CN-117104247-B - Method and system for estimating vehicle speed of bicycle and computer readable storage medium

CN117104247BCN 117104247 BCN117104247 BCN 117104247BCN-117104247-B

Abstract

The invention relates to a vehicle speed estimation method and system and a computer readable storage medium, wherein the vehicle speed estimation method comprises the steps of inputting a preset Kalman filtering model by taking wheel speed signals and acceleration signals as observables to obtain a first vehicle speed by vehicle speed estimation, clustering a plurality of radar targets relative to speed signals of a vehicle according to a preset clustering algorithm, obtaining a second vehicle speed according to a clustering result when the clustering is successful, identifying a current running condition when the clustering is successful, determining a weighting factor Z 1 and a weighting factor Z 2 according to the current running condition, carrying out fusion calculation according to the first vehicle speed, the weighting factor Z 1 , the second vehicle speed and the weighting factor Z 2 , outputting a fusion calculation result as a vehicle speed estimation result, and outputting the first vehicle speed as the vehicle speed estimation result when the clustering is failed. The invention ensures good estimation effect under the extreme nonlinear working condition, especially under the sudden acceleration and deceleration working condition and the low-attachment road surface.

Inventors

ZU GUOQIANG
WANG YI
HE QIAOJUN
LIU ZHEN
CAI LULONG

Assignees

广州汽车集团股份有限公司

Dates

Publication Date: 20260505
Application Date: 20220513

Claims (11)

