CN114084154B - Automatic driving system parameter configuration method, device and system - Google Patents
Automatic driving system parameter configuration method, device and system Download PDFInfo
- Publication number
- CN114084154B CN114084154B CN202111173647.0A CN202111173647A CN114084154B CN 114084154 B CN114084154 B CN 114084154B CN 202111173647 A CN202111173647 A CN 202111173647A CN 114084154 B CN114084154 B CN 114084154B
- Authority
- CN
- China
- Prior art keywords
- preset
- user
- multidimensional
- information
- automatic driving
- 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 27
- 239000013598 vector Substances 0.000 claims abstract description 179
- 238000011156 evaluation Methods 0.000 claims abstract description 92
- 238000005457 optimization Methods 0.000 claims description 52
- 238000012545 processing Methods 0.000 claims description 22
- 238000006243 chemical reaction Methods 0.000 claims description 15
- 230000005484 gravity Effects 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000007635 classification algorithm Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 description 14
- 230000006872 improvement Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000013507 mapping Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013210 evaluation model Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000033764 rhythmic process Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- 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
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0002—Automatic control, details of type of controller or control system architecture
- B60W2050/0004—In digital systems, e.g. discrete-time systems involving sampling
- B60W2050/0005—Processor details or data handling, e.g. memory registers or chip architecture
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a configuration method, a device and a system of parameters of an automatic driving system, which are characterized in that firstly, riding evaluation information of a user is collected, then the riding evaluation information is used as compensation quantity of a preset multidimensional vector, the preset multidimensional vector is corrected to generate a corresponding multidimensional vector, the multidimensional vector is classified, and the system parameters of the automatic driving system are optimized according to the classified multidimensional vector. According to the technical scheme, the personalized configuration accuracy of the parameters of the automatic driving system can be improved, so that the automatic driving experience of a user is improved.
Description
Technical Field
The present invention relates to the field of autopilot technologies, and in particular, to a method, an apparatus, and a system for configuring parameters of an autopilot system.
Background
In the field of autopilot, a large number of system parameters are required to be configured for the vehicle, which parameters are set according to the experience of the engineer. But in the context of autopilot landing, different users have the same or similar view of all parameters as the engineer, and depend on the subjective perception of the passenger in some personalized choices or ride experiences. In order to ensure that the automatic driving can fulfill the function of safely conveying passengers, the prior art only allows users to manually configure certain system parameters, and the user has rough parameter configuration, cannot reflect the actual feeling of the users, and cannot meet the requirements of optimizing riding or driving experience.
Disclosure of Invention
The invention provides a configuration method, a device and a system for parameters of an automatic driving system, which can improve the personalized configuration accuracy of the parameters of the automatic driving system, thereby improving the automatic driving experience of a user.
In order to solve the above technical problems, the present invention provides a method for configuring parameters of an autopilot system, including:
acquiring riding evaluation information of a preset user on an automatic driving system; wherein the preset user corresponds to one or more pieces of riding evaluation information;
taking each piece of riding evaluation information as a compensation quantity of a preset multidimensional vector, correcting the preset multidimensional vector, and generating a multidimensional vector corresponding to each piece of riding evaluation information;
classifying the generated multiple multidimensional vectors according to a preset classification algorithm, and determining multiple first multidimensional vectors corresponding to the preset user;
and optimizing the automatic driving system parameters of the preset user according to the first multidimensional vectors.
As an improvement of the above solution, the acquiring the riding evaluation information of the automatic driving system by the preset user specifically includes:
acquiring multiple pieces of riding evaluation information of the same user on the automatic driving system;
alternatively, one or more ride assessment information for the autopilot system by different users is obtained.
As an improvement of the above solution, the classifying, according to a preset classification algorithm, the generated multiple multidimensional vectors, and determining the multiple first multidimensional vectors corresponding to the preset user specifically includes:
when the preset user is the same user, center of gravity processing is carried out on the generated multiple multidimensional vectors, the category of the preset user is determined, and multiple first multidimensional vectors corresponding to the preset user are determined according to the determined category;
when the preset users are different users, clustering or clustering and then barycenter processing are carried out on the generated multiple multidimensional vectors, the group of the preset users is determined, and the multiple first multidimensional vectors corresponding to the preset users are determined according to the determined group.
As an improvement of the above solution, the optimizing the autopilot system parameter of the preset user according to the plurality of first multidimensional vectors specifically includes:
when the preset user is the same user, generating first optimization information corresponding to the preset user according to the configuration information corresponding to the determined category and combining the plurality of first multidimensional vectors, and optimizing the automatic driving system parameters of the preset user according to the first optimization information;
when the preset users are different users, generating second optimization information corresponding to the groups according to the configuration information corresponding to the determined groups and combining the plurality of first dimension vectors, and optimizing automatic driving system parameters corresponding to the groups according to the second optimization information;
the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
As an improvement of the above solution, before the clustering processing is performed on the generated multiple multidimensional vectors, the method further includes:
denoising the generated multidimensional vectors, and removing abnormal data which contain noise points and error information in the multidimensional vectors of which the change amount of the two multidimensional vectors of the same user is larger than a preset threshold value.
