KR20220034968A - Method and Apparatus for Tuning Vehicle Driving Device Based on Voice Recognition - Google Patents

Abstract

Disclosed are a tuning apparatus and method of a vehicle driving apparatus based on voice recognition. The present embodiment provides the apparatus and method for tuning a parameter set value of the vehicle driving apparatus to improve driving satisfactory of a driver by reinforcing personalization of the vehicle according to the driver. In addition, provided are the apparatus and method for tuning the parameter set value of the vehicle driving apparatus, using the voice recognition based on deep learning implemented by pre-existing elements in the vehicle as an input element without additional cost increase. The tuning method comprises the following steps of: recognizing key words from a voice signal by using the voice recognition based on deep learning; determining the parameter set value; and tuning an operation mode by using the parameter set value.

Description

BACKGROUND OF THE INVENTION Field of the Invention

The present disclosure relates to a method and apparatus for tuning a vehicle driving apparatus based on voice recognition. More specifically, tuning the parameter value of the vehicle driving device using deep learning-based voice recognition to enhance the driver's driving satisfaction by enhancing the personalization of the vehicle according to the driver. to a method and apparatus for

The content described below merely provides background information related to the present invention and does not constitute the prior art.

The vehicle driving system includes brake, steering, drivetrain (engine, transmission, electric motor, etc. of electric vehicle), suspension, etc., which are operated with SW planted in HW. It is operated in the form of an embedded system. For some components of the driving device, there are driving selection modes that can support 2 to 4 modes, but for most components, the tuning parameters are fixed for convenience in the development process and safety in the use process, so each user's request There is a problem that this is difficult to reflect.

In terms of enhancing the driver's convenience by enhancing the personalization of the vehicle according to the driver, there is an increasing demand for tuning the driving/operation mode to suit the driver's taste. Tunable items include brake sensitivity (braking force versus pedal effort) in the case of braking, the weight of the steering wheel in the case of steering, the degree of rigidity felt by the user in the case of suspension, and shift timing, driving force and fuel economy in the case of driving. correlation, etc.

On the other hand, the voice recognition technology may be usefully used as a means for simplifying the tuning operation for the vehicle driving apparatus. Deep learning-based voice recognition technology using a deep neural network of the Recurrent Neural Network (RNN) series is a technology that has already been proven useful in AI (Artificial Intelligence) speakers and kitchen appliances. On the other hand, the application of voice recognition technology in automobile-related fields is a rudimentary step (see Patent Document 1).

Fortunately, most of the components, including microphones, computing resources, and memory for voice acquisition and recognition, already exist in the electrified (HW → HW+SW embedded system) vehicle. In particular, since it is possible to recycle existing components for the operation of the multimedia device for voice recognition, an additional increase in cost can be suppressed in the process of implementing the tuning of the driving device.

Therefore, it is required to easily implement the tuning of the parameters of the driving device by combining the voice recognition technology while suppressing the additional cost increase.

Patent Document 1: Korean Patent Application No. 10-2018-0157680 Voice recognition led rear indicator light.

An object of the present disclosure is to provide an apparatus and method for tuning a parameter setting value of a vehicle driving apparatus in order to enhance the driving satisfaction of the driver by enhancing the personalization of the vehicle according to the driver.

In addition, an apparatus for tuning a parameter setting value of a vehicle driving apparatus using as an input means deep learning-based voice recognition that can be implemented with elements already present in the vehicle while suppressing an increase in additional cost; and There is an additional purpose of providing a method.

According to an embodiment of the present disclosure, in a method for tuning a driving device performed by a computing device, keywords ( the process of recognizing key words); a process of determining an operation mode of the driving device and related parameter setting values from the keywords, wherein the driving device includes a drive system, a steering system, a brake system, and a suspension as lower units; and tuning the operation mode of the corresponding lower unit by using the parameter setting value.

According to another embodiment of the present disclosure, a voice recognition unit for recognizing keywords (key words) using a deep learning-based voice recognition model from a voice signal; a parameter extracting unit for determining an operation mode of the driving device and related parameter setting values from the keywords, wherein the driving device includes a drive system, a steering system, a brake system, and a suspension as lower units; and a control unit for tuning an operation mode of a corresponding lower unit by using the parameter setting value.

According to another embodiment of the present disclosure, there is provided a computer program stored in a computer-readable recording medium to execute each step included in the method for tuning a driving device.