1. A method of estimating a vehicle speed of a host vehicle, the method comprising: Acquiring wheel speed signals And acceleration signal ; By the wheel speed signal of the wheel And the acceleration signal The vehicle speed is estimated by taking the observed quantity as an input of a preset Kalman filtering model to obtain a first vehicle speed ; Acquiring speed signals of a plurality of radar targets in the surrounding environment of the vehicle relative to the own vehicle; inputting speed signals of the radar targets relative to the vehicle into a preset clustering model to perform clustering estimation of the vehicle speed, and obtaining a second vehicle speed according to a clustering result when the clustering is successful ; When the clustering is successful, the current running working condition is identified, and the first own vehicle speed is determined according to the current running working condition Is a weighting factor Z 1 of the second own vehicle speed Is selected to be a weighting factor Z 2 , according to the first bicycle speed Weighting factor Z 1 , second own vehicle speed And a weighting factor Z 2 for fusion calculation, and outputting a fusion calculation result as a vehicle speed estimation result; when the clustering fails, the first own vehicle speed is determined And outputting the result as an estimation result of the vehicle speed of the own vehicle.
2. The method of claim 1, wherein the identifying the current driving condition comprises: The method comprises the steps of obtaining a current accelerator pedal opening signal and a current brake pedal opening signal, comparing the current accelerator pedal opening signal and the current brake pedal opening signal with a plurality of preset opening value intervals, and determining the current running working condition according to a comparison result.
3. The method of claim 1, wherein the acquiring wheel speed signal Comprising: And acquiring wheel speed signals of two front wheels and wheel speed signals of two rear wheels of the vehicle, which are acquired by the wheel speed sensor, respectively carrying out filtering processing on the four wheel speed signals, and then taking an average value and outputting the four wheel speed signals as the wheel speed signals.
4. The method of claim 1, wherein said generating said wheel speed signal is based on said wheel speed signal And the acceleration signal The vehicle speed is estimated by taking the observed quantity as an input of a preset Kalman filtering model to obtain a first vehicle speed Comprising: according to the wheel speed signal of the wheel And the acceleration signal Obtaining a current state vector , ; According to the current state vector Computing a Kalman gain matrix , wherein, , Is a preset state measurement matrix, and is used for measuring the state of the object, When the vehicle is powered on again and the own vehicle speed of the Kalman filtering model is estimated for the first time for the target covariance matrix, the method is that For a preset fixed value, when the vehicle speed of the Kalman filtering model is estimated subsequently, the vehicle speed is estimated The value updated for the last vehicle speed estimation process; According to the Kalman gain matrix For the current state vector And a target covariance matrix A correction is made, wherein, , , ; According to the corrected current state vector Obtaining a Kalman filtered state vector According to the state vector Obtaining the first own vehicle speed Wherein, the method comprises the steps of, , A preset state transition matrix; according to the corrected target covariance matrix For the target covariance matrix Updating, after updating For the next vehicle speed estimation of the kalman filter model, wherein, Q is a preset measurement noise value.
5. The method according to claim 1, wherein inputting the speed signals of the plurality of radar targets relative to the own vehicle into a preset clustering model for cluster estimation of the own vehicle speed comprises: Clustering speed signals of the radar targets relative to the vehicle based on a preset clustering algorithm, wherein when the distance between any two clusters is smaller than a preset threshold d 1 in each clustering iteration, increasing the number K of the clusters and entering the next clustering iteration, when the cluster of the cluster center of any one speed signal and the corresponding cluster is larger than a preset threshold d 2 , reducing the number K of the clusters and entering the next clustering iteration, and when the cluster meets a preset cluster ending condition, ending the clustering iteration; After finishing the clustering iteration, selecting one cluster with the largest speed signal quantity, comparing the speed signal quantity contained in the one cluster with a preset threshold value d 3 , if the speed signal quantity contained in the one cluster is larger than the threshold value d 3 , judging that the clustering is successful, and obtaining a second own vehicle speed according to the speed signal contained in the one cluster Otherwise, judging that the clustering fails.
6. The method of claim 5, wherein the acquiring speed signals relative to the host vehicle for a plurality of radar targets in the vehicle surroundings comprises: Acquiring longitudinal speed signals and transverse speed signals of a plurality of radar targets in the surrounding environment of a vehicle detected by a vehicle millimeter wave radar relative to a vehicle, and eliminating abnormal signals of the longitudinal speed signals and the transverse speed signals of the plurality of radar targets to obtain speed signals for clustering.
7. The method of claim 6, wherein the obtaining the second bicycle speed is based on a speed signal contained in the one cluster Comprising: Selecting one longitudinal speed signal and one transverse speed signal which are closest to the cluster center of the cluster, and calculating a second bicycle speed according to the one longitudinal speed signal and the one transverse speed signal 。
8. A vehicle speed estimation system, comprising: A first signal acquisition unit for acquiring wheel speed signals of the wheels And acceleration signal ; A first vehicle speed estimation unit for estimating a wheel speed signal of the vehicle wheel And the acceleration signal The vehicle speed is estimated by taking the observed quantity as an input of a preset Kalman filtering model to obtain a first vehicle speed ; A second signal acquisition unit configured to acquire speed signals of a plurality of radar targets in a surrounding environment of the vehicle with respect to the own vehicle; A second vehicle speed estimation unit for inputting the speed signals of the multiple radar targets relative to the vehicle into a preset clustering model to perform clustering estimation of the vehicle speed, and obtaining a second vehicle speed according to the clustering result when the clustering is successful ; The vehicle speed output unit is used for identifying the current running condition when the second vehicle speed estimation unit is clustered successfully, and determining the first vehicle speed according to the current running condition Is a weighting factor Z 1 of the second own vehicle speed Is selected to be a weighting factor Z 2 , according to the first bicycle speed Weighting factor Z 1 , second own vehicle speed And a weighting factor Z 2 , outputting the fusion calculation result as a vehicle speed estimation result, and when the second vehicle speed estimation unit fails to cluster, outputting the first vehicle speed And outputting the result as an estimation result of the vehicle speed of the own vehicle.
9. The system according to claim 8, wherein the vehicle speed output unit is specifically configured to: The method comprises the steps of obtaining a current accelerator pedal opening signal and a current brake pedal opening signal, comparing the current accelerator pedal opening signal and the current brake pedal opening signal with a plurality of preset opening value intervals, and determining the current running working condition according to a comparison result.
10. The system according to claim 8, wherein the second signal acquisition unit is specifically configured to: Acquiring longitudinal speed signals and transverse speed signals of a plurality of radar targets in the surrounding environment of a vehicle detected by a vehicle millimeter wave radar relative to a vehicle, and eliminating abnormal signals of the longitudinal speed signals and the transverse speed signals of the plurality of radar targets to obtain speed signals for clustering.
11. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the vehicle speed estimation method according to any one of claims 1 to 7.

