CN114919590A - Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium - Google Patents
Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114919590A CN114919590A CN202210663210.3A CN202210663210A CN114919590A CN 114919590 A CN114919590 A CN 114919590A CN 202210663210 A CN202210663210 A CN 202210663210A CN 114919590 A CN114919590 A CN 114919590A
- Authority
- CN
- China
- Prior art keywords
- speed
- vehicle speed
- vehicle
- preset
- calibration
- 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.)
- Pending
Links
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
- 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
- B60W40/105—Speed
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The application relates to a speed determination method and device for an automatic driving vehicle, an electronic device and a storage medium. The method comprises the following steps: acquiring a first vehicle speed and a second vehicle speed of the automatic driving vehicle, wherein the first vehicle speed is a vehicle running speed acquired when the automatic driving vehicle runs straight and the RTK differential signal quality is good, and the second vehicle speed is a calibration speed of the first speed; training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model; and acquiring a third vehicle speed when the RTK differential signal quality of the automatic driving vehicle is not good, and determining the calibration speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed. According to the scheme, the vehicle speed calibration model is obtained when the RTK differential signal quality is good, the preset speed interval is selected according to the vehicle speed when the RTK differential signal is not good, and the vehicle speed calibration model corresponding to the preset speed interval is selected to calibrate the vehicle speed in real time, so that the accuracy of vehicle speed calibration is improved.
Description
Technical Field
The present application relates to the field of navigation technologies, and in particular, to a method and an apparatus for determining a speed of an autonomous vehicle, an electronic device, and a storage medium.
Background
In the related art, the speed of a vehicle is an important control variable in autonomous driving. The control of the speed of the vehicle closely affects the safety of the occupants, and the accuracy of measuring the speed of the vehicle affects the judgment of the autonomous driving system. Measuring the speed of an autonomous vehicle is typically measured by an internal sensor of the autonomous vehicle, but there is often a deviation in the vehicle travel speed from an internal cross sensor, typically including the vehicle chassis speed and the autonomous vehicle wheel speed from the internal sensor. The electromagnetic sensor is usually adopted for measuring the chassis speed of the vehicle, the electromagnetic sensor collects electromagnetic induction signals through the rotation of wheels, and when the road surface is uneven or the vehicle vibrates, the measured speed has errors. The speed of the bottom of the vehicle can be measured by adopting a light sensor and determining the chassis speed of the vehicle by utilizing the reflection time of a light beam from the ground, but the method has higher requirements on the placing position and the stability of an optical device and has errors of a measuring device. Therefore, the calibration speed of the automatic driving vehicle can be obtained from the positioning system and the inertia measurement unit of satellite navigation positioning to calibrate the vehicle speed.
However, in some scenarios, such as under bridges, culverts, tunnels, dense buildings, and other locations with poor positioning signals, the calibration speed of the obtained autonomous vehicle is often not accurate enough, or even cannot be provided by the positioning system based on the RTK differential signal.
Therefore, it is necessary to provide a method for accurately calibrating the vehicle driving speed when the RTK differential signal is not good.
Disclosure of Invention
In order to solve or partially solve the problems in the related art, the application provides a method and a device for determining the speed of an autonomous vehicle, an electronic device and a storage medium, which can solve the problem that the speed of the autonomous vehicle cannot be accurately calibrated when an RTK differential signal is not good.
A first aspect of the present application provides a method of determining a speed of an autonomous vehicle, comprising:
acquiring a first vehicle speed of the automatic driving vehicle, wherein the first vehicle speed is a vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle meets a first preset condition and the course angle of the automatic driving vehicle meets a second preset condition, the first preset condition is a judgment condition when the RTK differential signal quality of the automatic driving vehicle is available, and the second preset condition is the course angle of the automatic driving vehicle in a straight-ahead state;
acquiring a second vehicle speed of the automatic driving vehicle, wherein the second vehicle speed is a calibration speed corresponding to the first vehicle speed;
training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, wherein the vehicle speed calibration model is used for receiving the first vehicle speed and outputting a second vehicle speed corresponding to the first vehicle speed;
and acquiring a third vehicle speed, wherein the third vehicle speed is the vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle does not meet the first preset condition, and determining the calibration speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed.
Preferably, the first vehicle speed is a vehicle chassis speed measured by a vehicle speed sensor of the autonomous vehicle, the second vehicle speed is a vehicle running speed calculated according to an RTK differential signal corresponding to the first vehicle speed, and a time of the second vehicle speed is the same as a time of the first vehicle speed corresponding to the second vehicle speed.
Preferably, training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, comprising:
determining a model error parameter of a preset linear model according to the speed error coefficient;
setting a preset speed interval, and initializing a model error parameter of a preset linear model in the preset speed interval;
acquiring a training sample set of a preset linear model in a preset speed interval according to a first vehicle speed in the preset speed interval and a second vehicle speed corresponding to the first vehicle speed;
and inputting the sample training set in the preset speed interval into a preset linear model, and optimizing model error parameters to obtain a vehicle speed calibration model corresponding to each preset speed interval.
