Bridgehead bumping rapid detection method based on one-dimensional convolution kernel feature extraction

Abstract

The invention discloses a bridgehead bumping rapid detection method based on one-dimensional convolution kernel feature extraction. The method comprises the following steps: acquiring vehicle motion state data by using a vehicle-mounted attitude sensor, extracting a feature vector from the acquired vehicle motion state data through one-dimensional convolution, and training the feature vector by using a full-connection neural network to obtain a one-dimensional convolutional neural network model; and based on the neural network model, inputting the motion state data of the vehicle passing through the to-be-detected road section into the model, and detecting whether the road section has a bridgehead bump condition and the severity of bridgehead bump.

Description

technical field [0001] The invention relates to the technical field of engineering detection, and more specifically relates to a rapid detection method for bridge head jumping based on one-dimensional convolution kernel feature extraction. Background technique [0002] Vehicle jumping at the bridge head is one of the common road diseases at present. The phenomenon of vehicle jumping at the bridge head refers to the obvious staggered platform at the connection between the bridge head and the road surface structure, which causes the vehicle to jump when passing the bridge head. When the vehicle passes the bridge head jump section, it will produce obvious vibration, which will make the occupants feel obvious discomfort. [0003] Vehicle jumping at the bridge head seriously affects driving safety and comfort. In the section where the vehicle jumps at the bridge head, the driving vehicle has to reduce the speed, thereby reducing the traffic capacity of the highway. Vehicle jum...

AI Technical Summary

Problems solved by technology

The mainstream detection method still stays at manual detection, with low efficiency and insufficient security

In recent years, the measurement methods such as level instrument and turbulence accumulator all have problems such as inconvenient operation and high cost.

Images