CN113895461B - Vehicle lateral control method, device, vehicle and medium - Google Patents
Vehicle lateral control method, device, vehicle and medium Download PDFInfo
- Publication number
- CN113895461B CN113895461B CN202111359721.8A CN202111359721A CN113895461B CN 113895461 B CN113895461 B CN 113895461B CN 202111359721 A CN202111359721 A CN 202111359721A CN 113895461 B CN113895461 B CN 113895461B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- transverse
- data
- control
- lateral
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000006870 function Effects 0.000 claims description 24
- 230000015654 memory Effects 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 12
- 238000013527 convolutional neural network Methods 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 19
- 230000000694 effects Effects 0.000 abstract description 4
- 238000012549 training Methods 0.000 description 13
- 239000002609 medium Substances 0.000 description 12
- 230000008901 benefit Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 125000004432 carbon atom Chemical group C* 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000036461 convulsion Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000012120 mounting media Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/10—Path keeping
- B60W30/12—Lane keeping
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/18—Propelling the vehicle
- B60W30/18009—Propelling the vehicle related to particular drive situations
- B60W30/18145—Cornering
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/18—Propelling the vehicle
- B60W30/18009—Propelling the vehicle related to particular drive situations
- B60W30/18163—Lane change; Overtaking manoeuvres
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0013—Planning or execution of driving tasks specially adapted for occupant comfort
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2540/00—Input parameters relating to occupants
- B60W2540/18—Steering angle
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2720/00—Output or target parameters relating to overall vehicle dynamics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
The embodiment of the invention discloses a method and a device for controlling the transverse direction of a vehicle, the vehicle and a medium. The method comprises the following steps: acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of a vehicle at the current moment; inputting the transverse planning track parameters, transverse state deviation data and transverse control driving data into a pre-trained transverse deviation prediction model, and predicting transverse control deviation data of the vehicle at the next moment; and performing lateral control on the vehicle by using the lateral control deviation data. By adopting the scheme, the output result of the transverse deviation prediction model is directly the transverse control deviation data of the vehicle at the next moment, and the technical effect of simplifying the control process is achieved. Meanwhile, when the transverse control deviation data of the vehicle at the next moment is predicted, the transverse planning track parameter of the vehicle is taken as the input of the prediction model, so that the prediction result is more accurate.
Description
Technical Field
The embodiment of the invention relates to the technical field of automatic driving, in particular to a method and a device for controlling a vehicle transversely, a vehicle and a medium.
Background
In the field of automatic driving, a vehicle control system is one of the most central parts of an automatic driving system, and is used for controlling a vehicle so that the vehicle travels along a desired trajectory. In a vehicle control system, lateral tracking control of a vehicle is generally the last loop of an autonomous vehicle, and is used to keep the vehicle running straight, control the steering of the vehicle, change the lane in which the vehicle is located, and the like. Automatic driving therefore places high demands on the accuracy and stability of the lateral control.
When a vehicle is transversely controlled based on a conventional vehicle dynamics model, firstly, kinetic parameters (such as vehicle body moment of inertia and tire cornering stiffness) of the vehicle are identified through a plurality of sensors, then an output track is predicted based on the vehicle dynamics model, a pre-planned track is subtracted from the predicted output track to obtain a state error equation, and then the state error equation is used for subsequent rolling optimization and feedback control and further used for transverse tracking control of the vehicle.
When realizing the horizontal tracking control to the vehicle based on traditional scheme, the implementation process is comparatively loaded down with trivial details, can not directly output the horizontal state at future moment of vehicle, and need realize based on multiple sensor, and the cost is higher.
Disclosure of Invention
The embodiment of the invention provides a vehicle transverse control method, a vehicle transverse control device, a vehicle and a medium, which can optimize the existing related scheme of vehicle transverse state control, and can directly output transverse control deviation data of the vehicle at the next moment so as to be used for transverse control of the vehicle, simplify the operation process of the vehicle transverse control and reduce the calculation overhead of the vehicle transverse control.
In a first aspect, an embodiment of the present invention provides a vehicle lateral control method, including:
acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of a vehicle at the current moment;
inputting the transverse planning track parameters, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model, and predicting transverse control deviation data of the vehicle at the next moment;
and performing lateral control on the vehicle by using the lateral control deviation data.
In a second aspect, an embodiment of the present invention provides a vehicle lateral control apparatus, including:
the acquisition module is used for acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of the vehicle at the current moment;
the prediction module is used for inputting the transverse planning track parameters, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model and predicting the transverse control deviation data of the vehicle at the next moment;
and the control module is used for carrying out transverse control on the vehicle by utilizing the transverse control deviation data.