As an improvement of the above solution, the method for obtaining the riding evaluation information includes one or more of the following combinations, specifically:
acquiring historical configuration information of a first user on the automatic driving system, and generating riding evaluation information corresponding to the first user according to the historical configuration information;
or generating one piece of riding evaluation information of the second user in a trigger scene according to a plurality of pieces of voice configuration information of the second user in the trigger scene under a preset condition;
or, in response to a parameter configuration operation input by a third user, adjusting a current system parameter of the automatic driving system, and generating one piece of riding evaluation information corresponding to the third user according to the adjusted system parameter.
As an improvement of the above solution, the correcting the preset multidimensional vector by using each piece of riding evaluation information as a compensation amount of the preset multidimensional vector to generate a multidimensional vector corresponding to each piece of riding evaluation information, specifically:
when the riding evaluation information is generated according to the plurality of voice configuration information, voice recognition is respectively carried out on each voice configuration information through a preset NLP model, each voice recognition result is used as a compensation quantity of a preset multidimensional vector to correct the preset multidimensional vector, and one multidimensional vector corresponding to each riding evaluation information is generated.
Correspondingly, the invention provides a configuration device of the parameters of the automatic driving system, which comprises the following components: the device comprises an acquisition module, a conversion module, a classification module and an optimization module;
the acquisition module is used for acquiring riding evaluation information of a preset user on the automatic driving system; wherein the preset user corresponds to one or more pieces of riding evaluation information;
the conversion module is used for taking each piece of riding evaluation information as a compensation quantity of a preset multidimensional vector, correcting the preset multidimensional vector and generating a multidimensional vector corresponding to each piece of riding evaluation information;
the classifying module is used for classifying the generated multiple multidimensional vectors according to a preset classifying algorithm and determining multiple first multidimensional vectors corresponding to the preset user;
the optimization module is used for optimizing the automatic driving system parameters of the preset user according to the first multidimensional vectors.
As an improvement of the above solution, the obtaining module is configured to obtain riding evaluation information of a preset user on the autopilot system, specifically:
the acquisition module acquires a plurality of pieces of riding evaluation information of the same user on the automatic driving system;
alternatively, the acquisition module acquires one or more ride assessment information for the autopilot system for different users.
As an improvement of the above solution, the classification module is configured to classify the generated multiple multidimensional vectors according to a preset classification algorithm, and determine multiple first multidimensional vectors corresponding to the preset user, where the multiple first multidimensional vectors are specifically:
when the preset user is the same user, the classification module performs gravity center solving on the generated multiple multidimensional vectors, determines the category of the preset user, and determines multiple first multidimensional vectors corresponding to the preset user according to the determined category;
when the preset users are different users, the classification module performs clustering processing or clustering processing and gravity center solving processing on the generated multiple multidimensional vectors, determines the group of the preset users, and determines multiple first multidimensional vectors corresponding to the preset users according to the determined group.
As an improvement of the above solution, the optimizing module is configured to optimize, according to the plurality of first multidimensional vectors, an autopilot system parameter of the preset user, specifically:
when the preset user is the same user, the optimization module combines the plurality of first multidimensional vectors according to the configuration information corresponding to the determined category to generate first optimization information corresponding to the preset user, and optimizes the automatic driving system parameters of the preset user according to the first optimization information;
when the preset users are different users, the optimization module generates second optimization information corresponding to the groups by combining the plurality of first dimension vectors according to the configuration information corresponding to the determined groups, and optimizes the automatic driving system parameters corresponding to the groups according to the second optimization information;
the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
As an improvement of the above solution, the configuration device further includes: a preprocessing module;
the preprocessing module is used for denoising the generated multidimensional vectors before the classifying module performs clustering processing on the generated multidimensional vectors, and eliminating abnormal data which contain noise points and error information in the multidimensional vectors of which the two multidimensional vector variable quantities of the same user are larger than a preset threshold value.
As an improvement of the above solution, the conversion module is configured to take each piece of riding evaluation information as a compensation amount of a preset multidimensional vector, correct the preset multidimensional vector, and generate a multidimensional vector corresponding to each piece of riding evaluation information, where the method specifically includes:
when the riding evaluation information is generated according to the historical configuration information, the conversion module respectively carries out voice recognition on each voice configuration information through a preset NLP model, and corrects the preset multidimensional vector by taking each voice recognition result as a compensation quantity of the preset multidimensional vector to generate one multidimensional vector corresponding to each riding evaluation information.
Correspondingly, the invention also provides an automatic driving system, which comprises: the automatic driving device comprises automatic driving equipment, a cloud server and the automatic driving system parameter configuration device;
wherein the acquisition module is configured on the autopilot device;
the conversion module is configured on the automatic driving equipment or the cloud server;
the classification module and the optimization module are both configured on the cloud server.