Preferably, the method includes inputting a sample training set in a preset speed interval into a preset linear model, and performing model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval, including:
inputting a first vehicle speed to the preset linear model, and acquiring a fourth vehicle speed output by the preset linear model according to the first vehicle speed;
calculating a value of a loss function of a preset linear model according to a fourth vehicle speed output by the first vehicle speed and a second vehicle speed corresponding to the first vehicle speed;
and optimizing the model error parameters of the preset linear model according to the values of the loss function values so as to obtain the vehicle speed calibration model corresponding to each preset speed interval.
Preferably, the model error coefficient includes a first calibration coefficient and a second calibration coefficient, and the setting of the model error parameter of the preset linear model according to the speed error coefficient includes:
acquiring a speed error coefficient, and setting a first calibration coefficient and a second calibration coefficient of a preset linear model according to the error coefficient of the automatic driving vehicle, wherein the speed error coefficient is an error coefficient between a first vehicle speed and a second vehicle speed;
and obtaining the product of the first calibration coefficient and the first vehicle speed, and determining the preset linear model according to the product of the first calibration coefficient and the first vehicle speed and the second calibration coefficient.
Preferably, within the preset speed interval, initializing an error parameter of the preset linear model, including:
in each preset speed interval, setting a first calibration coefficient of a preset linear model to be 1, and setting a second calibration coefficient of the preset linear model to be 0.
Preferably, determining a calibrated speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed comprises:
determining a preset speed interval where the third vehicle speed is located;
and inputting the third vehicle speed into a vehicle speed calibration model corresponding to the preset speed interval to obtain a calibration speed corresponding to the third vehicle speed.
A second aspect of the present application provides a speed determination device of an autonomous vehicle, comprising:
the automatic driving vehicle control system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring a first vehicle speed of the automatic driving vehicle, the first vehicle speed is a vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle meets a first preset condition and the course angle of the automatic driving vehicle meets a second preset condition, the first preset condition is a judgment condition when the RTK differential signal quality of the automatic driving vehicle is available, and the second preset condition is the course angle of the automatic driving vehicle in a straight running state;
the second acquiring unit is used for acquiring a second vehicle speed of the automatic driving vehicle, wherein the second vehicle speed is a calibration speed corresponding to the first vehicle speed;
the training unit is used for training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, and the vehicle speed calibration model outputs a second vehicle speed corresponding to the first vehicle speed according to the first vehicle speed;
and the calibration unit is used for acquiring a third vehicle speed, wherein the third vehicle speed is the vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle does not meet the first preset condition, and the calibration speed of the third vehicle speed is determined based on the vehicle speed calibration model and the third vehicle speed.
Preferably, the device further comprises a parameter calibration unit, wherein the parameter calibration unit is used for obtaining a model error parameter of the vehicle speed calibration model and adjusting a parameter value of the inertia measurement unit according to the model error parameter of the vehicle speed calibration model.
A third aspect of the present application provides an electronic device comprising:
a processor; and
a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method as described above.
A fourth aspect of the present application provides a computer-readable storage medium having stored thereon executable code, which, when executed by a processor of an electronic device, causes the processor to perform a method as described above.
The technical scheme provided by the application can comprise the following beneficial effects: on one hand, because the original sample set used for training the speed calibration model is obtained when the RTK differential signal of the automatic driving vehicle meets the preset condition, the accurate calibration of the chassis speed of the vehicle can be realized based on the trained speed calibration model when the RTK differential signal is not good; on the other hand, each preset speed interval corresponds to the original sample set, so that the problem that the target speed calibration model is not accurately calibrated due to the fact that the sample data span is large can be solved, namely, through the subdivided original sample sets, more accurate training samples can be provided for the speed calibration model, and therefore the target calibration model obtained accordingly can be used for accurately calibrating the vehicle running speed.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The foregoing and other objects, features and advantages of the application will be apparent from the following more particular descriptions of exemplary embodiments of the application as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the application.
FIG. 1 is a schematic flow chart diagram illustrating a method for determining a speed of an autonomous vehicle in accordance with an embodiment of the present application;
FIG. 2 is another schematic flow diagram of a method for determining a speed of an autonomous vehicle according to an embodiment of the application;
FIG. 3 is another schematic flow chart diagram illustrating a method for determining a speed of an autonomous vehicle in accordance with an embodiment of the present application;
FIG. 4 is another schematic flow chart diagram illustrating a method for determining a speed of an autonomous vehicle in accordance with an embodiment of the present application;
FIG. 5 is a schematic diagram of a speed determination device for an autonomous vehicle according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device shown in an embodiment of the present application.
Detailed Description
Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While embodiments of the present application are illustrated in the accompanying drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It should be understood that although the terms "first," "second," "third," etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
In the related art, the running speed of the vehicle obtained directly through the internal sensor of the autonomous vehicle often has deviation, such as the chassis speed of the vehicle and the wheel speed of the autonomous vehicle obtained through the internal sensor of the autonomous vehicle, so that the calibrated speed of the autonomous vehicle can be obtained from a positioning system and an inertial measurement unit of satellite navigation positioning. However, in some scenes, such as under bridges, culverts, tunnels, dense buildings and other places with poor positioning signals, the calibration speed of the obtained automatic driving vehicle is often not accurate enough or even cannot be provided by a positioning system based on the RTK differential signal.