As described above, according to the present embodiment, by providing an apparatus and method for tuning a parameter setting value of a vehicle driving apparatus using deep learning-based voice recognition, personalization of the vehicle according to the driver is strengthened and the driver It has the effect of making it possible to improve the driving satisfaction of

Also, according to the present embodiment, in the input method of the device for tuning the parameter setting value of the vehicle driving device, deep learning-based voice recognition that can be implemented with elements existing in the vehicle is used. By doing so, there is an effect that it becomes possible to suppress an additional increase in cost.

1 is a block diagram conceptually illustrating an apparatus for tuning a parameter of a vehicle driving apparatus according to an embodiment of the present disclosure.
2 is an exemplary diagram illustrating a relationship between input and output of a pedal simulator according to an embodiment of the present disclosure.
3 is an exemplary diagram schematically illustrating fine tuning of a braking device according to an embodiment of the present disclosure.
4 is an exemplary diagram illustrating a steering torque gain of an MDPS in a steering apparatus according to an embodiment of the present disclosure.
5 is an exemplary view illustrating a damping force gain of an electronic suspension according to an embodiment of the present disclosure.
6 is a flowchart of a method for tuning a parameter of a vehicle driving apparatus according to an embodiment of the present disclosure.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to exemplary drawings. In adding reference numerals to the components of each drawing, it should be noted that the same components are given the same reference numerals as much as possible even though they are indicated on different drawings. In addition, in the description of the present embodiments, if it is determined that a detailed description of a related well-known configuration or function may obscure the gist of the present embodiments, the detailed description thereof will be omitted.

Also, in describing the components of the present embodiments, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only for distinguishing the elements from other elements, and the essence, order, or order of the elements are not limited by the terms. Throughout the specification, when a part 'includes' or 'includes' a certain element, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated. . In addition, the '... Terms such as 'unit' and 'module' mean a unit that processes at least one function or operation, which may be implemented as hardware or software or a combination of hardware and software.

DETAILED DESCRIPTION The detailed description set forth below in conjunction with the appended drawings is intended to describe exemplary embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced.

This embodiment discloses the contents of a voice recognition-based vehicle driving apparatus tuning system. In more detail, tuning the parameter setting value of the vehicle driving device using deep learning-based voice recognition to enhance the driver's driving satisfaction by enhancing the personalization of the vehicle according to the driver. provide a device for

Hereinafter, it is assumed that the vehicle driving apparatus includes a driving system, or a power train (P/T), a steering system, a brake system, and a suspension as a lower unit.

1 is a block diagram conceptually illustrating an apparatus for tuning a parameter of a vehicle driving apparatus according to an embodiment of the present disclosure.

An apparatus 100 (hereinafter, a tuning device) for tuning a parameter of a driving device for a vehicle according to the present disclosure uses a deep learning-based voice recognition, a parameter for an operation mode of a lower unit of a vehicle driving device tune the value. The tuning apparatus 100 includes all or part of the voice recognition unit 102 , the parameter extraction unit 104 , and the control unit 106 . In addition, as a lower unit of the control unit 106, all or part of the drive control unit 110, the steering control unit 112, the braking control unit 114, and the suspension control unit 116 are included. The present invention is not necessarily limited thereto. The control unit 106 may further include a multimedia control unit (not shown) for tuning parameter values for an operation mode of a multimedia-related sub-unit in the vehicle using voice recognition.

All of the tuning apparatus 100 according to the present disclosure may be implemented on a vehicle control unit (VCU) in a vehicle. Here, the VCU represents an integrated system of an embedded system related to vehicle control. In addition, all or part of the tuning device 100 according to the present disclosure may be distributed and implemented on each ECU (Electronic Control Unit) that controls the embedded system for each sub-unit of the vehicle driving apparatus.

Meanwhile, all or part of the voice recognition unit 102 and the parameter extraction unit 104 of the tuning device 100 may be implemented on a separate AI controller existing outside the vehicle. When the voice recognition-related part exists outside the vehicle, the parameters for tuning may be transmitted to the VCU side of the vehicle using a controller area network (CAN). Here, CAN is a system for providing digital serial communication between ECUs, which are embedded systems for each sub-unit. 1 , the part of the voice recognition unit 102 and the parameter extraction unit 104 included in the VCU indicates that these components may exist outside the vehicle.

1 is an exemplary configuration according to the present embodiment, and various components including other components or different connections between components according to the structure and operation of the voice recognition method, the range of the tunable sub-element, and the structure and operation of the control unit, etc. implementation is possible.

The voice recognition unit 102 according to the present disclosure recognizes keywords related to tuning of the driving device from the driver's (or passenger's) voice signal using a deep learning-based voice recognition model.