Description

Method and system for estimating vehicle speed of bicycle and computer readable storage medium Technical Field The invention relates to the technical field of vehicle speed estimation, in particular to a method and a system for estimating the vehicle speed of a bicycle and a computer readable storage medium. Background The estimation algorithm of vehicle running speed estimation mainly comprises a direct integration method, a Kalman filtering method, a neural network method, a nonlinear state observer method and the like. The method has low requirements on a vehicle model, contains less vehicle parameters, has certain robustness, is not suitable for long-time use, integrates noise signals while integrating sensor measurement signals, and particularly can not obtain a reliable and accurate state estimation value due to low signal-to-noise ratio of the measurement signals due to accumulation along with time under the condition of poor road conditions. The Kalman filtering method is a pre-estimation-correction algorithm with a numerical solution, a pre-estimation equation and a vehicle dynamics model are combined to calculate the state parameter and the error covariance of the next moment, and a correction equation and an output variable newly measured by a vehicle system are combined to finally obtain a state estimation value of the current moment. The neural network method has good experimental effect in a nonlinear region of automobile operation, but the method has great dependence on experimental data, slow convergence speed of model parameters, poor robustness of working conditions and low practical application value of engineering. The nonlinear state observer method estimates the longitudinal speed, a nonlinear vehicle model and a dynamic tire model are required to be constructed, nonlinear iterative computation is completed, the calculated amount is large, and real-time performance is difficult to ensure. In summary, a single vehicle speed estimation algorithm has respective defects, which cannot well solve the technical problem of vehicle running speed estimation, and further improvement is needed. Disclosure of Invention The invention aims to provide a vehicle speed estimation method and system and a computer readable storage medium, which can still ensure good estimation effect under the limiting nonlinear working condition, especially under the working condition of inaccurate wheel speed measurement under the rapid acceleration and deceleration working conditions, have smaller calculated amount and meet the real-time requirement. To achieve the above object, a first embodiment of the present invention provides a vehicle speed estimation method, including: acquiring a wheel speed signal v and an acceleration signal a of a wheel; The wheel speed signal V and the acceleration signal a are used as observational quantity to input a preset Kalman filtering model to estimate the vehicle speed so as to obtain a first vehicle speed V 1; acquiring speed signals of a plurality of radar targets in the surrounding environment of the vehicle relative to the own vehicle; Inputting speed signals of the radar targets relative to the vehicle into a preset clustering model to perform clustering estimation of the vehicle speed, and obtaining a second vehicle speed V 2 according to a clustering result when the clustering is successful; When the clustering is successful, the current running condition is identified, a weighting factor Z 1 of a first own vehicle speed V 1 and a weighting factor Z 2 of a second own vehicle speed V 2 are determined according to the current running condition, fusion calculation is carried out according to the first own vehicle speed V 1, the weighting factor Z 1, the second own vehicle speed V 2 and the weighting factor Z 2, and a fusion calculation result is output as an own vehicle speed estimation result; and when the clustering fails, outputting the first own vehicle speed V 1 as an own vehicle speed estimation result. Preferably, the identifying the current driving condition includes: The method comprises the steps of obtaining a current accelerator pedal opening signal and a current brake pedal opening signal, comparing the current accelerator pedal opening signal and the current brake pedal opening signal with a plurality of preset opening value intervals, and determining the current running working condition according to a comparison result. Preferably, the acquiring the wheel speed signal v includes: and acquiring wheel speed signals of two front wheels and wheel speed signals of two rear wheels of the vehicle, which are acquired by the wheel speed sensor, respectively carrying out filtering processing on the four wheel speed signals, and then taking the average value and outputting the four wheel speed signals as the wheel speed signal v. Preferably, the estimating the vehicle speed by using the wheel speed signal V and the acceleration signal a as observational quantities and inpu