In order to solve or partially solve the problems in the related art, the present application provides a method and an apparatus for determining a speed of an autonomous vehicle, an electronic device, and a storage medium, which can solve the problem that the speed of the autonomous vehicle cannot be accurately calibrated when an RTK differential signal is not good.
Referring to fig. 1, a speed determination method of an autonomous vehicle includes:
step S101, a first vehicle speed of the automatic driving vehicle is obtained, wherein the first vehicle speed is a vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle meets a first preset condition and the course angle of the automatic driving vehicle meets a second preset condition, the first preset condition is a judgment condition when the RTK differential signal quality of the automatic driving vehicle is available, and the second preset condition is the course angle of the automatic driving vehicle in a straight running state of the automatic driving vehicle.
In step S101, the first vehicle speed is a vehicle chassis speed measured by a vehicle speed sensor of the autonomous vehicle. The satellite navigation System for acquiring the RTK differential signal may be any one of a GPS (Global Positioning System), a beidou satellite Positioning System, and an RTK (Real Time Kinematic) Positioning System. Under the condition that the RTK differential signal is good, the calibration speed of the automatic driving vehicle obtained according to the RTK differential signal is accurate. However, in a scenario where the RTK differential signal quality is poor, the calibration speed obtained according to the RTK differential signal is not very accurate, and therefore, a sample sampling set of each preset speed interval needs to be obtained in a scenario where the satellite quality signal is good. Due to the fact that the RTK differential signals with poor quality can cause data of the original sample set to be inaccurate, before the original sample set is collected, the quality of the RTK differential signals needs to be judged first, and the RTK differential signals meeting preset conditions are selected as data sources of calibration speed. Therefore, when the first vehicle speed and the second vehicle speed are adopted to train the preset linear model so as to determine the speed calibration model, the first vehicle speed and the second vehicle speed when the RTK differential signal quality is good need to be adopted,
specifically, the step of determining the quality of the RTK differential signal includes: and receiving an observed value of the RTK differential signal, and solving the RTK differential signal according to the observed value, wherein the solution of the observed value comprises a single-point fixed solution, a pseudo-range solution, a fixed solution and a floating solution. And if the solution of the obtained observation value is a fixed solution, judging that the RTK differential signal has better quality and accords with a preset condition.
In one embodiment, the autonomous vehicle traveling state includes a straight traveling state, a curve traveling state, and the like of the autonomous vehicle.
In one embodiment, the step S101 of determining the straight-ahead state of the heading angle of the autonomous vehicle is determined according to the RTK differential signal, and specifically includes: acquiring a course angle of the automatic driving vehicle according to the RTK differential signal of the automatic driving vehicle; and if the change angle of the course angle is smaller than the preset change angle, judging that the running state of the automatic driving vehicle is a straight running state. The course angle of the automatic driving vehicle can be calculated according to the RTK differential signal and can also be calculated according to data provided by the inertial measurement unit.
And S102, acquiring a second vehicle speed of the automatic driving vehicle, wherein the second vehicle speed is a calibration speed corresponding to the first vehicle speed.
In step S102, the time of the second vehicle speed is the same as the time of the first vehicle speed corresponding to the second vehicle speed. The second vehicle speed is obtained when the RTK differential signal quality is good, and the second vehicle speed is a calibration speed with more accurate first vehicle speed.
Step S103, training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, wherein the vehicle speed calibration model is used for receiving the first vehicle speed and outputting the second vehicle speed corresponding to the first vehicle speed.
In one embodiment, as shown in fig. 2, training a preset linear model according to a first vehicle speed and a second vehicle speed to obtain a vehicle speed calibration model, includes:
s201, determining a model error parameter of a preset linear model according to the speed error coefficient.
Step S201 eliminates an error between the first vehicle speed and the second vehicle speed by setting a linear model. The speed error coefficient is a calibration error coefficient between a first vehicle speed and a second vehicle speed, and a preset linear model is adopted between the first vehicle speed and the second vehicle speed in the application, so that a first calibration coefficient and a second calibration coefficient of the preset linear model can be assumed, and the first calibration coefficient and the second calibration coefficient are determined through the first vehicle speed, the second vehicle speed and the speed error coefficient. In step S201, determining a model error parameter of a preset linear model according to the speed error coefficient, including: acquiring an error coefficient of an automatic driving vehicle, and setting a first calibration coefficient and a second calibration coefficient of a preset linear model according to the error coefficient of the automatic driving vehicle; and obtaining the product of the first calibration coefficient and the first vehicle speed, and determining the preset linear model according to the product of the first calibration coefficient and the first vehicle speed and the second calibration coefficient.
In one embodiment, the pre-setting the linear model comprises:
v＝k 1 *V 0 +b 1 ；
wherein, V 0 And v is a calibrated speed corresponding to the first vehicle speed, namely a second vehicle speed. k is a radical of 1 For calibrating the first calibration coefficient of the model for speed, b 1 And calibrating a second calibration coefficient of the model for the speed, wherein the model error parameter comprises the first calibration coefficient and the second calibration coefficient.
S202, setting a preset speed interval, and initializing a model error parameter of a preset linear model in the preset speed interval.