The voice recognition unit 102 may include a component such as a microphone to acquire a voice signal. An appropriate pre-processing process may be used to convert a voice signal into voice data suitable for a voice recognition model. For example, as the preprocessing process, sampling, noise reduction, and conversion into a spectrogram, which is a signal in the frequency domain, may be used.

As the voice recognition model, a deep neural network of a Recurrent Neural Network (RNN) series suitable for processing a time-sequence signal such as voice may be used. RNN, Long Short-Term Memory (LSTM), and a transformer decoder are known as deep neural networks of the RNN series. In addition, as another example of the speech recognition model, a Convolutional Neural Network (CNN), which is a deep neural network suitable for a spectrogram in which a speech signal is preprocessed into two-dimensional data, may be used for speech recognition. In addition, as a speech recognition model different from the method using a deep neural network, a Gaussian Mixture Model-Hidden Markov Model (GMM-HMM)-based recognition model that recognizes a speech pattern based on a probabilistic model may be used. In addition, the above-described RNN-based model, CNN, GMM-HMM, etc. may be used as a speech recognition model while being coupled to each other.

When the voice recognition model is implemented as a deep neural network, the deep neural network for implementing the voice recognition model may be trained in advance using a training unit (not shown) based on a learning database. In this case, the training unit may train the voice recognition model using supervised learning and/or unsupervised learning according to the type of the deep neural network, the amount/form of available learning data, and the like.

When the voice recognition model according to the present disclosure is mounted on a VCU, the VCU includes at least one communication interface (eg, CAN) that can be connected to a server (not shown). Here, the server may be an external AI controller.

Training for the voice recognition model as described above may be performed in the server. The training unit of the server may perform training on the deep neural network model having the same structure as the voice recognition model, which is a component of the tuning device 100 mounted on the vehicle. Using a communication interface connected to the vehicle, the server transmits the parameters of the trained deep neural network model to the vehicle, and the tuning apparatus 100 may set the parameters of the speech recognition model using the received parameters. In addition, the parameters of the voice recognition model may be set at the time of shipment of the vehicle or the time when the voice recognition model is mounted on the vehicle.

In addition, when the voice recognition unit 102 is mounted on the external AI controller, training for the voice recognition model may be performed on the AI controller using the computing power of the external AI controller on which the voice recognition unit 102 is mounted. .

The keywords recognized using the voice recognition model may include a word indicating a lower unit of the driving device, a word expressing an operation/operation mode of the lower unit, a word indicating a setting for an operation/operation mode, and the like. For example, 'brake', 'soft', and 'hard' may be recognized as keywords from the driver's voice signal saying 'the brake is too soft, make it hard'.

The parameter extraction unit 104 according to the present disclosure extracts an operation mode and parameter setting value of a lower unit by using keywords. For example, if the pedal gain for operating the current brake is set to step 5 out of steps 1 to 10, the operation mode from the keywords 'brake', 'soft', and 'hard' is the brake pedal gain, and the parameter setting value may be determined in step 6.

The parameter extractor 104 may use a lookup table, a decision tree, or an additional deep neural network to determine an operation mode and a parameter setting value. When an additional deep neural network is used, the deep neural networks included in each of the voice recognition unit 102 and the parameter extraction unit 104 may be integrated and operated.

One or two main parameters for tuning can be selected for each sub-unit. For example, the main parameters include the pedal simulator gain of the regenerative brake system, the steering torque gain of the MDPS (Motor Driven Power Steering), the gain of the electronic suspension, and the shift of the transmission in the P/T. The timing, the torque of the electric motor (a trade-off may exist between the magnitude of the torque and the heating performance/fuel economy), etc. may be included. In particular, in the case of a braking device, parameter tuning according to the present disclosure is possible for a regenerative braking device including a pedal simulator, This is because the conventional brake pedal is a HW-based configuration, and the regenerative braking device included in the electric vehicle includes an electric motor for the brake pedal combined with the SW-based pedal simulator.

The control unit 106 according to the present disclosure tunes the operation mode of the corresponding lower unit by using the operation mode and parameter setting value of the lower unit. In this case, the parameter set value used for tuning may be stored in a memory in the VCU (or ECU of a corresponding lower unit), and may be maintained regardless of whether the vehicle is started on/off.