In a third aspect, an embodiment of the present invention provides a vehicle, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the vehicle lateral control method according to an embodiment of the present invention is implemented.
In a fourth aspect, embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements a vehicle lateral control method as provided by embodiments of the present invention.
According to the vehicle transverse control scheme provided by the embodiment of the invention, the transverse planning track parameter, the transverse state deviation data and the transverse control running data of the vehicle at the current moment are input into the transverse deviation prediction model, the output result is the predicted transverse control deviation data of the vehicle at the next moment, and the transverse control running data of the vehicle at the current moment is adjusted in time through the obtained transverse control deviation data, so that the vehicle runs at the next moment according to the adjusted transverse control running data, and the transverse control of the vehicle is realized. By adopting the technical scheme, the output result of the transverse deviation prediction model is directly the transverse control deviation data of the vehicle at the next moment, the technical effect of simplifying the control process is achieved, and the calculation overhead of the transverse control of the vehicle is reduced by simplifying the operation flow of the transverse control of the vehicle. Meanwhile, when the transverse control deviation data of the vehicle at the next moment is predicted, the transverse planning track parameter of the vehicle is taken as the input of the transverse deviation prediction model, so that the prediction result is more accurate, various sensors are not needed to be used for realizing the scheme, and the realization cost of the transverse control scheme of the vehicle can be reduced.
Drawings
Fig. 1 is a schematic flow chart of a vehicle lateral control method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating a vehicle lateral control method according to a second embodiment of the present invention;
fig. 3 is a structural block diagram of a vehicle lateral control device according to a third embodiment of the present invention;
fig. 4 is a block diagram of a vehicle according to a fourth embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some structures related to the present invention are shown in the drawings, not all of them.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the steps as a sequential process, many of the steps can be performed in parallel, concurrently or simultaneously. In addition, the order of the steps may be rearranged. The process may be terminated when its operations are completed, but could have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
Example one
Fig. 1 is a schematic flowchart of a vehicle lateral control method according to an embodiment of the present invention, where the method may be executed by a vehicle lateral control device, where the device may be implemented by software and/or hardware, and may be generally integrated in a computer device such as a server. As shown in fig. 1, the method includes:
and S110, acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of the vehicle at the current moment.
The vehicle control system is one of core technologies of automatic driving control execution, and mainly comprises longitudinal control and transverse control. Longitudinal control means driving and braking control of the vehicle, and accurate following of a desired vehicle speed is achieved through coordination of the accelerator and the brake. Lateral control refers to the control of the vehicle's steering wheel angle and tire force to achieve path tracking in an autonomous vehicle. The purpose of the transverse control of the vehicle is to enable the vehicle to run on a desired running route, and the vehicle has good riding comfort and stability under different speeds, loads, windage and road conditions.
Therefore, when the vehicle is controlled in the transverse direction, relevant factors influencing the transverse driving need to be selected and studied. In an embodiment of the present invention, the factors affecting the lateral driving may be: lateral planning trajectory parameters, lateral state deviation data, and lateral control travel data.
The transverse planned trajectory parameter may be a parameter related to a planned trajectory generated when the vehicle runs at the current time, and the current transverse planned trajectory may be obtained by acquiring, by a vehicle-mounted camera or an Inertial Measurement Unit (IMU), a center line of a lane line in a vehicle running process as a transverse planned trajectory, and a related parameter deviating from the center line as a transverse planned trajectory parameter, such as an angle parameter and a distance parameter.
The lateral state deviation data represents deviation data of the vehicle from the lateral planned trajectory during driving, and may be, for example and without limitation, lateral distance deviation data, heading angle deviation data, speed deviation data, acceleration deviation data, and the like of the vehicle from the planned trajectory of the vehicle.
The lateral control running data indicates data related to the vehicle lateral control affected by the vehicle running, and may be, for example, steering wheel angle data, steering wheel YAW RATE data (YAW RATE), vehicle speed data, vehicle acceleration data, and the like, without being limited thereto.
It should be noted that, in the vehicle lateral control method provided in the embodiment of the present invention, the specific acquisition manner and the contained content of the lateral planned trajectory parameter, the lateral state deviation data and the lateral control driving data of the acquired vehicle at the current time are not limited herein, and are mainly related to factors affecting the lateral control, subject to actual requirements.
And S120, inputting the transverse planning track parameters, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model, and predicting transverse control deviation data of the vehicle at the next moment.
The lateral deviation prediction model is one of machine learning models, and needs to be trained in advance before being used.
