CN-122009239-A - Vehicle control method and device

CN122009239ACN 122009239 ACN122009239 ACN 122009239ACN-122009239-A

Abstract

The embodiment of the application provides a vehicle control method and device. The control method comprises the steps of fitting a first motion increment sequence and a second motion increment sequence by adopting a basic spline curve to obtain a first continuous time track and a second continuous time track, performing time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track pair Ji Cancha, performing motion consistency evaluation according to the first continuous time track and the second continuous time track to obtain a motion consistency residual error, determining a comprehensive evaluation index according to the track alignment residual error and the motion consistency residual error, comparing the comprehensive evaluation index with a target threshold value, and triggering a control instruction of a vehicle according to a comparison result to control path planning of the vehicle. According to the embodiment of the application, the B-spline control points are used for generating the paths with continuous curvature, so that the uncertainty of prediction is quantified, and the safe driving of the vehicle in a dynamic driving environment is ensured.

Inventors

CUI HAOHAN
WANG CHUAN
MA YUNFENG
ZHANG ZIHAO
QIN QIANG
Shi Guangdie
WANG CHUNBO

Assignees

无锡北微传感科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260203
Priority Date: 20251216

Claims (10)

1. A control method of a vehicle, characterized by comprising: acquiring visual image stream data acquired by a camera in the vehicle and measurement data of an inertial measurement unit in the vehicle, wherein the visual image stream data comprises a first motion increment sequence of a visual image, and the measurement data comprises a second motion increment sequence of the inertial measurement unit; Fitting the first motion increment sequence and the second motion increment sequence by adopting a basic spline curve to obtain a first continuous time track and a second continuous time track, wherein the first continuous time track comprises the motion track of the vehicle induction characteristic points, and the second continuous time track comprises the motion track of the vehicle body; performing time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track pair Ji Cancha; Performing motion consistency assessment according to the first continuous time track and the second continuous time track to obtain a motion consistency residual; Determining a comprehensive evaluation index according to the track alignment residual error and the motion consistency residual error; and comparing the comprehensive evaluation index with a target threshold value, and triggering a control instruction of the vehicle according to a comparison result so as to control the path planning of the vehicle.
2. The control method of the vehicle according to claim 1, wherein the acquiring the visual image stream data collected by the camera in the vehicle and the measurement data of the inertial measurement unit in the vehicle includes: acquiring visual image stream data of the camera and measurement data of the inertial measurement unit through online subscription or offline importing a preset configuration file; Based on a sliding window buffer memory module, dynamically updating the visual image stream data and the measurement data acquired at the current moment according to a preset time length; processing the visual image stream data based on a visual odometer or a synchronous positioning and mapping module, and outputting the first motion increment sequence; And processing the measurement data through mechanized integration, and outputting the second motion increment sequence.
3. The method of controlling a vehicle of claim 1, wherein said fitting said first sequence of motion increments and said second sequence of motion increments with a base spline curve to obtain said first continuous time trace and said second continuous time trace comprises: according to the first motion increment sequence and the second motion increment sequence, node vector distribution is carried out, and a first node vector sequence corresponding to the first motion increment sequence and a second node vector sequence corresponding to the second motion increment sequence are obtained; Determining a first control node and a second control node according to the first node vector sequence, the second node vector sequence and the basis function matrix; Determining the first continuous time track through a Deboolean-Cox recursion formula according to the first control node; And determining the second continuous time track according to the second control node through a Deboolean-Cox recursion formula.
4. The method for controlling a vehicle according to claim 3, wherein the step of performing node vector assignment according to the first motion delta sequence and the second motion delta sequence to obtain a first node vector sequence corresponding to the first motion delta sequence and a second node vector sequence corresponding to the second motion delta sequence includes: determining a predicted angular velocity of the vehicle in a continuous time domain from the first sequence of motion increments and the second sequence of motion increments; Under the condition that the value of the predicted angular velocity is larger than a first preset value, carrying out node vector distribution based on chord length parameterization so as to make the parameter distribution of a first node vector sequence corresponding to the first motion increment sequence and a second node vector sequence corresponding to the second motion increment sequence proportional to displacement; And under the condition that the value of the predicted angular velocity is smaller than a second preset value, carrying out node vector distribution based on uniform parameterization so as to uniformly distribute parameters of a first node vector sequence corresponding to the first motion increment sequence and a second node vector sequence corresponding to the second motion increment sequence.
