scholar
- Lillicrap, T. P., Hunt, J. J., Pritzel, A., Heess, N., Erez, T., Tassa, Y., Silver, D. and Wierstra, D. Continuous control with deep reinforcement learning. arXiv preprint arXiv:1509.02971 (2015).Google Scholar
Recommendations
Simulation and Analysis of Derailment of Freight Train on Bridges
ICCMS '10: Proceedings of the 2010 Second International Conference on Computer Modeling and Simulation - Volume 03Taking into account the connecting conditions of wheel-rail displacement and the influence of clearance between wheel flange and gauge line, the calculation model of the spatial vibration of train bridge system is established. The whole course of train ...
Freight Telematics Systems: An Intelligent Wagon
Communication Technologies for VehiclesAbstractIn order to reach the EU’s objectives with regards to developing rail freight in the future, an evolution of the freight train is required. For that, telematics technologies play a crucial role as an enabler for the intelligent freight train. This ...
Deep Q-learning with Explainable and Transferable Domain Rules
Intelligent Computing Theories and ApplicationAbstractRecent research on deep reinforcement learning (RL) has shown its capability in automating sequential decision-making and control tasks. However, RL agents require a large number of interactions with the environment and are lacking in ...
Comments