The preset speed intervals are obtained according to the speed division of the vehicle chassis, the speed of the chassis is divided into a plurality of preset speed intervals, and then the speed calibration method is obtained according to the preset speed intervals, so that the problem that the obtained speed calibration method is not accurate enough due to the fact that the original sample set spans too large can be solved. Specifically, the chassis speed of the autonomous vehicle may be divided into a low-speed travel preset speed section, a medium-speed travel preset speed section, and a high-speed travel preset speed section according to the driving habits of the driver and the standards of the autonomous driving industry. For example, the low-speed running preset speed interval is 2m/s to 10m/s, the medium-speed running preset speed interval is 10m/s to 25m/s, and the high-speed running preset speed interval is more than 25 m/s.
In one embodiment, the preset speed interval may also be acquired according to a driving state of the autonomous vehicle. For example, according to the running state of the autonomous vehicle, the autonomous vehicle is divided into a starting preset speed section, a normal running preset speed section, an acceleration running preset speed section, and a high-speed running section. Each running state interval can be divided into a multi-gear preset speed interval according to the running speed of the vehicle. For example, for the starting preset speed interval, the preset speed interval can be divided into a first-gear vehicle speed, a second-gear vehicle speed and a third-gear vehicle speed, wherein the first-gear vehicle speed corresponds to a speed interval of 0km/h to 15km/h, and the second-gear preset speed interval is: 10 km/h-25 km/h, and the three-gear preset speed interval is as follows: 20km/h to 45 km/h. For the preset speed intervals in other states, the preset speed interval can be obtained according to the running state of the automatic driving vehicle by referring to the method for obtaining the preset speed interval from the starting preset speed interval.
Optionally, in an embodiment, an original sample set of a preset speed interval under a relevant environmental condition may be obtained in combination with an environmental condition and a preset speed interval under which the autonomous vehicle runs, so that the speed calibration model trained by the original sample set is more targeted. For example, the driving scene information of the autonomous vehicle, the road condition information and the driving environment information of the autonomous vehicle are acquired to determine the environmental condition of the autonomous vehicle, the maximum driving speed of the autonomous vehicle and the minimum driving speed of the autonomous vehicle are determined according to the environmental condition of the autonomous vehicle, and the preset speed interval of the autonomous vehicle is acquired according to the maximum driving speed and the minimum driving speed.
For example, the running speed of the autonomous vehicle may be divided into a plurality of preset speed intervals in the calibration process, a preset speed interval corresponding to the chassis speed is selected in the running process, a target speed calibration model corresponding to the preset speed interval is obtained, and the chassis speed is calibrated based on the target speed calibration model. By selecting the preset speed interval for the chassis speed of the automatic driving vehicle in real time, the target calibration speed model of the preset speed interval can be obtained, and therefore the vehicle speed can be calibrated in real time according to the preset speed interval.
For example, for a first vehicle speed V 0 Selecting V 0 Selecting a first calibration coefficient and a second calibration coefficient corresponding to the preset speed interval, substituting the first calibration coefficient and the second calibration coefficient into a speed calibration model, and inputting V according to the speed calibration model 0 To obtain V 0 Calibration ofSpeed.
In step S202, a preset linear model is trained to obtain a speed calibration model in each preset speed interval, so as to solve the problem of inaccurate calibration of a target speed calibration model caused by a large span of original sample data. The speed calibration model capable of being more accurate is calibrated for the vehicle running speed through the preset speed interval.
And in each preset speed interval, acquiring a first vehicle speed meeting the running state of the preset automatic driving vehicle and a second vehicle speed corresponding to the first vehicle speed, and generating an original sample set corresponding to each preset speed interval.
In step S202, within a preset speed interval, initializing an error parameter of a preset linear model, including: in each preset speed interval, setting a first calibration coefficient of a preset linear model to be 1, and setting a second calibration coefficient of the preset linear model to be 0.
In step S202, in the embodiment of the invention, only the vehicle speed calibration method in the straight-line running state of the autonomous vehicle will be discussed.
S203, according to a first vehicle speed in a preset speed range and a second vehicle speed corresponding to the first vehicle speed, a training sample set of a preset linear model in a preset speed range is obtained.
In one embodiment, generating the original sample set corresponding to each preset speed interval includes: and in each preset speed interval, acquiring a first vehicle speed and a second vehicle speed corresponding to the first vehicle speed, and printing corresponding labels on the first vehicle speed and the second vehicle speed corresponding to the first vehicle speed.
For example, for an original sample < V1, V2, θ >, where < V1, V2> indicates whether a pair of nominal speeds needs to be determined, V1 is the first vehicle speed, V2 is the nominal speed corresponding to the first vehicle speed, θ is the corresponding tag obtained according to V1 and V2, and θ is usually represented by similarity.
Alternatively, the original samples < V1, V2>, < V1, V2> may be directly obtained according to the first vehicle speed and the second vehicle speed, which indicate whether the pair of calibration speeds needs to be determined, where V1 is the first vehicle speed, and V2 is the calibration speed corresponding to the first vehicle speed.
And S204, inputting the sample training set in the preset speed interval into a preset linear model, and performing model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval.