Optionally, the lateral deviation prediction model provided in the embodiment of the present invention is composed of a Convolutional Neural Network (CNN) and a Long Short Term Memory Network (LSTM). The convolutional neural network has the characteristic of extracting key data features through convolutional kernels, and learning and training costs can be reduced; the long and short term memory network is a time recurrent neural network, and can solve the problems of gradient extinction and gradient explosion in the long sequence training process.
The training process of the lateral deviation prediction model can be as follows: the method comprises the steps of taking transverse planning track parameters and transverse state deviation data in an off-line state as training samples according to transverse control driving data, presetting sample labels of the training samples, inputting the training samples into a convolutional neural network to extract transverse control related characteristic data of a vehicle, inputting the related characteristic data into a long-short term memory network to extract time series characteristic data, and finally outputting transverse control deviation data related to the next moment through an output layer of a model. And then, after outputting the transverse control deviation data each time, comparing the transverse control deviation data with a sample label corresponding to a training sample to judge the accuracy of the current training of the model, and correspondingly adjusting each model parameter in the convolutional neural network and the long-short term memory network according to the difference between the output result and the sample label. And after the model parameters are adjusted, continuously inputting new transverse planning track parameters and transverse state deviation data in the training samples in an off-line state into the convolutional neural network by using transverse control driving data, and executing the process again to further continuously optimize the transverse deviation prediction model until the output result is within an error range after the relevant test data is input into the model, namely the difference loss between the final output result and the sample labels of the training samples reaches certain convergence, and then the current transverse deviation prediction model is trained.
And further inputting the acquired transverse planning track parameters, transverse state deviation data and transverse control driving data into a trained transverse deviation prediction model, thereby outputting transverse control deviation data of the predicted vehicle at the next moment. Compared with the conventional scheme, the vehicle transverse Control scheme provided by the embodiment of the invention simplifies the implementation process of predicting the transverse Control deviation data of the vehicle at the next moment and further shortens the prediction time.
And S130, performing lateral control on the vehicle by using the lateral control deviation data.
After the lateral deviation prediction model outputs the lateral control deviation data of the vehicle at the next time in step S120, the controller may perform relevant optimization adjustment on the lateral control driving data at the current time according to the lateral control deviation data, so that the adjusted lateral control driving data can control the vehicle to drive according to the optimized data, thereby implementing lateral control on the vehicle.
According to the vehicle transverse control method provided by the embodiment of the invention, the transverse planning track parameter, the transverse state deviation data and the transverse control running data of the vehicle at the current moment are input into the pre-trained transverse deviation prediction model, the output result is the transverse control deviation data of the predicted vehicle at the next moment, and the transverse control running data of the vehicle at the current moment is adjusted in time through the obtained transverse control deviation data, so that the vehicle runs at the next moment according to the adjusted transverse control running data, and the transverse control of the vehicle is realized. By adopting the technical scheme, the output result of the transverse deviation prediction model is directly the transverse control deviation data of the vehicle at the next moment, the technical effect of simplifying the control process is achieved, and the calculation overhead of the transverse control of the vehicle is reduced by simplifying the operation flow of the transverse control of the vehicle. Meanwhile, when the transverse control deviation data of the vehicle at the next moment is predicted, the transverse planning track parameter of the vehicle is taken as the input of the transverse deviation prediction model, so that the prediction result is more accurate, various sensors are not needed to be used for realizing the scheme, and the realization cost of the transverse control scheme of the vehicle can be reduced.
Example two
The embodiment of the invention is further optimized on the basis of the embodiment, and the step of inputting the transverse planning track parameter, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model to predict the transverse control deviation data of the vehicle at the next moment is optimized, wherein the step comprises the following steps: inputting the lateral planned trajectory parameters, the lateral state deviation data, and the lateral control travel data into an objective function of the lateral deviation prediction model; and adopting the constraint conditions to carry out optimization solution on the input objective function to obtain the transverse control deviation data of the vehicle at the next moment. The advantage of this arrangement is that the lateral deviation prediction model is optimized by the objective function and the constraint conditions, so that the lateral control deviation data of the output vehicle at the next moment is more accurate.
The steps of obtaining the transverse planning track parameter, the transverse state deviation data and the transverse control driving data of the vehicle at the current moment are optimized, and the method comprises the following steps: acquiring a vehicle planning track through a vehicle camera, wherein the vehicle planning track is a lane center line in a vehicle driving lane; determining a transverse planned track parameter and transverse state deviation data of the vehicle at the current moment based on the vehicle planned track; and acquiring the transverse control running data of the vehicle at the current moment through a Controller Area Network (CAN) bus on the vehicle. The advantage of setting up like this is used the vehicle camera to realize the collection to vehicle planning orbit correlation data, need not to use expensive sensor, can reduce scheme implementation cost.
