CN-117755311-B - Road bump type identification method, vehicle control method and related device

CN117755311BCN 117755311 BCN117755311 BCN 117755311BCN-117755311-B

Abstract

The application discloses a method for identifying the bump type of a road, a vehicle control method and a related device, wherein the method comprises the steps of acquiring real-time running information in the running process of a vehicle; the method comprises the steps of determining whether a vehicle is in a bumpy state or not according to running information, determining that the vehicle runs on a bumpy road when the vehicle is in the bumpy state, triggering and collecting a preset data set when the vehicle is on the bumpy road, wherein the preset data set comprises road surface image information and vehicle running information, respectively identifying the bumpy type of a road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the collected preset data set, and fusing identification results of the image identification algorithm, the neural network identification algorithm and the dynamic time planning algorithm for respectively identifying the bumpy type of the road to take the fused result as the bumpy type of the road. The application can efficiently and accurately identify the bump type of the road.

Inventors

WANG LIANXU
REN JINGYING
YANG YIXIN
LI YANG
GU ZHENGMIN
ZHANG FEIFEI
XIAO BAIHONG
SHEN XINYI

Assignees

蔚来汽车科技(安徽)有限公司

Dates

Publication Date: 20260505
Application Date: 20231220

Claims (9)

1. A method of identifying a jolt type of a road, the method comprising: acquiring real-time running information of a vehicle in running; determining whether the vehicle is in a bumpy state according to the driving information; when the vehicle is in a bumpy state, determining that the vehicle is running on a bumpy road surface; Triggering and collecting a preset data set when the vehicle is on a bumpy road surface, wherein the preset data set comprises road surface image information and vehicle driving information; Respectively identifying the bump type of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the acquired preset data set; Fusing the recognition results of the image recognition algorithm, the neural network recognition algorithm and the dynamic time planning algorithm for recognizing the road bump type respectively, and taking the fused result as the road bump type; the method for identifying the bump type of the road by using an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the collected preset data set comprises the following steps: And respectively identifying the confidence degrees of various possible bump types of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the acquired preset data set.
2. The method according to claim 1, wherein the fusing the recognition results of the image recognition algorithm, the neural network recognition algorithm, and the dynamic time planning algorithm for recognizing the bump type of the road, respectively, takes the fused result as the bump type of the road, includes: Taking the bump type with the largest matching degree after fusion as the bump type of the road, wherein the matching degree corresponding to any bump type i is Mi, M i =α i A i +β i B i +γ i C i Wherein alpha i is the confidence weight of the image recognition algorithm on the bump type i, A i is the recognition matching degree of the image recognition algorithm on the bump type i, beta i is the confidence weight of the neural network recognition algorithm on the bump type i, B i is the recognition matching degree of the neural network recognition algorithm on the bump type i, gamma i is the confidence weight of the dynamic time planning algorithm on the bump type i, C i is the recognition matching degree of the dynamic time planning algorithm on the bump type i, Wherein a 1 is an illumination condition influence factor of the image recognition algorithm, a 1 ≤1;a 2 is an image recognition algorithm visual output definition influence factor, a 2 ≤1;W i1 with a value range of 0 is an initial confidence weight of the image recognition algorithm to the bump type i, b 1 is a data quality influence factor input by the neural network recognition algorithm, b 1 ≤1;b 2 is a label data quality influence factor of the neural network recognition algorithm, b 2 ≤1;W i2 with a value range of 0 is an initial confidence weight of the neural network recognition algorithm to the bump type i, c is a dynamic time planning algorithm standard bump data quality influence factor, and c is an initial confidence weight of 0 is less than or equal to 1;W i3 to the bump type i.
3. The method as recited in claim 1, further comprising: Acquiring position information; and uploading the position information and the bump type of the road to a server.
4. A method according to any one of claims 1 to 3, wherein the type of jounce of the road comprises one or more of the group consisting of upper bridge, lower bridge, pit, small heave, medium heave, large heave and continuous jounce.
5. A vehicle control method characterized by comprising: identifying the bump type of the road by the method of any one of claims 1 to 4; And controlling the suspension to enable the state of the suspension to be matched with the identification result, so that the vehicle smoothly passes through the bumpy road section.
6. The method as recited in claim 5, further comprising: Obtaining bump information which is shared by other vehicle owners and is in a preset range of the current position of the vehicle, wherein the bump information comprises position information and bump types; predicting the bump type of a road section to be passed according to the real-time running information of the vehicle and the bump information; Controlling the suspension to match the bump type of the predicted upcoming road segment.
7. A device for identifying the type of jolt on a road, said device comprising: the acquisition unit is used for acquiring real-time running information in running of the vehicle; the device comprises a first determining unit, a second determining unit and a third determining unit, wherein the first determining unit is used for determining whether the vehicle is in a bumpy state according to the running information; The acquisition unit is also used for triggering and acquiring a preset data set when the vehicle is on a bumpy road surface, wherein the preset data set comprises road surface image information and vehicle driving information; The identification unit is used for respectively identifying the bump type of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the preset data set acquired by the acquisition unit; The fusion unit is used for fusing the recognition results of the image recognition algorithm, the neural network recognition algorithm and the dynamic time planning algorithm for recognizing the bump type of the road respectively, and taking the fused result as the bump type of the road; The identification unit is further used for respectively identifying the confidence degrees of various possible bump types of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the acquired preset data set.
8. A processing device comprising a memory and a processor, the memory storing a computer program executable by the processor, the processor implementing the method of any one of claims 1 to 4 and/or 5 or 6 when the computer program is executed.
9. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when read and executed, implements the method of any of claims 1 to 4 and/or of claims 5 or 6.