For the main parameters for tuning, minimum/maximum values and initial values that satisfy regulations and performance may be set during the development process. The operator can change the parameter settings in the range of min/max. In addition, the driver may change the set value for each step or change the set value for two or more steps at a time. Accordingly, the driver can change the parameters from the current set value to the desired value including the minimum/maximum values. When the driver requests a change to the set value out of the minimum/maximum values, the controller 106 may notify the driver of this fact after maintaining the current set value or restoring it to a preset initial value.

The control unit 106 includes all or a part of the driving control unit 110 , the steering control unit 112 , the braking control unit 114 , and the suspension control unit 116 as a lower unit. Hereinafter, an embodiment of parameter tuning performed by each sub-unit of the control unit 106 will be described.

In the case of the brake system of an existing vehicle, once the brake pressure of the pedal is determined according to the HW setting of the master booster cylinder in the brake system, no further change is possible. On the other hand, in the case of an integrated regenerative braking system equipped with a pedal simulator, when the driver steps on the pedal, the pedal simulator transmits the pedal repulsive force to the driver and transfers the braking pressure based on the motor to the brake. Accordingly, in a vehicle including a regenerative braking device with an integrated electronic brake, 'electronic tuning of the main brake' is possible by electronically tuning the boost ratio of the pedal.

2 is an exemplary diagram illustrating a relationship between input and output of a pedal simulator according to an embodiment of the present disclosure.

The bar illustrated in FIG. 2 is a curve between input and output simulated by the pedal simulator, and represents an output (actually output force, effect) with respect to a braking input (force applied to the pedal). The input/output curve goes through a short invalid section ① and a jump up section ②, passes through a boosting section ③ and passes through a knee point, and has a section ④ in which the brake boosting force decreases.

Vehicle manufacturers may limit the range of the boosting curve based on laws and regulations and safety margins. Within this limited range, the inflection point may be set in about 10 steps. By using the parameter set values extracted by the voice recognition unit 102 and the parameter extracting unit 104 , as shown in the example of FIG. 3 , the braking control unit 114 may perform fine tuning of the braking device.

As shown in (1) and (2) of FIG. 3 , the braking control unit 114 may tune the parameter in consideration of the minimum/maximum values (eg, 1/10). When the driver requests, or when the driver's request deviates from the minimum/maximum values, the braking controller 114 may restore the parameter set value to a preset initial value ((3) of FIG. 3). The braking control unit 114 may change the set value in two or more steps at a time or change the set value at a time in a specific step ((4) of FIG. 3). Also, the braking controller 114 may tune the parameter set value according to the driving mode (eg, 5 in the normal driving mode, 7 in the sports driving mode, etc.). As described above, the changed parameter set value may be continuously maintained even when the vehicle is turned off/on.

Although the examples of FIGS. 2 and 3 are limited to tuning the inflection point of the pedal simulator, if necessary, the slope of the graph illustrated in FIG. 2 or another location may be selected as a tuning target. In addition, the slope of the boosting section, the difference between the inflection point tuning steps, etc. may be adjusted by the vehicle designer as needed.

4 is an exemplary diagram illustrating a steering torque gain of an MDPS in a steering apparatus according to an embodiment of the present disclosure.

As illustrated in FIG. 4 , the steering torque gain when the steering wheel is manipulated according to the current driving speed may be set. Basically, it is set to be light (small gain) at low speed and heavy (large gain) at high speed. Steering torque gain of MDPS can be divided into about 10 steps. As indicated by the dotted line illustrated in FIG. 4 , an inclination, a difference between gain stages, and the like may be adjusted by a vehicle designer as needed.

The steering control unit 112 may tune the gain of the MDPS by using the parameter set value extracted according to the driver's voice instruction. At this time, the steering control unit 112 performs a similar operation to the braking control unit 114 in the range of the parameter setting value, the number of steps that can be changed at once, restoration to a preset initial value, operation of different setting values according to the driving mode, etc. can do.

5 is an exemplary view illustrating a damping force gain of an electronic suspension according to an embodiment of the present disclosure.

As illustrated in FIG. 5 , the damping force gain of the damper may be determined according to the size of the orifice in the damper (also called a shock absorber). Basically, the damping force gain during straight driving It is set to be small (the damping feels soft), and the damping force gain is large (so that the damping feels hard) when cornering and braking. The damping force gain of the electronic suspension can be divided into about 10 steps, and if necessary, the slope, the difference between the gain steps, etc. can be adjusted by the vehicle designer.

The suspension controller 116 may tune the damping force gain of the electronic suspension by using the parameter set value extracted according to the driver's voice instruction. At this time, in the range of the parameter set value, the number of steps that can be changed at once, restoration to a preset initial value, operation of different set values according to the driving mode, etc., the suspension controller 116 performs a similar operation to the braking controller 114 . can do.