5. The method for controlling a vehicle according to claim 1, wherein the performing time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track alignment residual comprises: performing frequency domain transformation on the first continuous time track phase response function to obtain a first frequency spectrum; Performing frequency domain transformation on the second continuous time track phase response function to obtain a second frequency spectrum; Calculating a cross spectrum phase and a group delay parameter of the cross spectrum phase according to the first frequency spectrum and the second frequency spectrum: Fitting a linear phase model in an effective frequency band of an inertial measurement unit, and extracting an average value and covariance of the group delay parameters; And determining the track alignment residual error according to the average value and the covariance of the group delay parameters.
6. The method for controlling a vehicle according to claim 1, wherein the performing time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track alignment residual comprises: performing frequency domain transformation on the first continuous time track phase response function to obtain a first frequency spectrum; Performing frequency domain transformation on the second continuous time track phase response function to obtain a second frequency spectrum; Calculating a cross spectrum phase and a group delay parameter of the cross spectrum phase according to the first frequency spectrum and the second frequency spectrum: fitting a linear phase model in an effective frequency band of an inertial measurement unit, and extracting an average value of the group delay parameters; carrying out frequency domain coherence calculation according to the average value of the group delay parameters to determine a peak value position, peak value energy corresponding to the peak value position and time domain offset; and determining the track alignment residual error according to the average value of the group delay parameters, the peak energy corresponding to the peak position and the time domain offset.
7. The method for controlling a vehicle according to claim 1, wherein performing the extrinsic parameter consistency assessment according to the first continuous time track and the second continuous time track to obtain a motion consistency residual, comprises: calculating curvature first derivative deviation of the first continuous time track and the second continuous time track, and generating a steering dynamic residual error; Calculating the second derivative deviation of acceleration of the first continuous time track and the second continuous time track to generate a braking inertia residual error; and determining the motion consistency residual according to the distribution average value of the steering dynamic residual and the distribution average value of the braking inertia residual.
8. The method of controlling a vehicle according to claim 1, wherein the determining a comprehensive evaluation index from the trajectory alignment residual and the motion consistency residual includes: Under the condition that the speed of the vehicle is outside a first preset interval, a first weighting coefficient, the track alignment residual error and the motion consistency residual error are fused, and a comprehensive evaluation index is determined; under the condition that the angular speed of the steering of the vehicle is outside a second preset interval, a second weighting coefficient, the track alignment residual error and the motion consistency residual error are fused, and a comprehensive evaluation index is determined; And under the condition that the acceleration of the vehicle is located outside a third preset interval, fusing a third weighting coefficient, the track alignment residual error and the motion consistency residual error, and determining a comprehensive evaluation index.
9. The control method of a vehicle according to any one of claims 1 to 8, characterized in that the control method of a vehicle further includes: Providing a visual interface, the visual interface being configurable to generate a format report; Based on the data updating signal provided by the vehicle, the visual interface displays the visual image stream data and the measurement data, the track alignment residual error, the motion consistency residual error, the comprehensive evaluation index and the driving state of the vehicle which are updated in real time.
10. A control device for a vehicle, comprising: A data acquisition module configured to acquire visual image stream data acquired by a camera in the vehicle and measurement data of an inertial measurement unit in the vehicle, the visual image stream data including a first sequence of motion increments of a visual image, the measurement data including a second sequence of motion increments of the inertial measurement unit; A fitting module configured to fit the first motion delta sequence and the second motion delta sequence with a base spline curve to obtain the first continuous time track and the second continuous time track, wherein the first continuous time track comprises the motion track of the vehicle sensing feature point, and the second continuous time track comprises the motion track of the vehicle body; The time synchronization evaluation module is configured to perform time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track pair Ji Cancha; The external parameter consistency evaluation module is configured to perform external parameter consistency evaluation according to the first continuous time track and the second continuous time track so as to obtain a motion consistency residual; The comprehensive evaluation module is configured to determine a comprehensive evaluation index according to the track alignment residual error and the motion consistency residual error; And the signal triggering module is configured to compare the comprehensive evaluation index with a target threshold value and trigger a control instruction of the vehicle according to a comparison result so as to control the path planning of the vehicle.