The embodiment of the invention has the following beneficial effects:
the invention provides a configuration method, a device and a system for parameters of an automatic driving system, which are characterized in that firstly, riding evaluation information of a user is collected, then the riding evaluation information is used as compensation quantity of a preset multidimensional vector, the preset multidimensional vector is corrected to generate a corresponding multidimensional vector, the multidimensional vector is classified, and the system parameters of the automatic driving system are optimized according to the classified multidimensional vector. Compared with the prior art that the real experience of the user cannot be reflected and the requirement of optimizing riding or driving experience cannot be met, the invention can evaluate the riding experience of the user to optimize the next automatic driving riding, not only can improve the automatic driving experience of the user, but also can improve the personalized configuration accuracy of the parameters of an automatic driving system.
Drawings
FIG. 1 is a flow chart of one embodiment of a method for configuring autopilot system parameters provided by the present invention;
FIG. 2 is a schematic diagram illustrating an embodiment of an autopilot system parameter configuration apparatus according to the present invention;
fig. 3 is a schematic structural diagram of an embodiment of a terminal device provided by the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made more apparent and fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flow chart of an embodiment of a method for configuring parameters of an autopilot system according to the present invention, as shown in fig. 1, the method includes steps 101 to 104, and the steps are specifically as follows:
step 101: acquiring riding evaluation information of a preset user on an automatic driving system; wherein the preset user corresponds to one or more pieces of riding evaluation information.
In the present embodiment, the riding assessment information may be obtained by, but is not limited to, combining one or more of the following: acquiring historical configuration information of a first user on the automatic driving system, and generating riding evaluation information corresponding to the first user according to the historical configuration information; or generating one piece of riding evaluation information of the second user in the triggering scene according to a plurality of pieces of voice configuration information of the second user in the triggering scene under the preset condition; or, in response to the parameter configuration operation input by the third user, adjusting the current system parameter of the automatic driving system, and generating one piece of riding evaluation information corresponding to the third user according to the adjusted system parameter. The first user, the second user and the third user may be the same user or different users.
In this embodiment, for the first acquisition mode, the application scenario is that the user is an old user, and the system stores the historical configuration information of the user. Historical configuration information is associated with the user account and may be invoked, but not limited to, from a remote server, as is system parameters of the autopilot system, such as acceleration, deceleration, lane change aggressiveness, lane change comfort, and the like. The passenger can log in the user account before sitting by means of keys, face recognition and the like, and load/add/modify corresponding historical configuration information.
As an example of the present embodiment, the history arrangement information of the user can reflect the riding evaluation of the user or be generated from the riding evaluation information of the last riding, so that the riding evaluation of the user can be reversely obtained from the history arrangement information of the user at the time of data acquisition.
In this embodiment, for the second mode, the application scenario is a preset trigger scenario in which emergency braking, emergency driving direction, deviation from the original planned route, etc. occur in automatic driving, the system pops up voice, and accesses the optimization suggestion of the passenger about whether this action or decision is understood, reasonable or relevant. After the passenger inputs the voice evaluation, the system generates voice configuration information such as information of a slow turning point, a fast lane changing point and the like in the scene. The system generates riding evaluation information corresponding to the riding experience of the user after voice configuration information in a plurality of trigger scenes is collected and integrated. The trigger scene can be designed according to actual needs or actual conditions so as to be suitable for various automatic driving systems.
In this embodiment, for the third mode, the application scenario is that the user experiences autopilot for the first time, and the user may fill in a corresponding form or input corresponding configuration information through an interactive interface, etc. to adjust the current system parameters of the autopilot system. And generating riding evaluation information of the user by the system according to the current system parameters.
In this embodiment, step 101 may obtain multiple pieces of riding evaluation information of the same user on the autopilot system, or obtain one or more pieces of riding evaluation information of different users on the autopilot system. In this embodiment, when the data source is acquired, the user may be initially screened, the data of the same user is used to optimize the own autopilot system parameter, and the data of different users is used to optimize the overall autopilot system parameter.
Step 102: and correcting the preset multidimensional vector by taking each piece of riding evaluation information as the compensation quantity of the preset multidimensional vector, and generating one multidimensional vector corresponding to each piece of riding evaluation information.
In this embodiment, the obtained riding evaluation information may be a specific configuration value of a system parameter, voice configuration information of a passenger, or both. At this time, before executing step 102, these data need to be converted into compensation amounts of multidimensional vectors according to a preset mapping relationship. The mapping relationship may be, but is not limited to, parameter configuration values, voice configuration information, and a mutual mapping relationship between multiple dimension vectors, and is used for converting data.
As an example of this embodiment, step 102 specifically includes: when the riding evaluation information is generated according to the plurality of voice configuration information, voice recognition is respectively carried out on each voice configuration information through a preset NLP model, each voice recognition result is used as a compensation quantity of a preset multidimensional vector to correct the preset multidimensional vector, and one multidimensional vector corresponding to each riding evaluation information is generated.
As an example of this embodiment, step 102 specifically includes: when the riding evaluation information simultaneously comprises voice configuration information and system parameter configuration information, the system parameter configuration information and the voice configuration information are respectively converted into a first multidimensional vector and a second multidimensional vector according to the mapping relation, the first multidimensional vector is used as a compensation quantity, the second multidimensional vector is used as a preset multidimensional vector, and therefore the first multidimensional vector and the second multidimensional vector are subjected to weighted superposition to generate a multidimensional vector.
