Optimization method, device and system of control system, vehicle, medium and equipment

Abstract

The invention relates to a control system optimization method, device and system, a vehicle, a medium and equipment. The optimization method comprises the steps of: obtaining optimization related parameters of a to-be-optimized control system in response to a received control system optimization request; determining a compensation model of the to-be-optimized control system according to the optimization related parameters; and according to the compensation model and the original control model of the to-be-optimized control system, constructing an optimal control algorithm of the to-be-optimized control system, so that the to-be-optimized control system performs control according to the optimal control algorithm. Therefore, the purposes of optimizing the control algorithm of the control system and improving the control margin of the control system can be achieved.

Description

technical field [0001] The present disclosure relates to the technical field of automatic control, in particular, to an optimization method, device, system, vehicle, medium and equipment of a control system. Background technique [0002] At present, the control algorithms used in most automatic driving solutions are mainly PID, LQR and MPC, and their development process includes algorithm design, code implementation, simulation testing and algorithm performance optimization. However, in actual engineering, it is difficult to control the accurate object model for the control system. Therefore, the performance optimization scheme of the current algorithm mainly uses real vehicle debugging to complete the optimization, which leads to long time-consuming optimization of the real vehicle, and it is often easy to judge the optimization effect based on intuition. There are no issues with optimal performance. [0003] The design of control algorithms in academia often uses simulati...

AI Technical Summary

Problems solved by technology

However, in actual engineering, it is difficult to control the accurate object model for the control system. Therefore, the performance optimization scheme of the current algorithm mainly uses real vehicle debugging to complete the optimization, which leads to long time-consuming optimization of the real vehicle, and it is often easy to judge the optimization effect based on intuition. Problems with suboptimal performance

[0003]Academic control algorithm design mostly uses simulation environment to rely on accurate models for algorithm construction, and realizes the optimization of algorithm design architecture through simulation, but because the vehicle is an extremely complex system, the simulation environment is difficult to simulate the real driving environment and vehicle dynamics, therefore, it is difficult to apply the method of optimizing the algorithm design architecture through simulation to the vehicle, especially the automatic driving scene of the vehicle

[0004]Usually, when designing a control algorithm, first set the algorithm parameters according to the requirements, and then repeatedly optimize the algorithm parameters in the actual test, while in the process of using the control algorithm Due to the delay in the control system or the strong nonlinearity of the control object, the control margin is insufficient and the coverage of the control system is small. At this time, the algorithm parameters in the control algorithm are not optimal.

The method of adjusting algorithm parameters in related technologies is easy to fall into problems such as repeated parameter adjustment and empirical judgment

Images