After the transverse planned trajectory parameters, the transverse state deviation data and the transverse control driving data of the vehicle at the current moment are obtained, the method further comprises the following steps: discarding abnormal data in the transverse control driving data, and filtering the residual transverse control driving data by using a preset filter; and after the transverse control driving data after the filtering processing is subjected to standardization processing, preprocessing the transverse control driving data. The advantage of this arrangement is that by preprocessing the lateral control travel data, the trained model can be made more accurate when using the preprocessed data for model training.
It is further optimized that, after the lateral control of the vehicle by using the lateral control deviation data, the method further comprises: and acquiring transverse control running data of the vehicle at the next moment, and combining transverse planned trajectory parameters and transverse state deviation data of the vehicle at the next moment, and continuously inputting the transverse planned trajectory parameters and the transverse state deviation data into the transverse deviation prediction model to realize transverse closed-loop feedback control of the vehicle. The method has the advantages that the transverse planning track parameters and the transverse state deviation data of the vehicle at the next moment are predicted according to the current vehicle state, so that the implementation of the transverse control scheme of the vehicle is continuous, and the implementation process of transverse control of the vehicle is simplified.
As shown in fig. 2, fig. 2 is a schematic flow chart of a vehicle lateral control method according to a second embodiment of the present invention; specifically, the method comprises the following steps:
and S210, acquiring a vehicle planning track through a vehicle camera, wherein the vehicle planning track is a lane central line in a vehicle driving lane.
In order to reduce the difficulty of data acquisition, the embodiment of the invention can acquire the planned track of the vehicle through the vehicle-mounted camera of the vehicle, and the specific acquisition mode can be that the center line of the lane in the driving lane of the vehicle is taken as the planned track of the transverse control, and the planned track is taken as the transverse target, so that the transverse deviation of the vehicle on the planned track can be judged, for example, the distance deviation, the course angle deviation and the like of the vehicle and the planned track can be realized.
And S220, determining transverse planned track parameters and transverse state deviation data of the vehicle at the current moment based on the vehicle planned track.
After the planned track is taken as a transverse target, a transverse planned track parameter and transverse state deviation data of the vehicle deviating from the planned track can be obtained at the current moment in the vehicle running process. For example, the transverse tracing parameters may be an angle parameter and a distance parameter; the lateral state deviation data can be lateral distance deviation data, heading angle deviation data, speed deviation data, acceleration deviation data, and the like of the vehicle from a planned trajectory of the vehicle.
And S230, acquiring transverse control running data of the vehicle at the current moment through a Controller Area Network (CAN) bus on the vehicle.
A Controller Area Network (CAN) bus is a serial communication Network that effectively supports distributed control or real-time control, so that transverse control driving data of a driving vehicle, such as steering wheel angle data, steering wheel yaw rate data, vehicle speed data, vehicle acceleration data, and the like, CAN be acquired in real time through the CAN bus.
And S240, discarding abnormal data in the transverse control driving data, and performing filtering processing on the rest transverse control driving data by using a preset filter.
For various data affecting the lateral control travel, abnormal data exceeding a preset set value may be discarded. The following are exemplary: discarding data in which the steering wheel turning angle data exceeds a set value δ, for example, 180 ° or more; discarding data of which the yaw rate data of the steering wheel exceeds a set value alpha, such as data above 0.05 degrees; by exceeding the vehicle acceleration by a set value beta, e.g. 4m/S 2 Discarding data above (meters per square second); the jerk exceeds a set value omega, e.g. 0.5m/S 3 More than (meters per cubic second) and more than a set value d, e.g., more than 20cm (inside), from the lane center line.
It should be noted that the values of δ, α, β, Ω, and d are not limited herein, and the values are generally different according to different vehicles and different usage scenarios.
And further, the residual transverse control driving data are subjected to filtering processing by using a preset filter, and the purpose of the filtering processing is to filter the data with larger difference in each group of obtained data, so that the obtained data are more accurate.
Alternatively, the preset filter may be a butterworth filter, a chebyshev filter, or an elliptic filter, and the like, which is not limited herein.
And S250, after the transverse control driving data after the filtering processing is subjected to standardization processing, preprocessing is carried out on the transverse control driving data.
On the basis of step S240, the remaining lateral control travel data (data of the steering wheel angle data, the steering wheel yaw rate data, the vehicle speed data, and the vehicle acceleration data) may be subjected to the normalization process.
Alternatively, when the normalization processing is performed, the processing may be performed in a Z-Score (Z-Score) manner, which may be expressed by the following expression:
in the above formula, Z represents a result of normalization processing of current data, X represents the current data, μ represents a data mean value corresponding to the current number, and V represents a data standard deviation corresponding to the current data.