Step 103: classifying the generated multiple multidimensional vectors according to a preset classification algorithm, and determining multiple first multidimensional vectors corresponding to a preset user.
In this embodiment, step 103 specifically includes: when the preset user is the same user, center of gravity processing is carried out on the generated multiple multidimensional vectors, the category of the preset user is determined, and multiple first multidimensional vectors corresponding to the preset user are determined according to the determined category; when the preset user is different users, clustering processing is carried out on the generated multiple multidimensional vectors or clustering is carried out first and then gravity center processing is carried out, the group of the preset user is determined, and according to the determined group, multiple first multidimensional vectors corresponding to the preset user are determined.
In this embodiment, the categories and groups are custom defined by the vehicle manufacturer, and the categories may include, but are not limited to, sports, comfort, or economy categories. If the comfort type indicates the comfort level of the passenger for priority to pay attention to the automatic driving, the system parameters in the type are preferably configured by taking the comfort level as the direction, and the other types are the same. The groups may include, but are not limited to, groups divided according to different rules, such as different geographic locations, different environmental characteristics, different time periods, working days and non-working days, different weather, different road conditions, different urban rhythms, and the like.
In this embodiment, if the collected data originates from the same user, the center of gravity of the generated multiple multidimensional vectors is calculated, the category of the user is determined, and then multiple first multidimensional vectors corresponding to the user are determined. In this step, the first multidimensional vector may be, but is not limited to, a plurality of multidimensional vectors generated for the same user, or a plurality of multidimensional vectors are selected from the plurality of multidimensional vectors generated. The screening rule can be set according to the category of the user, such as a sport type category screening multi-dimensional vector capable of reflecting sensitivity evaluation, so that the accuracy of configuration is improved for different user groups.
As an example of this embodiment, if the riding evaluation information corresponds to one riding experience of the passenger, the data amount is not large, and the matching degree calculation may be performed between a plurality of multidimensional vectors and vectors set in each category, and then the evaluation category with the highest matching degree may be used as the category of the user, thereby improving the classification speed. The matching degree calculating method is the prior art and is not described herein.
In this embodiment, if the collected data originate from different users, the generated multiple multidimensional vectors are clustered or clustered first and then barycenter processed, a group in which the preset user is located is determined, and then multiple first multidimensional vectors corresponding to the preset user are determined. Because the division rules of the groups have differences, the data acquisition amount of some groups is larger, but the data acquisition amount of some groups is not large, therefore, the user groups can be determined by using only clustering according to actual conditions, or the center of gravity is firstly obtained by clustering under the condition of larger data amount, and the accuracy of the group division is improved. In the clustering, the calculation may be performed by, but not limited to, machine learning, neural network, or the like.
In this embodiment, the specific calculation manner of clustering and barycenter calculation is the prior art, and will not be described herein.
In this embodiment, one multidimensional vector corresponds to one data element, and before cluster calculation, denoising processing can be performed on the generated multidimensional vectors, so that abnormal data containing noise points and error information in multidimensional vectors with the variation of two multidimensional vectors of the same user being greater than a preset threshold value are removed. The passenger's twice riding needs are greatly changed due to reasons such as false touch, drunk, unstable emotion and the like, and the original data become unreliable due to various reasons such as data transmission errors, data analysis errors and the like, so that the original data are required to be cleaned, abnormal data or noise points are removed, and effective data are reserved. Secondly, due to various reasons of weather, road conditions, urban rhythm and the like, data are required to be classified or pruned so as to achieve more targeted parameter configuration for training.
Step 104: and optimizing the automatic driving system parameters of the preset user according to the first multidimensional vectors.
In this embodiment, step 104 specifically includes: when the preset user is the same user, according to the configuration information corresponding to the determined category, combining a plurality of first multidimensional vectors to generate first optimization information corresponding to the preset user, and according to the first optimization information, optimizing the automatic driving system parameters of the preset user; when the preset users are different users, generating second optimization information corresponding to the groups by combining a plurality of first dimension vectors according to the configuration information corresponding to the determined groups, and optimizing automatic driving system parameters corresponding to the groups according to the second optimization information; the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
In this embodiment, step 104 may perform configuration of system parameters for a single user, or may perform configuration of system parameters for an entire group, where the system parameters of the group may be used for configuration of other users. If the group is the system parameters of the early peak, the system parameters of the group are optimized by collecting the riding evaluation information of different users in the same time period. When the system parameters of a certain user in the early peak are required to be configured, the system parameters of the group can be called as basic data, and the riding evaluation information of the user is used as compensation quantity for correction, so that the final system parameters of the user are obtained.
As an example of the embodiment, the acquired riding evaluation information and system parameters can be used as training data of an automatic driving evaluation model, so that training of the evaluation model can be more directed to user demand convergence on the premise of safety.