Specifically, inputting a sample training set in a preset speed interval into a preset linear model, and performing model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval, including: inputting a first vehicle speed to the preset linear model, and acquiring a fourth vehicle speed output by the preset linear model according to the first vehicle speed; calculating a value of a loss function of a preset linear model according to a fourth vehicle speed output by the first vehicle speed and a second vehicle speed corresponding to the first vehicle speed; and optimizing model error parameters of the preset linear model according to the value of the loss function value so as to obtain a vehicle speed calibration model corresponding to each preset speed interval.
In an embodiment, step S204 inputs the sample training set in the preset speed interval into a preset linear model, and performs model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval. Training the preset linear model includes: setting a loss function, inputting an original sample set into a speed calibration model to obtain a loss function value, performing iterative training on parameters of the speed calibration model through the loss function value, stopping training until the obtained loss function value converges to a preset threshold value, and obtaining the target speed calibration model as the speed calibration model. Specifically, as shown in step S301 to step S303 of fig. 3.
In one embodiment, a predetermined linear model is used for the velocity calibration model, and the predetermined linear model includes:
v＝k 1 *V 0 +b 1 ；
wherein, V 0 Is a first vehicle speed, v is a second vehicle speed corresponding to the first vehicle speed, k 1 For the first calibration factor of the velocity calibration model, b 1 And calibrating a second calibration coefficient of the model for the speed, wherein the model error parameter comprises the first calibration coefficient and the second calibration coefficient.
Step S301, initializing a preset linear model in each preset speed interval.
In one embodiment, the initializing the speed calibration model in each preset speed interval includes: in each preset speed interval, setting a first calibration coefficient of the speed calibration model as 1 and setting a second calibration coefficient of the speed calibration model as 0.
Step S302, inputting each first vehicle speed in the original sample set into a speed calibration model to obtain a loss function value of a second vehicle speed corresponding to each first vehicle speed;
specifically, in step S302, each original sample in the original sample set is obtained, a true similarity probability between a first vehicle speed and a second vehicle speed corresponding to the first vehicle speed in each original sample is obtained, and each original sample in the original sample set is input into the speed calibration model to obtain a similarity probability of each original sample; and calculating the value of the loss function of each original sample according to the similarity probability and the real similarity corresponding to each original sample.
For example, in each interval, the original sample < V1, V2> is input into the speed calibration model, and the true similarity label θ in the original sample < V1, V2> is obtained. The original samples < V1, V2, θ > may also be obtained directly. In the training process, each original sample < V1, V2> in the original sample set is input into the target model to obtain the probability (i.e. similarity probability) of being predicted as the label θ corresponding to each original sample, and then the value of the loss function is calculated according to the probability of being predicted as the label θ corresponding to each original sample and the true similarity θ.
And S303, optimizing the model error parameter of the speed calibration model according to the loss function value, and acquiring the speed calibration model corresponding to each preset speed interval according to the optimized model error parameter and the calibration model.
In step S303, a back propagation gradient of the velocity calibration model is obtained based on the value of the loss function obtained in step S302, and a model error parameter of the velocity calibration model is updated according to the gradient, so as to finally obtain a trained velocity calibration model, i.e., a target velocity calibration model.
In one embodiment, the maximum iteration number is set, the original sample set is input into the calibration model, a loss function corresponding to the chassis speed is obtained according to the chassis speed, a loss function value is obtained, and the model error parameter is updated according to the loss function value back propagation gradient, so that the trained speed calibration model is obtained. And obtaining parameters of the trained speed calibration model, and determining a target speed calibration model according to the parameters.
Specifically, the step S301 to the step S303 of obtaining the target calibration model by using the speed calibration model and the original sample set includes:
a) selecting a preset speed interval, giving an initialization model and an original sample, and then calculating output values of the model, wherein for example, the first input parameters are that the original sample is < V1, V2 and theta >, the V1 is a first vehicle speed, the V2 is a second vehicle speed, the theta is a similarity label between the V1 and the V2, and the output values are that the similarity probability of the V2 is obtained according to the V1 and a speed calibration model;
b) calculating the value of the loss function according to the output value of the model and the real similarity label;
c) according to the value back propagation gradient of the loss function, updating the error parameter of the model according to the gradient;
d) and giving iteration times, setting a similarity probability, and acquiring a model error parameter of a target model, wherein the model error parameter of which the output value is the preset similarity probability.
For example, the loss function value is obtained in step (b), the loss function value is updated to the initial number in step (c), then step b) is executed again, data in the initial data set are input step by step according to the preset iteration number, and the model is trained step by step until the iteration is carried out to the preset iteration number, so as to complete parameter updating.
In an embodiment, step S204, inputting the sample training set in the preset speed interval into a preset linear model, and performing model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval, further includes: and solving a first calibration coefficient and a second calibration coefficient of the speed calibration model in each preset speed interval according to the original sample set and the least square method.
For example, the original samples are collected within a preset time period, the original samples are classified according to a preset speed interval, and an original sample set corresponding to the speed of the vehicle in each interval is obtained, wherein the original sample set comprises a first speed and a second speed corresponding to the first speed. And in a preset speed interval, fitting the original sample set data set and the speed calibration model according to a least square method, and solving a first calibration coefficient and a second calibration coefficient of the target calibration model.
In an embodiment, the step S102 of obtaining a speed calibration model, training the speed calibration model according to an original sample set corresponding to each preset speed interval, to obtain a target speed calibration model corresponding to each preset speed interval, further includes: and acquiring a first calibration coefficient and a second calibration coefficient of the speed calibration model according to the optimization algorithm and the original number set.