By preprocessing the lateral control travel data in steps S240 and S250, the trained model can be more accurate when the preprocessed lateral control travel data is input to the lateral deviation prediction model.
And S260, inputting the transverse planning track parameter, the transverse state deviation data and the transverse control running data into an objective function of a transverse deviation prediction model.
The vehicle lateral deviation prediction model is composed of an objective function and constraint conditions which are well trained according to a non-convex optimization problem.
The benefit of using the non-convex optimization problem to train objective functions and constraints in the machine learning model is that the user can be made to apply the algorithm to the machine learning model according to the desired behavior, e.g., the objective function in the non-convex optimization can be expressed as a loss function that measures how good the fitting training data is. While the constraint on the objective function is an ability to allow the constraint model to encode behavior or knowledge, such as the size of the constraint model.
The vehicle transverse deviation prediction model provided by the embodiment of the invention consists of a convolutional neural network and a long-short term memory network, and when the vehicle transverse control is realized on the basis of the convolutional neural network and the long-short term memory network, a trained target function can be represented by the following expression:
in the above formula, p represents the prediction time domain, k represents the current time, u represents the input, y represents the output, m ∈ {1, …, p } is the control time domain,
and outputting the vehicle lateral deviation prediction model at the moment of step k.
After the obtained lateral planned trajectory parameters, lateral state deviation data, and lateral control travel data are input to the objective function, step S270 is further performed to optimize the objective function using the constraint condition.
And S270, optimizing and solving the input objective function by adopting constraint conditions to obtain the transverse control deviation data of the vehicle at the next moment.
The constraint condition may be expressed by the following expression:
u k ∈[u min,k ,u max,k ],k∈{0,...,m-1} (4)
Δu k ∈[Δu min,k ,Δu max,k ],k∈{0,...,m-1} (5)
in the above formula, the first and second carbon atoms are,
indicating a transverse directionInput of a deviation prediction model>
Is a related description of the lateral deviation prediction model, Δ u k ＝u k -u k-1 Represents the increment of step k time compared to input u at time k-1, { Q } y ,Q u Denotes a symmetric positive half-definite weight matrix corresponding to input and output, [ Δ u } min,k ,Δu max,k ]Represents the maximum and minimum values of the steering wheel allowable rotation at time k, [ Delta u ] min,k ,Δu max,k ]Representing the maximum and minimum values of the change in the allowable rotation of the steering wheel at time k compared to time k-1.
By using the three constraint conditions of the formula (3) to the formula (5), after the objective function is optimized and solved, the lateral deviation prediction model can output lateral control deviation data of the vehicle at the next moment.
And S280, acquiring transverse control running data of the vehicle at the next moment, and continuously inputting the transverse control running data into a transverse deviation prediction model by combining transverse planning track parameters and transverse state deviation data of the vehicle at the next moment so as to realize transverse closed-loop feedback control of the vehicle.
The method comprises the steps of acquiring transverse control driving data of a vehicle at the next moment (t + 1), combining transverse planning track parameters and transverse state deviation data of the vehicle at the next moment (t + 1), inputting the transverse control deviation data into a transverse deviation prediction model, outputting transverse control deviation data at the moment t +2, and transversely adjusting the vehicle driving at the moment t +1 based on the transverse control deviation data, so that the transverse control driving data output by the vehicle at the moment t +2 can be in a planning range, and transverse control of the vehicle is realized. Therefore, the predicted lateral control driving data output at the next moment can be combined with the lateral planning track parameter at the next moment and the output lateral state deviation data to serve as the input of the current moment, and the lateral closed-loop feedback control of the vehicle is achieved. The purpose of doing so can make the realization of this vehicle lateral control scheme possess the continuity, has simplified the implementation process of carrying out the lateral control to the vehicle.
According to the vehicle transverse control method provided by the embodiment of the invention, the vehicle planning track is obtained through the vehicle camera, so that the implementation cost of the vehicle transverse control scheme is reduced. And the acquired transverse control driving data is preprocessed, so that the trained model is more accurate when model training is carried out by using the preprocessed transverse control driving data. Furthermore, the transverse planning track parameters of the vehicle are taken into consideration as the input of the prediction model when the model is input, so that the prediction result is more accurate. And the transverse control prediction model is composed of a convolutional neural network and a long-short term memory network, so that the output of the prediction model is the transverse control deviation data directly at the next moment of the vehicle, and the control flow is simplified. And the realization of the vehicle transverse control scheme has continuity by a transverse closed-loop feedback control mode of a transverse planned track parameter and transverse state deviation data of the vehicle at the next moment by using the current vehicle state prediction.