As an example of this embodiment, the configuration method of the present invention may be completed by a vehicle-mounted terminal, or may be responsible for data acquisition by the vehicle-mounted terminal, and the operation of the configuration method may be completed by a background server. Steps 102 to 104 are completed by the background server as step 101 is implemented by the in-vehicle terminal, or steps 101 and 102 are implemented by the in-vehicle terminal, and steps 103 and 104 are completed by the background server.
Accordingly, referring to fig. 2, fig. 2 is a schematic structural diagram of an embodiment of an autopilot system parameter configuration apparatus provided in the present invention. The configuration device comprises: an acquisition module 201, a conversion module 202, a classification module 203 and an optimization module 204.
The acquiring module 201 is configured to acquire riding evaluation information of a preset user on the autopilot system; wherein the preset user corresponds to one or more pieces of riding evaluation information.
The conversion module 202 is configured to take each piece of riding evaluation information as a compensation amount of a preset multidimensional vector, correct the preset multidimensional vector, and generate a multidimensional vector corresponding to each piece of riding evaluation information.
The classification module 203 is configured to classify the generated multiple multidimensional vectors according to a preset classification algorithm, and determine multiple first multidimensional vectors corresponding to the preset user.
The optimizing module 204 is configured to optimize the autopilot system parameters of the preset user according to the plurality of first multidimensional vectors.
In this embodiment, the obtaining module 201 is configured to obtain riding evaluation information of an autopilot system by a preset user, specifically: the acquisition module acquires 201 a plurality of pieces of riding evaluation information of the same user on the automatic driving system; alternatively, the acquisition module 201 acquires one or more ride assessment information for the autopilot system for different users.
In this embodiment, the classification module 203 is configured to classify the generated multiple multidimensional vectors according to a preset classification algorithm, and determine multiple first multidimensional vectors corresponding to the preset user, which is specifically:
when the preset user is the same user, the classification module 203 performs gravity center solving on the generated multiple multidimensional vectors, determines the category of the preset user, and determines multiple first multidimensional vectors corresponding to the preset user according to the determined category; when the preset user is a different user, the classification module 203 performs clustering processing or clustering-first and then barycenter processing on the generated multiple multidimensional vectors, determines a group in which the preset user is located, and determines multiple first multidimensional vectors corresponding to the preset user according to the determined group.
As an example of the present embodiment, the configuration device of the autopilot system parameters further includes: and a preprocessing module. The preprocessing module is configured to, before the classifying module 203 performs clustering processing on the generated multiple multidimensional vectors, perform denoising processing on the generated multiple multidimensional vectors, and reject abnormal data containing noise points and error information in the multidimensional vectors, where the variation of the two multidimensional vectors of the same user is greater than a preset threshold.
In this embodiment, the optimizing module 204 is configured to optimize, according to the plurality of first multidimensional vectors, parameters of an autopilot system of a preset user, specifically:
when the preset user is the same user, the optimization module 204 combines the plurality of first multidimensional vectors according to the configuration information corresponding to the determined category to generate first optimization information corresponding to the preset user, and optimizes the autopilot system parameters of the preset user according to the first optimization information; when the preset user is a different user, the optimization module 204 combines the plurality of first dimension vectors according to the configuration information corresponding to the determined group, generates second optimization information corresponding to the group, and optimizes the autopilot system parameter corresponding to the group according to the second optimization information; the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
In this embodiment, the conversion module 202 is configured to take each piece of riding evaluation information as a compensation amount of a preset multidimensional vector, correct the preset multidimensional vector, and generate a multidimensional vector corresponding to each piece of riding evaluation information, which specifically is: when the riding evaluation information is generated according to the historical configuration information, the conversion module 202 performs voice recognition on each voice configuration information through a preset NLP model, and corrects the preset multidimensional vector by taking each voice recognition result as a compensation amount of the preset multidimensional vector to generate one multidimensional vector corresponding to each riding evaluation information.
Correspondingly, the embodiment of the invention also provides an automatic driving system, which comprises: the automatic driving device, the cloud server and the automatic driving system parameter configuration device are used for configuring the automatic driving system parameters; wherein the acquisition module 201 is configured on the autopilot device; the conversion module 202 is configured on the autopilot device or the cloud server; the classification module 203 and the optimization module 204 are both configured on the cloud server.
From the above, the invention provides a configuration method, a device and a system for parameters of an automatic driving system, which are characterized in that firstly, riding evaluation information of a user is collected, then the riding evaluation information is used as compensation quantity of a preset multidimensional vector, the preset multidimensional vector is corrected to generate a corresponding multidimensional vector, the multidimensional vector is classified, and the system parameters of the automatic driving system are optimized according to the classified multidimensional vector. Compared with the prior art that the real experience of the user cannot be reflected and the requirement of optimizing riding or driving experience cannot be met, the invention can evaluate the riding experience of the user to optimize the next automatic driving riding, not only can improve the automatic driving experience of the user, but also can improve the personalized configuration accuracy of the parameters of an automatic driving system.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an embodiment of a terminal device provided by the present invention.
The terminal device provided by the embodiment of the invention comprises a processor 71, a memory 72 and a computer program stored in the memory 72 and configured to be executed by the processor 71, wherein the configuration method of the autopilot system parameters is realized when the processor 71 executes the computer program.
