阅读说明：本技术 用于复杂路况的主动悬架控制系统和控制方法 (Active suspension control system and control method for complex road conditions ) 是由 李玉芳 卢小丁 倪铭 徐国放 于 2019-08-21 设计创作，主要内容包括：本发明公开一种用于复杂路况的主动悬架控制系统和控制方法,其中控制系统包括环境感知模块,包括速度传感器、红外结构光组件和单目摄像头；实时处理控制模块,包括BP神经网络,其接收来自速度传感器和红外结构光组件的数据,输出适合当前路况的阻尼和刚度值,再结合卷积神经网络输出的数据,得到修正的阻尼和刚度值；调节模块,用于将修正的阻尼和刚度值传送到主动悬架,并对主动悬架的阻尼和刚度进行调整。本发明在保证实时性和精度的前提下,降低了路面信息获取的成本,提升了对路面的适应能力。(The invention discloses an active suspension control system and a control method for complex road conditions, wherein the control system comprises an environment sensing module, a speed sensor, an infrared structure light assembly and a monocular camera; the real-time processing control module comprises a BP neural network, a speed sensor and an infrared structure optical assembly, wherein the BP neural network receives data from the speed sensor and the infrared structure optical assembly, outputs a damping and rigidity value suitable for the current road condition, and combines the data output by the convolutional neural network to obtain a corrected damping and rigidity value; and the adjusting module is used for transmitting the corrected damping and rigidity values to the active suspension and adjusting the damping and rigidity of the active suspension. The invention reduces the cost of acquiring the road surface information and improves the adaptability to the road surface on the premise of ensuring the real-time property and the precision.)

1. An active suspension control system for complex road conditions, comprising

The environment sensing module comprises a speed sensor, an infrared structured light assembly and a monocular camera, wherein the speed sensor is used for measuring the current speed of the vehicle, the infrared structured light assembly is used for three-dimensional identification of the road surface, and the monocular camera is positioned above the front part of the vehicle and used for collecting road surface images and then sending the road surface images to the convolutional neural network to judge the hardness and softness of the road surface;

the real-time processing control module comprises a BP neural network, a speed sensor and an infrared structural optical assembly, wherein the BP neural network receives data from the speed sensor and the infrared structural optical assembly, outputs a damping and rigidity value suitable for the current road condition, and combines the data output by the convolutional neural network to obtain a corrected damping and rigidity value;

and the adjusting module is used for transmitting the corrected damping and rigidity values to the active suspension and adjusting the damping and rigidity of the active suspension.

2. The active suspension control system of claim 1 wherein the speed sensor is a wheel speed sensor or an OBD interface.

3. The active suspension control system of claim 1 or 2 wherein the environmental awareness module further comprises a vehicle distance meter for sensing distance to an adjacent vehicle, and wherein the time threshold for changing the damping and stiffness values is set based on the current vehicle speed.

4. An active suspension control method for complex road conditions, comprising the steps of:

acquiring current environment and road condition information including a road surface three-dimensional structure and a road surface image;

extracting the height value of the road surface in front of the vehicle and the current speed, and inputting the height value and the current speed into a BP neural network to obtain the damping and rigidity values suitable for the current road condition;

processing the road surface image through a convolutional neural network, identifying the road surface type, and evaluating to obtain a road surface hardness degree parameter;

correcting the damping and rigidity values by using the road surface soft and hard degree parameters to obtain corrected damping and rigidity values;

and transmitting the corrected damping and stiffness values to an active suspension, and adjusting the damping and stiffness of the active suspension.

5. The active suspension control method of claim 4 wherein the identifying the road surface type is accomplished by convolutional neural network interval sampling.

6. The active suspension control method according to claim 4, wherein the correction of the damping and stiffness value using the road surface softness and stiffness parameter is performed by multiplying the damping and stiffness value by the road surface softness and stiffness parameter, and the road surface softness and stiffness parameter is in a range of 0.5 to 1.5.

Technical Field

The invention relates to the technical field of automobile control, in particular to an active suspension control system and method for complex road conditions.

Background

The smoothness of the automobile is an important index for measuring the performance of the automobile, and is the most intuitive experience of a driver of the automobile. Poor ride comfort then influences the working efficiency of people and car, leads to passenger fatigue, goods wearing and tearing, and the whole car spare part is premature failure and produces great noise in the car.

