Tracked vehicle slip rate estimation method based on pavement classification and machine learning

Abstract

The invention relates to a tracked vehicle slip rate estimation method based on pavement classification and machine learning, and belongs to the technical field of slip rate estimation. Based on pavement classification and segmented support vector regression, a crawler-type platform is used for collecting a large amount of data, a pavement classification model is established, a slip rate discrete estimation model is established, a slip rate regression model is established, driving wheel torque and rotating speed collected by the crawler-type platform, vehicle transverse and longitudinal acceleration and vehicle yaw velocity and pitch angular velocity are used as input characteristics of the slip rate discrete estimation model and the slip rate regression model, and finally an estimated value of the slip rate of the crawler-type vehicle is obtained. The method is suitable for tracked vehicles, the pavement types are accurately distinguished through the pavement classification model, the influence of the pavement types on slip rate estimation is eliminated, and the slip rate estimation value is further quickly and accurately obtained through the slip rate discrete estimation model and the slip rate regression model; and the method can be applied to tracked vehicle model correction and traction force control.

Description

technical field [0001] The patent of the invention relates to a method for estimating the slippage rate of tracked vehicles based on road surface classification and machine learning, in particular to a method for estimating the slippage rate of tracked vehicles based on road surface classification and segmental support vector regression, which belongs to the field of vehicle slippage rate estimation . Background technique [0002] Compared with wheeled vehicles, tracked vehicles have stronger steering ability and passability, making them play an important role in both military and civilian fields. During the running of a tracked vehicle, the track needs to perform compound motion: on the one hand, it needs to move along with the vehicle body in a straight line relative to the ground. q It means that, on the other hand, the crawler has to do a rotational movement relative to the car body, and this movement speed is called the relative speed, expressed as v x Indicates that ...

AI Technical Summary

Problems solved by technology

[0005] The wheel speed in the direct estimation method is generally measured by an encoder, and the vehicle speed can be measured by non-driving wheel speed, visual odometer, GPS sensor, and inertial measurement unit, but these sensors have some limitations in speed measurement: non-driving wheel speed is not Applicable to tracked vehicles; the visual odometer has a large amount of data and is greatly affected by light, so it is not applicable in an environment with few features; the global satellite positioning system has a large error in an open environment; the inertial measurement unit has cumulative errors

However, this method cannot be applied to tracked vehicles at present because of the following reasons: first, it only considers low-speed and straight-going conditions, and does not consider high-speed and turning conditions; second, it only estimates single-wheel slip, and does not consider multiple differences in wheel slip; finally, it is currently only for wheeled vehicles, whether it is applicable to tracked vehicles remains to be seen

In addition, it does not consider the influence of road type on the slip rate. When the vehicle state is the same, different road types will produce different degrees of slip. Therefore, it is necessary to consider the influence of road type when estimating the slip rate. Except In addition, the current slip rate estimation method based on machine learning has low accuracy

Images