In addition, an embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium includes a stored computer program, where when the computer program runs, a device where the computer readable storage medium is controlled to execute the method for configuring the parameters of the autopilot system according to any one of the embodiments above.
It will be appreciated by those skilled in the art that the schematic diagram is merely an example of a terminal device and does not constitute a limitation of the terminal device, and may include more or less components than illustrated, or may combine certain components, or different components, e.g., the terminal device may further include an input-output device, a network access device, a bus, etc.
The processor 71 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, and the processor 71 is a control center of the terminal device, and connects various parts of the entire terminal device using various interfaces and lines.
The memory 72 may be used to store the computer program and/or module, and the processor 71 implements various functions of the terminal device by running or executing the computer program and/or module stored in the memory 72 and invoking data stored in the memory 72. The memory 72 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the terminal device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
Wherein the terminal device integrated modules/units may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as stand alone products. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Those skilled in the art will appreciate that implementing all or part of the above-described embodiments may be accomplished by way of computer programs, which may be stored on a computer readable storage medium, which when executed may comprise the steps of the above-described embodiments. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like.
Claims (13)
1. A method of configuring parameters of an autopilot system, comprising:
acquiring riding evaluation information of a preset user on an automatic driving system; wherein the preset user corresponds to one or more pieces of riding evaluation information;
taking each piece of riding evaluation information as a compensation quantity of a preset multidimensional vector, correcting the preset multidimensional vector, and generating a multidimensional vector corresponding to each piece of riding evaluation information;
classifying the generated multiple multidimensional vectors according to a preset classification algorithm, and determining multiple first multidimensional vectors corresponding to the preset user; the classifying algorithm classifies the generated multiple multidimensional vectors according to a preset classifying algorithm, and determines multiple first multidimensional vectors corresponding to the preset user, specifically: when the preset user is the same user, center of gravity processing is carried out on the generated multiple multidimensional vectors, the category of the preset user is determined, and multiple first multidimensional vectors corresponding to the preset user are determined according to the determined category; when the preset users are different users, clustering or clustering and then barycenter processing are carried out on the generated multiple multidimensional vectors, the group of the preset users is determined, and multiple first multidimensional vectors corresponding to the preset users are determined according to the determined group;
and optimizing the automatic driving system parameters of the preset user according to the first multidimensional vectors.
2. The method for configuring parameters of an autopilot system according to claim 1, wherein the acquiring the riding evaluation information of the autopilot system by the preset user specifically includes:
acquiring multiple pieces of riding evaluation information of the same user on the automatic driving system;
alternatively, one or more ride assessment information for the autopilot system by different users is obtained.
3. The method for configuring autopilot system parameters of claim 1 wherein said optimizing said autopilot system parameters of said preset user based on said plurality of first multidimensional vectors is:
when the preset user is the same user, generating first optimization information corresponding to the preset user according to the configuration information corresponding to the determined category and combining the plurality of first multidimensional vectors, and optimizing the automatic driving system parameters of the preset user according to the first optimization information;
when the preset users are different users, generating second optimization information corresponding to the groups according to the configuration information corresponding to the determined groups and combining the first multidimensional vectors, and optimizing automatic driving system parameters corresponding to the groups according to the second optimization information;
the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
4. The method for configuring parameters of an autopilot system of claim 1 further comprising, prior to said clustering of the generated plurality of multidimensional vectors:
denoising the generated multidimensional vectors, and removing abnormal data which contain noise points and error information in the multidimensional vectors of which the change amount of the two multidimensional vectors of the same user is larger than a preset threshold value.
5. The method for configuring parameters of an automatic driving system according to any one of claims 1 to 4, wherein the manner of acquiring the riding evaluation information includes one or more of the following combinations, in particular:
acquiring historical configuration information of a first user on the automatic driving system, and generating riding evaluation information corresponding to the first user according to the historical configuration information;
or generating one piece of riding evaluation information of the second user in a trigger scene according to a plurality of pieces of voice configuration information of the second user in the trigger scene under a preset condition;
or, in response to a parameter configuration operation input by a third user, adjusting a current system parameter of the automatic driving system, and generating one piece of riding evaluation information corresponding to the third user according to the adjusted system parameter.
6. The method for configuring parameters of an automatic driving system according to claim 5, wherein the compensation amount of the preset multidimensional vector is taken as each piece of riding evaluation information, the preset multidimensional vector is corrected, and one multidimensional vector corresponding to each piece of riding evaluation information is generated, specifically:
when the riding evaluation information is generated according to the plurality of voice configuration information, voice recognition is respectively carried out on each voice configuration information through a preset NLP model, each voice recognition result is used as a compensation quantity of a preset multidimensional vector to correct the preset multidimensional vector, and one multidimensional vector corresponding to each riding evaluation information is generated.