The drive control unit 110, using a method similar to the method performed by the steering control unit 112, the braking control unit 114, and the suspension control unit 116, may tune the shift timing of the transmission, the torque of the electric motor, etc. there is. For example, the shift timing may be determined depending on the current number of stages, the target number of stages, the total number of stages, the current speed, and the like. The driving controller 110 may tune the shift timing of the transmission by using the parameter set value extracted according to the driver's voice instruction.

6 is a flowchart of a method for tuning a parameter of a vehicle driving apparatus according to an embodiment of the present disclosure.

The tuning apparatus 100 according to the present disclosure obtains a voice signal from the driver (or passenger) (S600). The tuning apparatus 100 may include a component such as a microphone to obtain a voice signal.

The tuning device 100 recognizes key words from a voice signal using a deep learning-based voice recognition model (S602).

An appropriate pre-processing process may be used to convert a voice signal into voice data suitable for a voice recognition model. For example, as a preprocessing process, sampling, noise removal, and conversion into a spectrogram, which is a signal in the frequency domain, may be used.

As a speech recognition model, an RNN-based deep neural network suitable for time-series signal processing can be used. Such deep neural networks include RNNs, LSTMs, and transformer decoders.

The deep neural network constituting the speech recognition model may be trained in advance using a training unit (not shown). In this case, a speech recognition model may be trained based on supervised learning or unsupervised learning according to the type of deep neural network.

The recognized keywords may include a word indicating a lower unit of the driving device, a word expressing an operation/operation mode of the lower unit, a word indicating setting of an operation/operation mode, and the like.

The tuning device 100 extracts the operation mode of the driving device and related parameter setting values from the keywords (S604). Here, the driving device includes a drive system, a steering device, a braking device, and a suspension as a lower unit.

The parameter extractor 104 may use a lookup table to determine the operation mode and parameter setting values, or may use an additional deep neural network. Therefore, when an additional deep neural network is used, the deep neural network used for the speech recognition model and the deep neural network used for parameter determination can be integrated and operated.

The tuning apparatus 100 tunes the operation mode of the corresponding lower unit by using the extracted parameter set value (S606). In this case, the parameter set value used for tuning may be stored in a memory in the VCU (or ECU of a corresponding lower unit), and may be maintained regardless of whether the vehicle is started on/off.

One or two main parameters for tuning can be selected for each sub-unit. For the main parameters for tuning, minimum/maximum values and initial values that satisfy regulations and performance may be set during the development process. The operator can change the parameter settings in the range of min/max. In addition, the driver may change the set value for each step or change the set value for two or more steps at a time. When the driver requests a change to a set value out of the minimum/maximum values, the tuning apparatus 100 may notify the driver of this fact after maintaining the current set value or restoring it to a preset initial value. In addition, the tuning apparatus 100 may tune the parameter set value according to the driving mode (eg, 3 in the normal driving mode, 7 in the sports driving mode, etc.).

As described above, according to the present embodiment, by providing an apparatus and method for tuning a parameter value of a vehicle driving apparatus using deep learning-based voice recognition, personalization of the vehicle according to the driver is strengthened, and the driver's There is an effect that it becomes possible to improve the driving satisfaction.

In addition, according to the present embodiment, in the input method of the device for tuning the parameter value of the vehicle driving device, by using deep learning-based voice recognition that can be implemented with elements already existing in the vehicle, , there is an effect that it becomes possible to suppress an additional increase in cost.

Although it is described that each process is sequentially executed in each flowchart according to the present embodiment, the present invention is not limited thereto. In other words, since it may be applicable to change and execute the processes described in the flowchart or to execute one or more processes in parallel, the flowchart is not limited to a time-series order.

Various implementations of the systems and techniques described herein may be implemented in digital electronic circuitry, integrated circuitry, field programmable gate array (FPGA), application specific integrated circuit (ASIC), computer hardware, firmware, software, and/or combination can be realized. These various implementations may include being implemented in one or more computer programs executable on a programmable system. The programmable system includes at least one programmable processor (which may be a special purpose processor) coupled to receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device. or may be a general-purpose processor). Computer programs (also known as programs, software, software applications or code) contain instructions for a programmable processor and are stored on a "computer-readable recording medium".