Specifically, an original sample set is collected within a preset time period, the original sample set is classified according to preset speed intervals, an original sample set corresponding to each preset speed interval is obtained, and sample data is sorted according to the size of a calibration speed in the original sample set. And setting an objective function of the original sample set of the sample set and the speed calibration model. In the optimization algorithm, a minimum value of an objective function needs to be obtained, and in each iteration step, the objective function value is required to be decreased, that is, from an initial point, a maximum displacement s which can be trusted is assumed, and then in a trust region which takes the current point as a center and takes s as a radius, a true displacement is obtained by searching an optimal point of an approximate function (quadratic) of the objective function. After the displacement is obtained, calculating an objective function value, if the reduction of the objective function value meets a certain condition, the displacement is reliable, and continuing to carry out iterative calculation according to the rule; if the reduction of the objective function value can not meet a certain condition, the range of the confidence domain is reduced, and then the solution is carried out again.
For example, an objective function of the velocity calibration model is obtained, the original sample set is updated, the objective function is iterated step by step until a minimum value of the objective function is obtained in the trust domain, and at this time, parameters of the objective model can be solved according to the minimum value of the objective function and the velocity calibration model.
And step S104, acquiring a third vehicle speed, wherein the third vehicle speed is the vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle does not meet the first preset condition, and determining the calibration speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed.
As shown in FIG. 4, determining a calibrated speed of a third vehicle speed based on the vehicle speed calibration model and the third vehicle speed includes:
step S401: and determining a preset speed interval where the third vehicle speed is located.
And in the calibration process, the running speed of the automatic driving vehicle can be divided into a plurality of preset speed intervals, a preset speed interval corresponding to a third vehicle speed is selected in the running process, a target speed calibration model corresponding to the preset speed interval is obtained, and the third vehicle speed is calibrated based on the target speed calibration model. By selecting the preset speed interval for the third speed of the automatic driving vehicle in real time, the target calibration speed model of the preset speed interval can be obtained, so that the vehicle speed can be calibrated in real time according to the preset speed interval.
Step S402: and inputting the third vehicle speed into a speed calibration model corresponding to the preset speed interval to obtain a calibration speed corresponding to the third vehicle speed.
For example, for a third vehicle speed V 0 Selecting V 0 Corresponding preset speed interval, selecting a first calibration coefficient and a second calibration coefficient corresponding to the preset speed interval, substituting the first calibration coefficient and the second calibration coefficient into the target calibration model, and inputting V according to the target calibration model 0 To obtain V 0 The calibration speed of (2).
Corresponding to the embodiment of the application function implementation method, the application also provides a speed determination device of the automatic driving vehicle, electronic equipment and a corresponding embodiment.
Fig. 5 is a schematic structural diagram of a speed determination device of an autonomous vehicle according to an embodiment of the present application.
Referring to fig. 5, the speed determination device of an autonomous vehicle includes:
the first obtaining unit 501 is configured to obtain a first vehicle speed of the autonomous vehicle, where the first vehicle speed is a vehicle running speed obtained when an RTK differential signal quality of the autonomous vehicle meets a first preset condition and a heading angle of the autonomous vehicle meets a second preset condition, the first preset condition is a determination condition when the RTK differential signal quality of the autonomous vehicle is available, and the second preset condition is the heading angle of the autonomous vehicle in a straight-ahead state of the autonomous vehicle.
A second obtaining unit 502, configured to obtain a second vehicle speed of the autonomous vehicle, where the second vehicle speed is a calibrated speed corresponding to the first vehicle speed;
a training unit 503, configured to train a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, where the vehicle speed calibration model outputs a second vehicle speed corresponding to the first vehicle speed according to the first vehicle speed;
a calibration unit 504, configured to obtain a third vehicle speed, where the third vehicle speed is a vehicle running speed obtained when an RTK differential signal quality of an autonomous vehicle does not meet the first preset condition, and determine a calibration speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed.
In one embodiment, the apparatus further includes an RTK differential signal acquiring unit, and the RTK differential signal acquiring unit is configured to acquire an RTK differential signal that meets a preset condition, and determine a driving state of the autonomous vehicle according to the RTK differential signal. The RTK differential signal acquisition unit is used for judging that the automatic driving vehicle accords with the running state of the preset automatic driving vehicle, and the first acquisition unit is also used for acquiring the chassis speed meeting the running state of the preset automatic driving vehicle and the calibration speed corresponding to the chassis speed meeting the running state of the preset automatic driving vehicle in each preset speed interval and generating an original sample set corresponding to each preset speed interval.
In one embodiment, the device further comprises a parameter calibration unit, wherein the parameter calibration unit is used for obtaining the model error parameter of the vehicle speed calibration model and adjusting the parameter value of the inertia measurement unit according to the model error parameter of the vehicle speed calibration model.