7. An automatic driving system parameter configuration apparatus, comprising: the device comprises an acquisition module, a conversion module, a classification module and an optimization module;
the acquisition module is used for acquiring riding evaluation information of a preset user on the automatic driving system; wherein the preset user corresponds to one or more pieces of riding evaluation information;
the conversion module is used for taking each piece of riding evaluation information as a compensation quantity of a preset multidimensional vector, correcting the preset multidimensional vector and generating a multidimensional vector corresponding to each piece of riding evaluation information;
the classifying module is used for classifying the generated multiple multidimensional vectors according to a preset classifying algorithm and determining multiple first multidimensional vectors corresponding to the preset user; the classification module is specifically configured to: when the preset user is the same user, the classification module performs gravity center solving on the generated multiple multidimensional vectors, determines the category of the preset user, and determines multiple first multidimensional vectors corresponding to the preset user according to the determined category; when the preset users are different users, the classification module performs clustering processing or clustering and gravity center solving processing on the generated multiple multidimensional vectors, determines the group of the preset users, and determines multiple first multidimensional vectors corresponding to the preset users according to the determined group;
the optimization module is used for optimizing the automatic driving system parameters of the preset user according to the first multidimensional vectors.
8. The autopilot system parameter configuration apparatus of claim 7 wherein the acquisition module is configured to acquire ride assessment information of a autopilot system for a preset user, specifically:
the acquisition module acquires a plurality of pieces of riding evaluation information of the same user on the automatic driving system;
alternatively, the acquisition module acquires one or more ride assessment information for the autopilot system for different users.
9. The autopilot system parameter configuration arrangement of claim 7 wherein the optimization module is configured to optimize the autopilot system parameters of the preset user based on the plurality of first multidimensional vectors, in particular:
when the preset user is the same user, the optimization module combines the plurality of first multidimensional vectors according to the configuration information corresponding to the determined category to generate first optimization information corresponding to the preset user, and optimizes the automatic driving system parameters of the preset user according to the first optimization information;
when the preset users are different users, the optimization module generates second optimization information corresponding to the groups by combining the plurality of first multidimensional vectors according to the configuration information corresponding to the determined groups, and optimizes the automatic driving system parameters corresponding to the groups according to the second optimization information;
the first optimization information and the second optimization information comprise one or more types of system parameters and parameter adjustment values corresponding to the system parameters respectively.
10. The automatic driving system parameter configuration device according to claim 7, further comprising: a preprocessing module;
the preprocessing module is used for denoising the generated multidimensional vectors before the classifying module performs clustering processing on the generated multidimensional vectors, and eliminating abnormal data which contain noise points and error information in the multidimensional vectors of which the two multidimensional vector variable quantities of the same user are larger than a preset threshold value.
11. The automatic driving system parameter configuration device according to any one of claims 7 to 10, wherein the manner of acquiring the riding evaluation information includes one or more of the following combinations, in particular:
acquiring historical configuration information of a first user on the automatic driving system, and generating riding evaluation information corresponding to the first user according to the historical configuration information;
or generating one piece of riding evaluation information of the second user in a trigger scene according to a plurality of pieces of voice configuration information of the second user in the trigger scene under a preset condition;
or, in response to a parameter configuration operation input by a third user, adjusting a current system parameter of the automatic driving system, and generating one piece of riding evaluation information corresponding to the third user according to the adjusted system parameter.
12. The automatic driving system parameter configuration device according to claim 11, wherein the conversion module is configured to take each piece of seating evaluation information as a compensation amount of a preset multidimensional vector, correct the preset multidimensional vector, and generate one multidimensional vector corresponding to each piece of seating evaluation information, specifically:
when the riding evaluation information is generated according to the historical configuration information, the conversion module respectively carries out voice recognition on each voice configuration information through a preset NLP model, and corrects the preset multidimensional vector by taking each voice recognition result as a compensation quantity of the preset multidimensional vector to generate one multidimensional vector corresponding to each riding evaluation information.
