Rollover early warning and control method based on quadratic prediction type zero moment point

Abstract

The invention discloses a rollover early warning and control method based on a quadratic prediction type zero moment point, and the method comprises the following steps: building a vehicle three-degree-of-freedom rollover dynamic model which comprises transverse motion, yawing motion and roll motion; establishing a rollover early warning index zero moment point by combining the established vehicle three-degree-of-freedom model, and calculating primary and secondary change rates of the transverse position of the zero moment point; based on the secondary change rate of the transverse position of the zero moment point, an analytical solution of the rollover early warning time is solved; and when the current rollover early warning time is smaller than the upper limit value, vehicle rollover early warning is conducted, and rollover prevention brake pressure control is conducted through a front wheel differential brake module based on fuzzy PI control. According to the method, a driver can be effectively warned of the rollover risk of the vehicle, and meanwhile, a good rollover prevention control effect can be achieved by applying fuzzy self-adaptive control.

Description

technical field [0001] The invention relates to the technical field of automobile safety, in particular to a rollover early warning and control method based on a quadratic predictive zero moment point. Background technique [0002] Vehicle rollover is a very serious traffic accident. In automobile accidents, the degree of harm of rollover is second only to collision accidents, ranking second. Rollover accidents can cause serious casualties and property damage. Timely and accurate prediction of the driving state of the vehicle, and timely transmission of the real-time state parameters of the vehicle to the driver and the active anti-rollover control system can prevent the vehicle from rolling over to a certain extent. [0003] The rollover early warning index is an indicator that characterizes the rollover state of the vehicle. The traditional rollover index uses the lateral load transfer rate, roll angle, lateral acceleration, etc. However, the traditional early warning ind...

AI Technical Summary

Problems solved by technology

[0003] The rollover early warning index is an indicator that characterizes the rollover state of the vehicle. The traditional rollover index uses the lateral load transfer rate, roll angle, lateral acceleration, etc. However, the traditional early warning index can only represent the rollover state of the vehicle at the current moment. It is impossible to characterize the future rollover state of the vehicle in a timely manner. For example, Jin Liqiang et al. used in "Vehicle Rollover Early Warning Based on Zero Moment Point Index and Rollover Time Algorithm [J]. Automotive Engineering. 2017.39(3).281-287" The zero-moment point position is used as a rollover warning indicator to characterize the vehicle state, and the rollover warning time is calculated based on the zero-moment point. However, since the zero-moment point can only represent the current vehicle state, the rollover warning time calculated by using it cannot Forecasting at any moment in the future leads to time-lag in vehicle early warning, and the accuracy of early warning is not high, so that the active anti-rollover control system cannot control the vehicle in time

However, the traditional active anti-rollover control system uses a simple PID controller, and the parameter adjustment is cumbersome and poor in applicability, resulting in poor control effect

Images