The technical scheme provided by the application can comprise the following beneficial effects: on one hand, because the original sample set used for training the speed calibration model is obtained when the RTK differential signal of the automatic driving vehicle accords with the preset condition, when the RTK differential signal is not good, the accurate calibration of the chassis speed of the vehicle can be realized based on the trained speed calibration model; on the other hand, each preset speed interval corresponds to the original sample set, so that the problem that the target speed calibration model is not accurately calibrated due to the fact that the sample data span is large can be solved, namely, through the subdivided original sample sets, more accurate training samples can be provided for the speed calibration model, and therefore the target calibration model obtained accordingly can be used for accurately calibrating the vehicle running speed.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 6 is a schematic structural diagram of an electronic device shown in an embodiment of the present application.
Referring to fig. 6, an electronic device 600 includes a memory 610 and a processor 620.
The Processor 620 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 610 may include various types of storage units such as system memory, Read Only Memory (ROM), and permanent storage. Wherein the ROM may store static data or instructions that are required by the processor 620 or other modules of the computer. The persistent storage device may be a read-write storage device. The persistent storage may be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off. In some embodiments, the persistent storage device employs a mass storage device (e.g., magnetic or optical disk, flash memory) as the persistent storage device. In other embodiments, the permanent storage may be a removable storage device (e.g., floppy disk, optical drive). The system memory may be a read-write memory device or a volatile read-write memory device, such as a dynamic random access memory. The system memory may store instructions and data that some or all of the processors require at runtime. In addition, the memory 610 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic and/or optical disks, as well. In some embodiments, memory 610 may include a removable storage device that is readable and/or writable, such as a Compact Disc (CD), a digital versatile disc read only (e.g., DVD-ROM, dual layer DVD-ROM), a Blu-ray disc read only, an ultra-dense disc, a flash memory card (e.g., SD card, min SD card, Micro-SD card, etc.), a magnetic floppy disk, or the like. Computer-readable storage media do not contain carrier waves or transitory electronic signals transmitted by wireless or wired means.
The memory 610 has stored thereon executable code that, when processed by the processor 620, may cause the processor 620 to perform some or all of the methods described above.
Furthermore, the method according to the present application may also be implemented as a computer program or computer program product comprising computer program code instructions for performing some or all of the steps of the above-described method of the present application.
Alternatively, the present application may also be embodied as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) having executable code (or a computer program or computer instruction code) stored thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the various steps of the above-described method according to the present application.
Having described embodiments of the present application, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or improvements to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (10)
1. A method of determining a speed of an autonomous vehicle, comprising:
acquiring a first vehicle speed of an automatic driving vehicle, wherein the first vehicle speed is a vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle meets a first preset condition and the course angle of the automatic driving vehicle meets a second preset condition, the first preset condition is a judgment condition when the RTK differential signal quality of the automatic driving vehicle is available, and the second preset condition is the course angle of the automatic driving vehicle in a straight running state;
acquiring a second vehicle speed of the automatic driving vehicle, wherein the second vehicle speed is a calibration speed corresponding to the first vehicle speed;
training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, wherein the vehicle speed calibration model is used for receiving the first vehicle speed and outputting a second vehicle speed corresponding to the first vehicle speed;
and acquiring a third vehicle speed, wherein the third vehicle speed is the vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle does not meet the first preset condition, and determining the calibration speed of the third vehicle speed based on the vehicle speed calibration model and the third vehicle speed.
2. The method according to claim 1, wherein the first vehicle speed is a vehicle chassis speed measured by a vehicle speed sensor of an autonomous vehicle, the second vehicle speed is a vehicle traveling speed calculated from an RTK differential signal corresponding to the first vehicle speed, and a time of the second vehicle speed is the same as a time of the first vehicle speed corresponding to the second vehicle speed.
3. The method of claim 1, wherein training a predetermined linear model based on the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model comprises:
determining a model error parameter of the preset linear model according to the speed error coefficient;
setting a preset speed interval, and initializing a model error parameter of the preset linear model in the preset speed interval;
acquiring a training sample set of the preset linear model in a preset speed interval according to a first vehicle speed in the preset speed interval and a second vehicle speed corresponding to the first vehicle speed;
and inputting the sample training set in the preset speed interval into the preset linear model, and performing model error parameter optimization to obtain a vehicle speed calibration model corresponding to each preset speed interval.
4. The method according to claim 3, wherein the inputting the sample training set in the preset speed interval into the preset linear model for model error parameter optimization to obtain the vehicle speed calibration model corresponding to each preset speed interval comprises:
inputting the first vehicle speed to the preset linear model, and acquiring a fourth vehicle speed output by the preset linear model according to the first vehicle speed;
calculating a value of a loss function of the preset linear model according to a fourth vehicle speed output by the first vehicle speed and a second vehicle speed corresponding to the first vehicle speed;
and optimizing the model error parameters of the preset linear model according to the value of the loss function value so as to obtain a vehicle speed calibration model corresponding to each preset speed interval.
5. The method of claim 3, wherein the model error coefficient comprises a first calibration coefficient and a second calibration coefficient, and the setting the model error parameter of the preset linear model according to the speed error coefficient comprises:
acquiring the speed error coefficient, and setting a first calibration coefficient and a second calibration coefficient of the preset linear model according to the error coefficient of the automatic driving vehicle, wherein the speed error coefficient is an error coefficient between the first vehicle speed and the second vehicle speed;
and obtaining the product of the first calibration coefficient and the first vehicle speed, and determining the preset linear model according to the product of the first calibration coefficient and the first vehicle speed and the second calibration coefficient.