13. An autopilot system comprising: autopilot equipment, a cloud server and configuration means of autopilot system parameters according to any one of claims 7 to 12;
wherein the acquisition module is configured on the autopilot device;
the conversion module is configured on the automatic driving equipment or the cloud server;
the classification module and the optimization module are both configured on the cloud server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111173647.0A CN114084154B (en) | 2021-09-30 | 2021-09-30 | Automatic driving system parameter configuration method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111173647.0A CN114084154B (en) | 2021-09-30 | 2021-09-30 | Automatic driving system parameter configuration method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114084154A CN114084154A (en) | 2022-02-25 |
| CN114084154B true CN114084154B (en) | 2024-03-15 |
Family
ID=80296559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111173647.0A Active CN114084154B (en) | 2021-09-30 | 2021-09-30 | Automatic driving system parameter configuration method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114084154B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018081622A (en) * | 2016-11-18 | 2018-05-24 | トヨタ自動車株式会社 | Driving support apparatus and driving support method |
| CN109711009A (en) * | 2018-12-13 | 2019-05-03 | 北京掌行通信息技术有限公司 | Pilotless automobile method of evaluating performance, device, electronic equipment and medium |
| CN110304068A (en) * | 2019-06-24 | 2019-10-08 | 中国第一汽车股份有限公司 | Acquisition method, device, equipment and the storage medium of running car environmental information |
| CN111610781A (en) * | 2019-02-06 | 2020-09-01 | 哲内提 | Method and system for controlling an autopilot system of a vehicle |
| CN111619576A (en) * | 2020-06-03 | 2020-09-04 | 中国第一汽车股份有限公司 | Control method, device, equipment and storage medium |
| CN112353392A (en) * | 2020-11-09 | 2021-02-12 | 清华大学 | Method for evaluating comfort of intelligent driving automobile passenger |
| KR20210048969A (en) * | 2019-10-24 | 2021-05-04 | 네이버 주식회사 | Method and system for optimizing reinforcement learning based navigation to human preference |
| CN113095344A (en) * | 2020-01-08 | 2021-07-09 | 奥迪股份公司 | Evaluation and optimization device, system and method, vehicle, server and medium |
| CN113370984A (en) * | 2021-06-30 | 2021-09-10 | 中国科学技术大学先进技术研究院 | Multi-index-based comprehensive evaluation method and system for comfort of automatic driving vehicle |
-
2021
- 2021-09-30 CN CN202111173647.0A patent/CN114084154B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018081622A (en) * | 2016-11-18 | 2018-05-24 | トヨタ自動車株式会社 | Driving support apparatus and driving support method |
| CN109711009A (en) * | 2018-12-13 | 2019-05-03 | 北京掌行通信息技术有限公司 | Pilotless automobile method of evaluating performance, device, electronic equipment and medium |
| CN111610781A (en) * | 2019-02-06 | 2020-09-01 | 哲内提 | Method and system for controlling an autopilot system of a vehicle |
| CN110304068A (en) * | 2019-06-24 | 2019-10-08 | 中国第一汽车股份有限公司 | Acquisition method, device, equipment and the storage medium of running car environmental information |
| KR20210048969A (en) * | 2019-10-24 | 2021-05-04 | 네이버 주식회사 | Method and system for optimizing reinforcement learning based navigation to human preference |
| CN113095344A (en) * | 2020-01-08 | 2021-07-09 | 奥迪股份公司 | Evaluation and optimization device, system and method, vehicle, server and medium |
| CN111619576A (en) * | 2020-06-03 | 2020-09-04 | 中国第一汽车股份有限公司 | Control method, device, equipment and storage medium |
| CN112353392A (en) * | 2020-11-09 | 2021-02-12 | 清华大学 | Method for evaluating comfort of intelligent driving automobile passenger |
| CN113370984A (en) * | 2021-06-30 | 2021-09-10 | 中国科学技术大学先进技术研究院 | Multi-index-based comprehensive evaluation method and system for comfort of automatic driving vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114084154A (en) | 2022-02-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220169258A1 (en) | Event determination for vehicles and occupants of mobility provider on maas platform | |
| US20210024083A1 (en) | Method and device for determining unmanned vehicle running scene | |
| WO2019051941A1 (en) | Method, apparatus and device for identifying vehicle type, and computer-readable storage medium | |
| CN111341098B (en) | Congestion state prediction method and device | |
| DE102020208191A1 (en) | Acquisition and processing of vehicle occupant data using artificial intelligence | |
| CN114084154B (en) | Automatic driving system parameter configuration method, device and system | |
| CN113919517A (en) | Vehicle maintenance prompting method, device, server, storage medium and vehicle | |
| CN112462759B (en) | Evaluation method, system and computer storage medium of rule control algorithm | |
| DE102021131054A1 (en) | Method, system and computer program product for evaluating a driving situation for the predictive control of an automated driving function | |
| CN113954836B (en) | Sectional navigation channel changing method and system, computer equipment and storage medium thereof | |
| CN113516847A (en) | Expressway anti-congestion control method and system based on Internet of vehicles V2X | |
| CN108839710A (en) | Vehicle and its power steering control method, system | |
| KR20200102370A (en) | Vehicle performance evaluation method, device and terminal | |
| CN114724011B (en) | Behavior determination method and device, storage medium and electronic device | |
| CN115195728A (en) | Vehicle following control method, system, equipment and storage medium | |
| CN113448806B (en) | Database cluster anomaly detection method and device, terminal device and storage medium | |
| CN117284302A (en) | User-specific driving mode generation method, system, vehicle, electronic equipment and storage medium | |
| CN114239645A (en) | Driving mode identification method and device, terminal equipment and storage medium | |
| US5988844A (en) | Method for arranging a fuzzy controller or a system of interlinked fuzzy controllers | |
| CN110166767A (en) | A kind of the picture quality adjusting method and device of the unused resource of combination vehicle device | |
| US20230005244A1 (en) | Neural network compression device and method for same | |
| CN118301003A (en) | Automatic configuration method and system for traffic priority of vehicle-mounted Ethernet | |
| CN116820738A (en) | Cloud edge-based decision method, device, equipment and storage medium | |
| CN115878976A (en) | Method, device, equipment and medium for identifying uncomfortable working conditions in automatic driving | |
| US20230177112A1 (en) | Method and system for generating a logical representation of a data set, as well as training method |
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 |