Description

Vehicle control method and device Technical Field The application relates to the technical field of vehicles, in particular to a vehicle control method and device. Background In the technical field of intelligent driving of vehicles, a perception and positioning system depends on the cooperation of a vision-inertia combination (VI) system and a path planning algorithm. In the related art, the space-time external parameter calibration of the camera and the IMU needs to jointly optimize internal parameters, distortion and time offset through an offline tool chain (such as Kalibr), but the special motion excitation and the calibration target are relied on, the data quality is difficult to ensure in a dynamic driving scene (such as high-speed lane change and sharp turning), the on-line self-calibration causes slow convergence or hidden deviation due to state coupling, the positioning accuracy is influenced, the path planning of the vehicle cannot be fused with real-time perception data to dynamically adjust a control point, and response delay is caused when the vehicle faces to a temporary obstacle, so that the potential safety hazard of driving of the vehicle is increased. In addition, the existing evaluation link severely depends on a dynamic capture system, a laser map and other truth sources to calculate Absolute Track Errors (ATE) and Relative Pose Errors (RPE), and the existing non-truth method (such as reprojection errors and gravity direction consistency) is characterized by single index (only capable of verifying rotation external parameters), lack of a unified base line (residual norms are difficult to transversely compare) and a long-term drift monitoring blind area, so that time-space alignment advantages and disadvantages cannot be quantified, and the reference track distortion is caused by drift of a VI system, so that the path planning errors of vehicles are further amplified, and safe driving is influenced. Therefore, a systematic non-truth evaluation framework is needed to be established for intelligent driving of vehicles at present so as to unify standard quantification camera-IMU alignment quality and solve the problems of positioning drift, infeasibility of paths and the like caused by calibration misalignment, track distortion and control constraint deletion in a dynamic driving environment. Disclosure of Invention The embodiment of the application provides a vehicle control method and a vehicle control device, wherein a path with continuous curvature is generated through a B-spline control point, so that the problem of curvature mutation of a traditional straight line-circular arc combination is avoided, steering jitter is reduced, a B-spline curve is fitted based on a history track point, the curve prediction precision is improved by combining lane line information, and the problem of error of a fixed curvature algorithm in a curve is solved. The multi-index fusion scoring system establishes a unified quantization standard to realize quantitative and comparable alignment evaluation of the camera and the inertial measurement unit. Therefore, even in the scene that the dynamic capture or the laser map cannot be acquired, the alignment quality score can be directly acquired, the comparison of different algorithms or parameter configurations is supported, and the test cost is remarkably reduced. Finally, two indexes are fused to generate a comprehensive evaluation index, and when the comprehensive evaluation index is lower than a target threshold value, a navigation correction instruction is triggered, so that the safe driving of the vehicle in a dynamic driving environment is ensured. The vehicle control method comprises the steps of obtaining visual image flow data collected by a camera in a vehicle and measurement data of an inertial measurement unit in the vehicle, wherein the visual image flow data comprise a first motion increment sequence of a visual image, the measurement data comprise a second motion increment sequence of the inertial measurement unit, fitting the first motion increment sequence and the second motion increment sequence by adopting a basic spline curve to obtain a first continuous time track and a second continuous time track, the first continuous time track comprises a motion track of a vehicle induction characteristic point, the second continuous time track comprises a motion track of a vehicle body, performing time synchronization evaluation according to the first continuous time track and the second continuous time track to obtain a track pair Ji Cancha, performing motion consistency evaluation according to the first continuous time track and the second continuous time track to obtain a motion consistency residual, determining a comprehensive evaluation index according to the track alignment residual and the motion consistency residual, comparing the comprehensive evaluation index with a target, and performing a control instruction to control path planning and tr