The suspension is of great importance to the smoothness, and the active suspension has better smoothness due to the characteristics of adjustable rigidity and damping. However, most of the existing methods for controlling damping and stiffness of active suspensions are to adjust the damping of the active suspension by inputting a road load into a preset adjustment system according to a vertical load from a road received by a wheel. The adjusting method has a limited adjusting range and cannot cope with the road surface with great mutation.

Secondly, the active suspension is adjusted by taking the output of the laser radar as road surface information, but the laser radar is high in overall price and is not suitable for being popularized in vehicles at all levels, and the real-time performance and the measurement precision of computer vision are difficult to be considered at the same time.

The structured light depth detection utilizes the influence of the surface depth information of the target object on the structured light projection pattern to obtain the depth information of the target object, and the depth information of the road surface can be obtained by analyzing a plurality of structured light coding modes, so that the precision and the real-time performance are good.

In the traditional control method based on wheel load, the response speed and accuracy are difficult to ensure in the processing of road surface input. The road surface sudden change such as deceleration strip can not be treated well.

In the existing control method, a road surface is considered as a rigid body in a modeling process, and the influence of the load deformation degree caused by the soft road surface on the vibration parameters of the suspension is not considered.

Disclosure of Invention

The invention aims to provide an active suspension control system and a control method for complex road conditions, which consider the influence of the hardness degree of a road surface on suspension parameters, improve the adaptability to an abrupt change road surface and reduce the system cost on the premise of ensuring quick response and high precision.

In order to achieve the above object, an aspect of the present invention provides an active suspension control system for complex road conditions, comprising: the environment sensing module comprises a speed sensor, an infrared structured light assembly and a monocular camera, wherein the speed sensor is used for measuring the current speed of the vehicle, the infrared structured light assembly is used for three-dimensional identification of the road surface, and the monocular camera is positioned above the front part of the vehicle and used for collecting road surface images and then sending the road surface images to the convolutional neural network to judge the hardness and softness of the road surface; the real-time processing control module comprises a BP neural network, a speed sensor and an infrared structural optical assembly, wherein the BP neural network receives data from the speed sensor and the infrared structural optical assembly, outputs a damping and rigidity value suitable for the current road condition, and combines the data output by the convolutional neural network to obtain a corrected damping and rigidity value; and the adjusting module is used for transmitting the corrected damping and rigidity values to the active suspension and adjusting the damping and rigidity of the active suspension.

Further, the speed sensor is a wheel speed sensor or an OBD interface.

Furthermore, the environment sensing module further comprises a vehicle distance measuring instrument which is used for sensing the distance between the environment sensing module and an adjacent vehicle and setting a time threshold value for changing the damping and rigidity values according to the current vehicle speed.

The invention also provides an active suspension control method for complex road conditions, which comprises the following steps:

acquiring current environment and road condition information including a road surface three-dimensional structure and a road surface image; extracting the height value of the road surface in front of the vehicle and the current speed, and inputting the height value and the current speed into a BP neural network to obtain the damping and rigidity values suitable for the current road condition; processing the road surface image through a convolutional neural network, identifying the road surface type, and evaluating to obtain a road surface hardness degree parameter; correcting the damping and rigidity values by using the road surface soft and hard degree parameters to obtain corrected damping and rigidity values; and transmitting the corrected damping and stiffness values to an active suspension, and adjusting the damping and stiffness of the active suspension.

Further, the identification of the pavement type is completed by sampling the convolutional neural network at intervals.

Further, the damping and rigidity value is corrected by using the road surface soft and hard degree parameter, the damping and rigidity value is multiplied by the road surface soft and hard degree parameter, and the range of the road surface soft and hard degree parameter is 0.5-1.5.

According to the invention, the front road information is collected through the infrared structured light assembly and the monocular camera, the optimal damping and rigidity of the vehicle about to pass through the road active suspension are calculated through the BP neural network and the convolution neural network model, and the adjustment is carried out after the correction is carried out by combining the road hardness degree. Compared with the prior art, the method reduces the cost for acquiring the road surface information and improves the adaptability to the road surface on the premise of ensuring the timeliness and the precision.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the invention and not to limit the invention. In the drawings, there is shown in the drawings,

FIG. 1 is a block diagram of an active suspension control system for complex road conditions according to an embodiment of the present invention;

FIG. 2 is a flowchart of an active suspension control method for complex road conditions according to another embodiment of the present invention;

FIG. 3 is a schematic diagram of the operation of a monocular camera and an infrared structured light assembly in the embodiment of FIG. 2;

FIG. 4 is a diagram of a convolutional neural network architecture;

fig. 5 is a simplified seven-degree-of-freedom model diagram of a suspension.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.