Joint module inequality constraint optimal robust control method based on reinforcement learning

Abstract

The invention relates to a joint module inequality constraint optimal robust control method based on reinforcement learning, and the method comprises the steps: building a joint module virtual simulation environment on a simulation platform, and constructing a reinforcement learning neural network model; initializing a virtual simulation environment of the joint module; adjusting connection weights among multiple layers of neurons in the reinforcement learning neural network model; randomly inactivating a part of neurons of the reinforcement learning neural network model, and outputting control parameter information of the joint module; collecting a training data set of the current joint module in a virtual simulation environment; and inputting an optimal parameter obtained by reinforcement learning neural network training into an inequality constraint optimal robust controller, constraining a motor moving trajectory within a specified range, and remarkably improving the control precision of the motor moving trajectory. According to the method, the particle swarm optimization algorithm is adopted to adjust the connection weight among multiple layers of neurons in the reinforcement learning neural network model, the motor control precision can be remarkably improved, and the moving trajectory of the motor can be restrained within a specified range.

Description

technical field [0001] The invention relates to the technical field of robot control, in particular to an optimal robust control method based on reinforcement learning for joint module inequality constraints. Background technique [0002] Joint modules have been widely used in small and medium-sized electric drive fields, such as aerospace, robotics, electric vehicles and other fields. The high-performance control about it is a multivariable, highly coupled and time-varying nonlinear system, and the traditional control method needs to obtain the precise system parameters of the motor. However, structural uncertainties, such as system parameter changes, insufficient system modeling, and non-structural uncertainties, such as load torque disturbances, control target diversity, etc., may affect the performance of joint modules, especially permanent magnet synchronous motor systems. Servo performance. Improving the robustness and dynamic performance of joint modules is an effec...

AI Technical Summary

Problems solved by technology

[0003] The existing control methods for joint modules mainly include the following two aspects: first, after the controller is built, the parameters need to be adjusted blindly from scratch, without knowing the optimal threshold of motor control; second, the existing The robust control method cannot well control the joint module within a specified range, and the accuracy is not enough, which may cause the joint module to run off the track, which may lead to some major accidents in real production and life

Images