EXAMPLE III
Fig. 3 is a block diagram of a vehicle lateral control apparatus according to a third embodiment of the present invention; the device can be realized by software and/or hardware, can be generally integrated in computer equipment such as a server and the like, and can realize the transverse control of the vehicle by executing the transverse control method of the vehicle provided by any embodiment of the invention. As shown in fig. 3, the apparatus includes: an acquisition module 31, a prediction module 32 and a control module 33, wherein:
the acquiring module 31 is configured to acquire a transverse planned trajectory parameter, transverse state deviation data, and transverse control driving data of the vehicle at a current time;
a prediction module 32, configured to input the transverse planned trajectory parameter, the transverse state deviation data, and the transverse control driving data into a pre-trained transverse deviation prediction model, and predict transverse control deviation data of the vehicle at a next time;
a control module 33 for performing lateral control of the vehicle using the lateral control deviation data.
According to the vehicle transverse control device provided by the embodiment of the invention, the transverse planning track parameter, the transverse state deviation data and the transverse control running data of the vehicle at the current moment are input into the pre-trained transverse deviation prediction model, the output result is the transverse control deviation data of the predicted vehicle at the next moment, and the transverse control running data of the vehicle at the current moment is adjusted in time through the obtained transverse control deviation data, so that the vehicle runs at the next moment according to the adjusted transverse control running data, and the transverse control of the vehicle is realized. By adopting the technical scheme, the output result of the transverse deviation prediction model is directly the transverse control deviation data of the vehicle at the next moment, the technical effect of simplifying the control process is achieved, and the calculation cost of the transverse control of the vehicle is reduced by simplifying the operation flow of the transverse control of the vehicle. Meanwhile, when the transverse control deviation data of the vehicle at the next moment is predicted, the transverse planning track parameter of the vehicle is taken as the input of the transverse deviation prediction model, so that the prediction result is more accurate, various sensors are not needed to be used for realizing the scheme, and the realization cost of the transverse control scheme of the vehicle can be reduced.
Optionally, the vehicle lateral deviation prediction model is composed of an objective function and constraint conditions trained according to a non-convex optimization problem.
Accordingly, the control module 33 includes: an input unit and an obtaining unit, wherein:
an input unit for inputting the lateral planned trajectory parameter, the lateral state deviation data, and the lateral control travel data into an objective function of the lateral deviation prediction model;
and the obtaining unit is used for adopting the constraint conditions to carry out optimization solution on the input objective function so as to obtain the transverse control deviation data of the vehicle at the next moment.
Optionally, the lateral deviation prediction model is composed of a convolutional neural network and a long-short term memory network.
Optionally, the obtaining module 31 includes: an acquisition unit and a determination unit, wherein:
the vehicle planning system comprises an acquisition unit, a display unit and a control unit, wherein the acquisition unit acquires a vehicle planning track through a vehicle camera, and the vehicle planning track is a lane center line in a vehicle driving lane;
the determining unit is used for determining transverse planning track parameters and transverse state deviation data of the vehicle at the current moment based on the vehicle planning track;
and the acquisition unit is also used for acquiring the transverse control running data of the vehicle at the current moment through a Controller Area Network (CAN) bus on the vehicle.
Optionally, the lateral state deviation data includes: at least one of lateral distance deviation data, course angular deviation data, speed deviation data, and acceleration deviation data of the vehicle from the vehicle planned trajectory;
the lateral control travel data includes: at least one of steering wheel angle data, steering wheel yaw rate data, vehicle speed data, and vehicle acceleration data.
Optionally, the apparatus further comprises: a processing module, wherein:
the processing module is used for discarding abnormal data in the transverse control driving data and filtering the rest transverse control driving data by using a preset filter; and after the transverse control driving data after the filtering processing is subjected to standardization processing, preprocessing the transverse control driving data.
Optionally, the obtaining module 31 is further configured to obtain lateral control driving data of the vehicle at the next time, and combine the lateral planned trajectory parameter and the lateral state deviation data of the vehicle at the next time to continuously input the lateral deviation prediction model, so as to implement lateral closed-loop feedback control of the vehicle.
The vehicle transverse control device provided by the embodiment of the invention can execute the vehicle transverse control method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects for executing the method.
Example four
The embodiment of the invention provides a vehicle, and the vehicle transverse control device provided by the embodiment of the invention can be integrated in the vehicle. Fig. 4 is a block diagram of a vehicle according to a fourth embodiment of the present invention. The vehicle 400 may include: a memory 401, a processor 402 and a computer program stored on the memory 401 and executable on the processor, the processor 402 implementing the vehicle lateral control method according to an embodiment of the present invention when executing the computer program.
