The invention relates to a nonlinear control method for importing a neural network in the field of artificial intelligence. The nonlinear control method is characterized in that a person does not think that an automobile is designed by a person with biological consciousness when driving the automobile. Although non-linear adjustments are required for the operation of the steering wheel, accelerator, and brake, yet would not feel anomalies. External information is sensed from eyes of people, and hands and feet are freely commanded to operate a steering wheel or an accelerator through biological consciousness, so that a nonlinear closed-loop biological control system is formed. The problem cannot be solved by the modern control theory. And according to the application experience of the combinatorial theory, the problem belongs to the NP-hard problem when more than 40 combined elements are included. Therefore, for complex road conditions, it can be considered that an automatic driving automobile also has an NP-hard problem in control. The reason is that an automatic driving automobile cannot reach the L4 level at present. The invention provides a construction method of an autonomous learning neural network capable of simulating human driving, and the method can solve the nonlinear and random control problem required by automatic driving of an automobile. And the automatic driving automobile can be in an optimal control state.

    本发明涉及人工智能领域中的一种导入神经网络的非线性控制方法,其特征在于:人类在驾驶汽车的时候没有想到汽车是由具有生物意识的人设计的。尽管方向盘,油门,以及刹车的操作需要非线性的调节,但是谁也没有感觉到异常。因为从人眼睛那里感知到外界信息,同时通过生物意识自如的指挥手和脚操作方向盘或油门,形成一个非线性的闭环生物控制系统。这是现代控制理论所解决不了的。再有根据组合理论的应用经验,超过40个以上组合要素时就属于NP‑hard问题。因此面对复杂的路况可以认为自动驾驶汽车也是存在控制上的NP‑hard问题。这就是目前自动驾驶汽车不可能达到L4级别的理由。本专利申请提出一个可以模仿人的驾驶的自律学习神经网络的构成方法,可以解决自动驾驶汽车所需的非线性,随机性的控制问题。可以使自动驾驶汽车处于最佳的控制状态。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Nonlinear control method of neural network imported with self-discipline machine learning model


    Weitere Titelangaben:

    一种导入自律机器学习模型的神经网络的非线性控制方法


    Beteiligte:
    GU ZECANG (Autor:in)

    Erscheinungsdatum :

    27.12.2022


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion



    Imported machine tools

    Engineering Index Backfile | 1931


    Self-discipline running mine vehicle

    MACHII KIMIYOSHI / UOTSU SHINICHI | Europäisches Patentamt | 2023

    Freier Zugriff

    Nonlinear Internal Model Control Based on Extreme Learning Machine

    Huang, Y. | British Library Online Contents | 2013


    Handling imported ore

    Atwood, G.H. | Engineering Index Backfile | 1951