Description

Road bump type identification method, vehicle control method and related device Technical Field The application relates to the technical field of automobiles, in particular to a road bump type identification method, a vehicle control method and a related device. Background The vehicle can experience various pavements in the running process, when the conditions of fluctuation, potholes and the like exist on the road, the vehicle jolt can be caused, the driving comfort can be influenced by jolt, and meanwhile, potential safety hazards are brought to driving. The road is identified, the road section with jolt and the jolt type are determined, the control of the vehicle according to the jolt type of the road is facilitated, and the driving safety and comfort can be improved. At present, the identification of the bump type of the road is mainly completed by manual observation, and the method has the defects of low efficiency, poor reliability and the like. Therefore, how to efficiently and accurately identify the bump type of the road is a technical problem to be solved at present. Disclosure of Invention The application provides a road bump type identification method, a vehicle control method and a related device, which can efficiently and accurately identify the road bump type. In a first aspect, an embodiment of the present application provides a method for identifying a bump type of a road, where the method includes: acquiring real-time running information of a vehicle in running; determining whether the vehicle is in a bumpy state according to the driving information; when the vehicle is in a bumpy state, determining that the vehicle is running on a bumpy road surface; Triggering and collecting a preset data set when the vehicle is on a bumpy road surface, wherein the preset data set comprises road surface image information and vehicle driving information; respectively identifying the bump type of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning (DYNAMIC TIME WRAPPING, DTW) algorithm according to the acquired preset data set; and merging the recognition results of the image recognition algorithm, the neural network recognition algorithm and the dynamic time planning algorithm for recognizing the bump type of the road respectively, and taking the merged result as the bump type of the road. When the embodiment is adopted, whether the vehicle is in a bumpy state or not is firstly determined according to real-time running information, a preset data set is acquired when the vehicle is in the bumpy state, the bumpy type of a road is respectively identified according to the acquired data set by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm, and finally, identification results of the image identification algorithm, the neural network identification algorithm and the dynamic time planning algorithm for respectively identifying the bumpy type of the road are fused, and the fused result is used as the bumpy type of the road. Compared with manual identification in the prior art, the embodiment can identify the bumpy type of the road more timely. In addition, since the recognition results of three different algorithms are fused, the recognition of the bump type of the road by the embodiment is more accurate. In some possible implementations, the identifying, according to the collected preset data set, the bump type of the road by using an image identification algorithm, a neural network identification algorithm, and a dynamic time planning algorithm includes: and respectively identifying the confidence degrees of various possible bump types of the road by utilizing an image identification algorithm, a neural network identification algorithm and a dynamic time planning algorithm according to the acquired preset data set. In some possible implementations, the fusing the image recognition algorithm, the neural network recognition algorithm, and the dynamic time planning algorithm respectively recognize the bump type of the road, and taking the fused result as the bump type of the road, where the method includes: Taking the bump type with the largest matching degree after fusion as the bump type of the road, wherein the matching degree corresponding to any bump type i is Mi, Mi＝αiAi+βiBi+γiCi Wherein alpha i is the confidence weight of the image recognition algorithm on the bump type i, A i is the recognition matching degree of the image recognition algorithm on the bump type i, beta i is the confidence weight of the neural network algorithm on the bump type i, B i is the recognition matching degree of the neural network algorithm on the bump type i, gamma i is the confidence weight of the DTW algorithm on the bump type i, C i is the recognition matching degree of the dynamic time planning algorithm on the bump type i, Wherein a 1 is an illumination condition influence