The vehicle provided by the embodiment of the invention can execute the vehicle transverse control method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects for executing the method.
EXAMPLE five
Embodiments of the present invention also provide a storage medium containing computer-executable instructions for a vehicle lateral control method when executed by a computer processor, the method comprising:
acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of a vehicle at the current moment;
inputting the transverse planning track parameter, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model to predict transverse control deviation data of the vehicle at the next moment;
and performing lateral control on the vehicle by using the lateral control deviation data.
Storage medium-any of various types of memory devices or storage devices. The term "storage medium" is intended to include: mounting media such as CD-ROM, floppy disk, or tape devices; computer system memory or random access memory such as DRAM, DDRRAM, SRAM, EDORAM, lanbas (Rambus) RAM, etc.; non-volatile memory such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in a first computer system in which the program is executed, or may be located in a different second computer system connected to the first computer system through a network (such as the internet). The second computer system may provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations, such as in different computer systems that are connected via a network. The storage medium may store program instructions (e.g., embodied as a computer program) that are executable by one or more processors.
Of course, the storage medium containing the computer-executable instructions provided by the embodiments of the present invention is not limited to the vehicle lateral control operation described above, and may also perform related operations in the vehicle lateral control method provided by any embodiments of the present invention.
The vehicle transverse control device, the vehicle and the storage medium provided in the above embodiments can execute the vehicle transverse control method provided in any embodiment of the present invention, and have corresponding functional modules and beneficial effects for executing the method. For technical details that are not described in detail in the above embodiments, reference may be made to a vehicle lateral control method provided in any embodiment of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in some detail by the above embodiments, the invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the invention, and the scope of the invention is determined by the scope of the appended claims.
Claims (9)
1. A vehicle lateral control method, characterized by comprising:
acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of a vehicle at the current moment;
inputting the transverse planning track parameter, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model to predict transverse control deviation data of the vehicle at the next moment;
performing lateral control on the vehicle by using the lateral control deviation data;
the vehicle transverse deviation prediction model is composed of a target function and constraint conditions which are trained according to a non-convex optimization problem;
correspondingly, the inputting the transverse planning track parameter, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model to predict the transverse control deviation data of the vehicle at the next moment comprises:
inputting the lateral planned trajectory parameters, the lateral state deviation data, and the lateral control travel data into an objective function of the lateral deviation prediction model;
and adopting the constraint conditions to carry out optimization solution on the input objective function to obtain the transverse control deviation data of the vehicle at the next moment.
2. The method of claim 1, wherein the lateral bias prediction model consists of a convolutional neural network and a long-short term memory network.
3. The method of claim 1, wherein the obtaining of the lateral planned trajectory parameters, lateral state deviation data, and lateral control driving data of the vehicle at the current time comprises:
acquiring a vehicle planning track through a vehicle camera, wherein the vehicle planning track is a lane center line in a vehicle driving lane;
determining a transverse planned track parameter and transverse state deviation data of the vehicle at the current moment based on the vehicle planned track;
and acquiring the transverse control running data of the vehicle at the current moment through a Controller Area Network (CAN) bus on the vehicle.
4. The method of claim 3, wherein the lateral state deviation data comprises: at least one of lateral distance deviation data, course angular deviation data, speed deviation data, and acceleration deviation data of the vehicle from the vehicle planned trajectory;
the lateral control travel data includes: at least one of steering wheel angle data, steering wheel yaw rate data, vehicle speed data, and vehicle acceleration data.
5. The method of claim 1, after acquiring the lateral planned trajectory parameters, the lateral state deviation data, and the lateral control driving data of the vehicle at the current time, further comprising:
discarding abnormal data in the transverse control driving data, and filtering the residual transverse control driving data by using a preset filter;
and after the transverse control driving data after the filtering processing is subjected to standardization processing, preprocessing the transverse control driving data.
6. The method of claim 1, further comprising, after laterally controlling the vehicle using the lateral control deviation data:
and acquiring transverse control running data of the vehicle at the next moment, and combining transverse planned trajectory parameters and transverse state deviation data of the vehicle at the next moment, and continuously inputting the transverse planned trajectory parameters and the transverse state deviation data into the transverse deviation prediction model to realize transverse closed-loop feedback control of the vehicle.
