PID locomotive automatic driving optimal control method based on reinforcement learning

Abstract

The invention provides a PID locomotive automatic driving optimal control method based on reinforcement learning. The PID locomotive automatic driving optimal control method comprises the steps of: regarding current route information and a speed difference between a locomotive actual running speed and an optimal speed as input of reinforcement PID at first, acquiring a group of optimal PID controlparameters through reinforcement learning, realizing PID control according to the optimal PID control parameters provided by a reinforcement learning module, providing a control quantity, and controlling operation of a locomotive. The PID locomotive automatic driving optimal control method applies reinforcement learning to PID parameter adjustment, can interact with the environment by means of reinforcement learning, has the self-learning capability, can realize the PID control having the optimal parameter group, improves the optimization effect, and reduces manual design difficulty.

Description

technical field [0001] The invention relates to locomotive operation optimization manipulation technology, in particular to a PID (proportional, integral, differential) locomotive automatic driving optimization control method based on reinforcement learning based on reinforcement. Background technique [0002] The optimal control of locomotive automatic driving is one of the core functions of the locomotive automatic control system. Its main function is to calculate the corresponding control input quantity based on the optimal speed curve trajectory, based on the actual motion state and combined with the following control algorithm, and convert the control input quantity It acts on the actual operation process of the locomotive to complete the speed control. The locomotive automatic driving system needs to achieve indicators such as punctuality, comfort and energy-saving operation. The optimal generation of the optimal speed curve is the guarantee for the locomotive automat...

AI Technical Summary

Problems solved by technology

The classic control algorithm and the improved PID control algorithm are easy to implement and have good robustness, but there are difficulties in parameter tuning; intelligent control algorithms also have their own characteristics, such as fuzzy control, which is difficult to establish an accurate model but can be controlled according to experience. The control effect is better , but the design of fuzzy rules is too dependent on manual design and cannot be applied to systems with large delays; expert systems can make better use of expert experience and knowledge, but there are deficiencies in knowledge acquisition relying on manual work and weak reasoning ability

Because various intelligent control algorithms have their own characteristics and have the possibility of complementarity, the integrated intelligent control algorithm is combined according to the advantages of different intelligent control algorithms, but it still cannot completely avoid the shortcomings of the combined intelligent control algorithm itself.

Such as fuzzy predictive control, although the control effect is improved, but the fuzzy rules still need to be manually designed

[0004] Although there have been a lot of theoretical research results on the control algorithm, most of the locomotive control algorithms currently in use still use the PID control algorithm, but the limitations of the PID control algorithm itself determine that the control performance cannot be optimal.

In addition, due to the complex and changeable characteristics of the locomotive operating environment, it will also affect the performance of PID control

Images