The computer-readable recording medium includes all types of recording devices in which data readable by a computer system is stored. These computer-readable recording media are non-volatile or non-transitory, such as ROM, CD-ROM, magnetic tape, floppy disk, memory card, hard disk, magneto-optical disk, and storage device. media, and may further include transitory media such as carrier waves (eg, transmission over the Internet) and data transmission media. In addition, the computer-readable recording medium is distributed in network-connected computer systems, and computer-readable codes may be stored and executed in a distributed manner.

Various implementations of the systems and techniques described herein may be implemented by a programmable computer. Here, the computer includes a programmable processor, a data storage system (including volatile memory, non-volatile memory, or other types of storage systems or combinations thereof), and at least one communication interface. For example, a programmable computer may be one of a server, a network appliance, a set-top box, an embedded device, a computer expansion module, a personal computer, a laptop, a Personal Data Assistant (PDA), a cloud computing system, or a mobile device.

The above description is merely illustrative of the technical idea of this embodiment, and various modifications and variations will be possible without departing from the essential characteristics of the present embodiment by those skilled in the art to which this embodiment belongs. Accordingly, the present embodiments are intended to explain rather than limit the technical spirit of the present embodiment, and the scope of the technical spirit of the present embodiment is not limited by these embodiments. The protection scope of this embodiment should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be interpreted as being included in the scope of the present embodiment.

100: tuning device
102: voice recognition unit 104: parameter extraction unit
106: control unit 110: drive control unit
112: steering control unit 114: braking control unit
116: suspension control unit

Claims (13)

In the method for tuning the driving device performed by the computing device,
A process of recognizing keywords (key words) from a voice signal related to a user's request using a deep learning-based voice recognition model;
a process of determining an operation mode of the driving device and related parameter setting values from the keywords, wherein the driving device includes a drive system, a steering system, a brake system, and a suspension as lower units; and
The process of tuning the operation mode of the corresponding lower unit using the parameter set value
A method for tuning a driving device comprising a. According to claim 1,
The keywords are
A method for tuning the driving apparatus, characterized in that it includes a word indicating the lower unit, a word expressing an operation/operation mode of the lower unit, and a word meaning a setting for the operation/operation mode. According to claim 1,
The parameter setting value is
The method for tuning the driving device, characterized in that it is stored in the memory in the computing device, and is maintained regardless of whether the vehicle is started off/on. According to claim 1,
The parameter setting value is
A method for tuning the driving device, characterized in that it is set to one of a plurality of steps within the range of the minimum/maximum value, and is changed step by step according to the user's request or is changed in two or more steps at a time. According to claim 1,
In the braking device, a knee point that is a boundary between a boosting section and a boosting force reduction section among an output (actual output force) curve for a braking input (force applied to the pedal) using the parameter set value, the A method for tuning the driving device, characterized in that the slope of the boost section and the boost reduction section, the starting point of the boost section and/or the end point of the boost reduction section are set. The method of claim 1,
The method for tuning the driving system, characterized in that, with respect to the drive system, the shift timing of the transmission and/or the torque of the electric motor are tuned by using the parameter set value. According to claim 1,
The method for tuning the driving device, characterized in that for the steering device, a steering torque gain when the steering wheel is operated according to the current driving speed is set using the parameter setting value. According to claim 1,
A method for tuning a traveling device, characterized in that for the suspension, a damping force gain of a damper according to the size of an orifice in the damper is set using the parameter setting value. According to claim 1,
The voice recognition model is
A method for tuning a driving device, characterized in that it is implemented as a deep neural network of a Recurrent Neural Network (RNN) series or a deep neural network of a Convolutional Neural Network (CNN) type, which is trained in advance using data for learning. a voice recognition unit for recognizing key words from a voice signal using a deep learning-based voice recognition model;
a parameter extracting unit for determining an operation mode of a driving device and related parameter setting values from the keywords, wherein the driving device includes a drive system, a steering system, a brake system, and a suspension as lower units; and
A control unit that tunes the operation mode of the corresponding lower unit by using the parameter setting value
Apparatus for tuning a driving device comprising a. 11. The method of claim 10,
The control unit is
and a drive control unit, a steering control unit, a braking control unit, and a suspension control unit to perform tuning for each of the lower units. 11. The method of claim 10,
The voice recognition model is
A device for tuning a driving device, characterized in that it is implemented as a deep neural network of a Recurrent Neural Network (RNN) series or a deep neural network of a Convolutional Neural Network (CNN) type, and is trained in advance using data for learning. A computer program stored in a computer-readable recording medium to execute each step included in the method for tuning the driving device according to any one of claims 1 to 9.

Images