6. The method according to claim 4, wherein initializing the error parameters of the preset linear model within a preset speed interval comprises:
and in each preset speed interval, setting a first calibration coefficient of the preset linear model to be 1, and setting a second calibration coefficient of the preset linear model to be 0.
7. The method of claim 1, wherein said determining a calibrated speed for said third vehicle speed based on said vehicle speed calibration model and said third vehicle speed comprises:
determining the preset speed interval where the third vehicle speed is located;
and inputting the third vehicle speed into a speed calibration model corresponding to the preset speed interval to obtain a calibration speed corresponding to the third vehicle speed.
8. A speed determination apparatus of an autonomous vehicle, comprising:
the automatic driving vehicle control system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring a first vehicle speed of an automatic driving vehicle, the first vehicle speed is a vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle meets a first preset condition and the course angle of the automatic driving vehicle meets a second preset condition, the first preset condition is a judgment condition when the RTK differential signal quality of the automatic driving vehicle is available, and the second preset condition is the course angle of the automatic driving vehicle in a straight-ahead state;
the second obtaining unit is used for obtaining a second vehicle speed of the automatic driving vehicle, and the second vehicle speed is a calibration speed corresponding to the first vehicle speed;
the training unit is used for training a preset linear model according to the first vehicle speed and the second vehicle speed to obtain a vehicle speed calibration model, and the vehicle speed calibration model outputs a second vehicle speed corresponding to the first vehicle speed according to the first vehicle speed;
and the calibration unit is used for acquiring a third vehicle speed, wherein the third vehicle speed is the vehicle running speed obtained under the condition that the RTK differential signal quality of the automatic driving vehicle does not meet the first preset condition, and the calibration speed of the third vehicle speed is determined based on the vehicle speed calibration model and the third vehicle speed.
9. An electronic device, comprising:
a processor; and
a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method of any one of claims 1-7.
10. A computer-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to perform the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210663210.3A CN114919590A (en) | 2022-06-13 | 2022-06-13 | Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210663210.3A CN114919590A (en) | 2022-06-13 | 2022-06-13 | Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114919590A true CN114919590A (en) | 2022-08-19 |
Family
ID=82814093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210663210.3A Pending CN114919590A (en) | 2022-06-13 | 2022-06-13 | Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114919590A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115293301A (en) * | 2022-10-09 | 2022-11-04 | 腾讯科技（深圳）有限公司 | Estimation method and device for lane change direction of vehicle and storage medium |
-
2022
- 2022-06-13 CN CN202210663210.3A patent/CN114919590A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115293301A (en) * | 2022-10-09 | 2022-11-04 | 腾讯科技（深圳）有限公司 | Estimation method and device for lane change direction of vehicle and storage medium |
| CN115293301B (en) * | 2022-10-09 | 2023-01-31 | 腾讯科技（深圳）有限公司 | Estimation method and device for lane change direction of vehicle and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101214474B1 (en) | Navigation apparatus and driving route information offering method using by it, automatic driving system and its method | |
| CN111077555B (en) | Positioning method and device | |
| WO2020184013A1 (en) | Vehicle control device | |
| WO2012062339A1 (en) | Technique for calibrating dead reckoning positioning data | |
| CN114624741A (en) | Positioning accuracy evaluation method and device | |
| CN111982158B (en) | Inertial measurement unit calibration method and device | |
| CN114919590A (en) | Method and device for determining speed of automatic driving vehicle, electronic equipment and storage medium | |
| CN109188478A (en) | A kind of urban environment satellite positioning navigation method and device assisted using map | |
| JP2008008628A (en) | Apparatus for determining position of self-vehicle | |
| WO2021033312A1 (en) | Information output device, automated driving device, and method for information output | |
| CN113917512B (en) | Positioning method and device for automatic driving vehicle, electronic equipment and storage medium | |
| CN110967023B (en) | Automobile positioning method and automobile positioning device | |
| CN113375668B (en) | Antenna installation angle calibration method and device of satellite navigation system | |
| US20220042802A1 (en) | Extended dead reckoning accuracy | |
| CN114088102A (en) | Road surface quality evaluation method and device based on vehicle-mounted inertia measurement unit | |
| JP7209912B2 (en) | DRIVING ASSISTANCE CONTROL DEVICE AND DRIVING ASSISTANCE CONTROL METHOD | |
| CN113280813B (en) | Inertial measurement data compensation method and device based on neural network model | |
| JP2018181301A (en) | Information processing method and information processing apparatus | |
| CN113763483B (en) | Method and device for calibrating pitch angle of automobile data recorder | |
| CN114623839A (en) | Navigation method, system, device and storage medium | |
| JP2008241592A (en) | Navigation device, navigation method, and navigation program | |
| JP7064001B2 (en) | Vehicle control unit | |
| CN114248782B (en) | Unmanned vehicle pitch angle determination method, unmanned vehicle pitch angle determination device and computer readable storage medium | |
| CN116559899B (en) | Fusion positioning method and device for automatic driving vehicle and electronic equipment | |
| CN113895446B (en) | Method, device, equipment and storage medium for determining speed of automatic driving vehicle |
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 |