Intelligent automobile transverse control method based on sensor measurement reconstruction under network attack

Abstract

The invention discloses an intelligent automobile transverse control method based on sensor measurement reconstruction under network attacks, and relates to automobile intelligent safety and automatic driving. A model prediction control MPC controller based on sensor measurement reconstruction is constructed, and accurate control over transverse movement of the vehicle is achieved. The system monitors sensor data in real time, detects sensor data abnormity caused by network attack, starts a measurement reconstruction mechanism, and corrects the data by using a redundant sensor or an algorithm model. A lower-layer torque optimal distribution controller is designed, the controller intelligently adjusts output torques of four hub motors according to expected yaw moment output by an upper-layer MPC and vehicle state information, so that accurate control over the yaw moment of the vehicle is achieved, transverse tracking errors are reduced, and the driving stability and the dynamic performance of the vehicle are improved. A transverse cooperative control rule is provided, and comprehensive optimization of transverse control of the vehicle is achieved by coordinating work of an upper-layer MPC controller and work of a lower-layer torque optimization distribution controller.

Images