7. A vehicle lateral control apparatus, characterized by comprising:
the acquisition module is used for acquiring transverse planning track parameters, transverse state deviation data and transverse control driving data of the vehicle at the current moment;
the prediction module is used for inputting the transverse planning track parameters, the transverse state deviation data and the transverse control driving data into a pre-trained transverse deviation prediction model and predicting the transverse control deviation data of the vehicle at the next moment;
a control module for performing lateral control on the vehicle using the lateral control deviation data;
the vehicle transverse deviation prediction model is composed of a target function and constraint conditions which are trained according to a non-convex optimization problem;
accordingly, the control module comprises: an input unit and an obtaining unit, wherein:
an input unit for inputting the lateral planned trajectory parameter, the lateral state deviation data, and the lateral control travel data into an objective function of the lateral deviation prediction model;
and the obtaining unit is used for adopting the constraint conditions to carry out optimization solution on the input objective function so as to obtain the transverse control deviation data of the vehicle at the next moment.
8. A vehicle comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-6 when executing the computer program.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111359721.8A CN113895461B (en) | 2021-11-17 | 2021-11-17 | Vehicle lateral control method, device, vehicle and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111359721.8A CN113895461B (en) | 2021-11-17 | 2021-11-17 | Vehicle lateral control method, device, vehicle and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113895461A CN113895461A (en) | 2022-01-07 |
| CN113895461B true CN113895461B (en) | 2023-04-14 |
Family
ID=79194460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111359721.8A Active CN113895461B (en) | 2021-11-17 | 2021-11-17 | Vehicle lateral control method, device, vehicle and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113895461B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019023628A1 (en) * | 2017-07-27 | 2019-01-31 | Waymo Llc | Neural networks for vehicle trajectory planning |
| CN110221611B (en) * | 2019-06-11 | 2020-09-04 | 北京三快在线科技有限公司 | Trajectory tracking control method and device and unmanned vehicle |
| WO2021004437A1 (en) * | 2019-07-05 | 2021-01-14 | Huawei Technologies Co., Ltd. | Method and system for predictive control of vehicle using digital images |
| CN111121777A (en) * | 2019-11-26 | 2020-05-08 | 北京三快在线科技有限公司 | Unmanned equipment trajectory planning method and device, electronic equipment and storage medium |
| CN111930015B (en) * | 2020-09-16 | 2021-06-04 | 北京三快在线科技有限公司 | Unmanned vehicle control method and device |
-
2021
- 2021-11-17 CN CN202111359721.8A patent/CN113895461B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113895461A (en) | 2022-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hima et al. | Trajectory tracking for highly automated passenger vehicles | |
| CN110794851B (en) | Vehicle remote control safety protection method and device and unmanned vehicle | |
| CN104691609A (en) | Algorithm for steering angle command to torque command conversion | |
| CN110967991A (en) | Method and device for determining vehicle control parameters, vehicle-mounted controller and unmanned vehicle | |
| CN110525436A (en) | Vehicle lane-changing control method, device, vehicle and storage medium | |
| CN108710294A (en) | A kind of unmanned vehicle path trace feed forward control method | |
| CN112099378B (en) | Front vehicle lateral motion state real-time estimation method considering random measurement time lag | |
| CN114942642A (en) | Unmanned automobile track planning method | |
| CN110027547B (en) | Vehicle lateral control method and device | |
| CN112937551B (en) | Vehicle control method and system considering input characteristics of driver | |
| US20240067185A1 (en) | Technology for dead time compensation during transverse and longitudinal guidance of a motor vehicle | |
| CN113895461B (en) | Vehicle lateral control method, device, vehicle and medium | |
| Vörös et al. | Lane keeping control using finite spectrum assignment with modeling errors | |
| CN111959506A (en) | Vehicle and control method and device for vehicle formation driving | |
| CN108919798B (en) | Hybrid function domain-based networked intelligent vehicle formation driving control method | |
| CN116834754A (en) | Transverse and longitudinal cooperative control method for self-adaptive speed regulation of automatic driving vehicle | |
| JP2023017703A (en) | Method and device for optimally parameterizing vehicle dynamic control system of vehicle | |
| Hu et al. | A model predictive control based path tracker in mixed-domain | |
| CN111857112A (en) | Automobile local path planning method and electronic equipment | |
| US20240059325A1 (en) | Autonomous racetrack driver coach and demonstrator | |
| CN115476881B (en) | Vehicle track tracking control method, device, equipment and medium | |
| CN111627247B (en) | Multi-vehicle formation control method and device | |
| US20230084702A1 (en) | System and method of adaptive, real-time vehicle system identification for autonomous driving | |
| Li et al. | Lane Departure Risk Assessment for Hands-free Driving Functions | |
| CN115098821A (en) | Trajectory reference curvature determination method, apparatus, device, medium, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |