KR20240006987A - Terminal having driving environment determination function of mobility and method of driving environment determination of mobility using the same - Google Patents

Abstract

A terminal equipped with a function for determining the driving environment of mobility according to a preferred embodiment of the present invention includes a transmitter and receiver that communicates and connects to a mobility controller that controls mobility, and learns an image of the ground surrounding the mobility to determine at least one surrounding environment of the mobility. It includes a modeling unit that classifies into a class, and a determination unit that determines whether the mobility is to be driven based on the classification result of the modeling unit, and transmits the determination result to the mobility controller.

Description

Terminal equipped with a mobility driving environment determination function, and a mobility driving environment determination method using the same {TERMINAL HAVING DRIVING ENVIRONMENT DETERMINATION FUNCTION OF MOBILITY AND METHOD OF DRIVING ENVIRONMENT DETERMINATION OF MOBILITY USING THE SAME}

The present invention relates to a terminal equipped with a mobility driving environment determination function and a method of determining the mobility driving environment using the same.

Personal mobility is a term referring to a means of personal transportation. It is called by various names depending on its composition and appearance, but mainly refers to a means of transportation for one person that is powered by electricity.

Such personal mobility includes electric bicycle, electric kickboard, electric scooter, electric kick-bike, Segway, etc.

However, here, ordinary bicycles, which are not powered by electricity but are a means of transportation for one person, are also considered to be included in personal mobility.

The use of personal mobility is increasing due to the increase in single-person households and convenience, but there is little consideration of the safety of personal mobility driving.

In particular, in the case of personal mobility, not only road driving but also sidewalk driving is frequent, and there is a problem of reduced safety due to the high possibility of accidents due to contact with obstacles or people while driving on sidewalks.

Accordingly, there is a need for a method to reduce the possibility of accidents occurring by limiting the sidewalk driving of personal mobility.

Republic of Korea Patent No. 10-2379970

Accordingly, the present invention was developed in consideration of the above circumstances, and uses a deep learning model to determine whether or not the mobility will be driven on the road, and determines the driving environment of the mobility by limiting the maximum speed of the mobility when driving on the road to encourage the driver to drive on the road. The purpose is to provide a terminal with functions and a method for determining the driving environment of mobility using the same.

A terminal equipped with a mobility driving environment determination function according to a preferred embodiment of the present invention for achieving the above object includes a transceiver unit for communication connection to a mobility controller that controls mobility; a modeling unit that learns images of the surrounding ground of the mobility and classifies the surrounding environment of the mobility into at least one class; and a determination unit that determines whether the mobility is to be driven based on the classification result of the modeling unit and transmits the determination result to the mobility controller.

It may further include a camera unit that photographs the ground surrounding the mobility and transmits the captured ground image to the modeling unit.

It may further include a sensor unit that detects the current location of the mobility and transmits the detected location signal to the determination unit.

The determination unit may determine whether the mobility is to be driven based on the classification result of the modeling unit and the location signal of the sensor unit.

The determination unit calculates the delivery driving distance of the mobility, compares the calculated delivery distance with a preset reference distance, and, if the delivery driving distance exceeds the reference distance according to the comparison result, determines the delivery driving distance of the mobility. You can judge.

The determination unit may limit the maximum speed of the mobility by transmitting a driving environment bit signal related to the guided driving of the mobility to the mobility controller.

The modeling unit can learn the ground image using a pre-prepared deep learning model and output scores for two classes representing a sidewalk and a road.

The modeling unit may compare the maximum score among the scores of each of the two classes with a preset threshold.

If the maximum score exceeds the threshold, the modeling unit classifies the surrounding environment of the mobility into classes corresponding to the maximum score, and if the maximum score is less than the threshold, determines whether the two classes have initial scores. can be judged.

If the modeling unit does not have the initial score, it can suspend class classification for the surrounding environment of the mobility, and if it does have the initial score, it can calculate the score change rate for each of the two classes.

The modeling unit may compare the score change rates of each of the two classes, and classify the class with a relatively large score change rate as the surrounding environment of the mobility according to the comparison result.

A method of determining the driving environment of mobility using a terminal according to a preferred embodiment of the present invention for achieving the above object includes a connection step of communication between the transmitter and receiver to a mobility controller that controls mobility; A classification step in which a modeling unit learns ground images surrounding the mobility and classifies the surrounding environment of the mobility into at least one class; and a determination step in which a determination unit determines whether or not the mobility is to be driven based on the classification result of the modeling unit.

The classification step may include, after the connection step, an input step in which a camera unit photographs the ground surrounding the mobility and inputs the captured ground image to the modeling unit.

The classification step includes: a learning step of learning the ground image using a deep learning model prepared in advance by the modeling unit after the input step; and an output step in which the modeling unit outputs scores for two classes representing a sidewalk and a road according to the learning results of the learning step.

The classification step may include a threshold comparison step in which the modeling unit compares a maximum score among scores of each of the two classes with a preset threshold value.

The classification step includes: a first class determination step in which the modeling unit classifies the surrounding environment of the mobility into a class corresponding to the maximum score when the maximum score exceeds the threshold; and an initial value retention determination step in which the modeling unit determines whether the initial scores for each of the two classes are retained when the maximum score is less than the threshold.

The classification step includes a holding step of withholding class classification for the surrounding environment of the mobility when the modeling unit does not hold the initial score; and a rate of change calculation step of calculating a rate of change in scores for each of the two classes when the modeling unit holds the initial score.

The classification step includes a second class determination step in which the modeling unit compares the score change rates of each of the two classes, and classifies the class with a relatively large score change rate as the surrounding environment of the mobility according to the comparison result. can do.

According to a terminal equipped with a function for determining the driving environment of mobility according to a preferred embodiment of the present invention, and a method for determining the driving environment of mobility using the same, whether the mobility is guided is determined using a deep learning model, and based on the determination result. By limiting the maximum speed of mobility when driving on sidewalks, drivers can be encouraged to drive on the road.

In addition, safe driving scores can be given to mobility users based on the basis for driving judgment, and in connection with this, benefits can be provided in the form of insurance premium discounts, usage fee surcharges, or discounts.

1 is a block diagram of a terminal equipped with a mobility driving environment determination function according to a preferred embodiment of the present invention.
Figure 2 is a diagram showing the configuration of a mobility controller.
Figure 3 is a diagram showing an example of calculating driving scores and usage amounts related to a user's mobility driving.
Figure 4 is a diagram showing an example of an application map related to a mobility driving path.
Figure 5 is a flowchart of a method for determining a mobility driving environment using a terminal according to a preferred embodiment of the present invention.
FIG. 6 is a flowchart showing the detailed configuration of the driving environment determination step of FIG. 5.
Figure 7 is a flowchart of a process for allowing starting of shared mobility in the method of determining the driving environment of mobility using a terminal according to a preferred embodiment of the present invention.
Figure 8 is a flowchart of a method for limiting the maximum speed of mobility during sidewalk driving.
Figure 9 is a flowchart of a method of generating a warning sound according to guided driving of mobility.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. First, when adding reference signs to components in each drawing, it should be noted that the same components are given the same reference numerals as much as possible even if they are shown in different drawings. In addition, preferred embodiments of the present invention will be described below, but the technical idea of the present invention is not limited or restricted thereto, and of course, it can be modified and implemented in various ways by those skilled in the art.

1 is a block diagram of a terminal equipped with a mobility driving environment determination function according to a preferred embodiment of the present invention.

Referring to FIG. 1, the terminal 100 equipped with a mobility driving environment determination function according to a preferred embodiment of the present invention uses a deep learning model to determine whether the mobility is to be driven and determines the maximum speed of the mobility during guided driving. It is characterized by encouraging drivers to drive on the road by limiting .

The terminal 100, which has a mobility driving environment determination function, is mounted on a mobility device (not shown) to acquire mobility driving environment information, and includes a transceiver unit 210, a camera unit 220, and a modeling unit 230. , includes a sensor unit 240, a determination unit 250, and a calculation unit 260. The terminal 100 may be a smartphone, but is not limited thereto.

The transceiver unit 210 may be configured to communicate with the mobility controller 100 and the mobility operation server 300. The transceiver unit 210 may be connected to the mobility controller 100 through Bluetooth or Near Field communication (NFC) communication. The transceiver unit 210 may be connected to the mobility operation server 30 through 5G or Wifi communication.

The transceiver 210 may transmit the decision result of the determination unit 250 related to the mobility driving environment to the mobility controller 100. The transceiver unit 210 may transmit information related to mobility usage fees to the mobility operation server 300.

The mobility controller 100 can start the mobility when connected to the terminal 200 via Bluetooth or NFC. When receiving a signal related to guided driving of the mobility from the terminal 200, the mobility controller 100 may limit the maximum speed of the mobility. The mobility controller 100 may transmit mobility driving data to the terminal 200 according to the user's selection.

The mobility operation server 200 can receive and store mobility driving data. Depending on the user's selection, the mobility operation server 200 may not store the driving data or may receive and store only part of the driving data and provide services accordingly.

The camera unit 220 is provided at the rear of the terminal 100 and can take pictures of the rear of the terminal 100. The camera unit 220 can obtain a ground image by photographing the ground on which the mobility is traveling. The camera unit 220 may transmit the acquired ground image to the modeling unit 230.

The modeling unit 230 can determine whether the surrounding environment of the mobility is a road or sidewalk by learning the ground image acquired by the camera unit 220. The modeling unit 230 may include a deep learning model for recognizing the road surface in the surrounding environment of the driving mobility. In one embodiment, a deep learning model can be prepared by combining two Relu Dense and two Softmax Dense based on the MobileNet V2 model. A deep learning model can be configured to classify two classes (sidewalk, driveway).

The deep learning model can be operated on the terminal 100 by reducing the amount of calculation and size. In one embodiment, the deep learning model can be converted to a model with a small size and computational amount through Float16 and int8 quantization. Here, the quantization level may be selected considering the inference time and accuracy of the deep learning model.

The sensor unit 240 may be a Global Positioning System (GPS). The sensor unit 240 may detect the current location of the mobility and transmit the current location signal to the determination unit 250.

The determination unit 250 may determine whether guided driving of the mobility is performed over a preset reference distance based on the class classification result of the modeling unit 230 and the position signal of the sensor unit 240. If the determination unit 250 determines that the mobility is guided beyond the reference distance, it may transmit a driving environment bit signal of 1 to the mobility controller 100 through the transceiver 210. The driving environment bit signal may have a bit signal of 0 for roads and 1 for sidewalks.

The calculation unit 260 may calculate a mobility driving score related to the user's mobility safety driving. The calculation unit 260 may selectively calculate the mobility driving score for the purpose of discounting insurance premiums.

The calculation unit 260 may calculate the mobility driving score by reflecting the mobility driving rate in the factor. Equation 1 below shows an example of a driving score calculation formula.

<Equation 1>

In Equation 1, the driving score _raw represents the initial driving score, the driving distance _roadway represents the road driving distance of the mobility, the driving distance _walkway represents the sidewalk driving distance of the mobility, and the driving distance _total represents the total driving distance of the mobility. .

If the mobility is shareable, the calculation unit 260 can calculate the mobility usage fee. The calculation unit 260 may apply a discount or surcharge to the final usage fee according to the road operation. Equation 2 below shows an example of a usage fee calculation formula.

<Equation 2>

In Equation 2, the usage fee _raw represents the basic usage fee for mobility.

Figure 2 is a diagram showing the configuration of a mobility controller.

Referring to FIG. 2, the mobility controller 100 is characterized by limiting the maximum speed of the mobility in order to guide the mobility to the roadway when driving on the sidewalk.

Mobility controller 100 may include a vehicle control unit 110 and a motor control unit 120.

When receiving a driving environment bit signal related to guided driving of mobility from the terminal 200, the vehicle control unit 110 may set a maximum speed value to limit the maximum speed of mobility. In one embodiment, the maximum speed value may be 10 km/h. The maximum speed value is not limited to the above example and can be appropriately set depending on the traffic environment or user needs.

The vehicle control unit 110 may limit and set the maximum speed by considering the area, number of people, user-set limit information, and parking location information.

Vehicle control unit 110 may include logic used to set the required torque in relation to maximum speed. Logic can calculate the required torque to drive mobility at a limited maximum speed in ASW (Application SW). The logic can calculate the final required torque by considering the required torque according to the driver's willingness to accelerate and the level of limitations depending on the battery and mobility status. When determining the torque limit level, logic can consider limiting factors based on driving information.

The motor control unit 120 may control the driving motor of the mobility using the calculated required torque to limit the driving speed of the mobility to the maximum speed. The driving motor may be a BLDC motor. Mobility can run at a maximum speed limited by the travel motor. Through this, the user can feel the intention of moving the location according to the low-speed driving of the mobility, and can move from the sidewalk to the road.

Figure 3 is a diagram showing an example of calculating driving scores and usage amounts related to a user's mobility driving.

Figure 3(a) shows an example of a safe driving score related to a user's mobility driving, and Figure 3(b) shows an example of a usage amount related to a user's mobility driving.

In Figure 3(a), the terminal 200 may calculate a safety score related to the user's mobility driving by considering the ratio of road driving and sidewalk driving during a preset reference driving distance. The standard driving distance may be approximately 15 km, but may be appropriately changed depending on the user's needs.

In Figure 3(b), when the mobility being used by the user is public mobility, the terminal 200 calculates the usage amount related to the user's mobility driving by considering the ratio of road driving and sidewalk driving during the preset reference driving distance. It can be calculated. The standard driving distance may be approximately 15 km, but may be appropriately changed depending on the user's needs.

Figure 4 is a diagram showing an example of an application map related to a mobility driving path.

Referring to FIG. 4, the terminal 200 can display the entire mobility driving route on the application map. The terminal 200 can display the entire driving route divided into roads and sidewalks. The terminal 200 can display the time when the mobility is located in the sidewalk driving section. The user can change the driving direction of the mobility from the sidewalk to the road according to the sidewalk driving information displayed on the terminal 200.

Figure 5 is a flowchart of a method for determining a mobility driving environment using a terminal according to a preferred embodiment of the present invention.

Referring to Figures 1 and 5, the method of determining the driving environment of mobility using a terminal according to a preferred embodiment of the present invention learns the surrounding environment of mobility using a deep learning model and guides mobility based on the learning results. It is characterized by determining whether to drive and inducing road driving of the mobility based on the driving distance of the mobility.

In the safety score determination step (S510), the determination unit 250 determines whether the mobility can be used based on the safety score related to driving of the mobility prepared in advance. The determination unit 250 may determine that mobility is unusable if the safety score is less than a preset standard score. The terminal 200 may determine that mobility is available if the safety score is greater than or equal to the standard score. This safety score determination step (S510) can be selectively applied by the user. Additionally, the safety score determination step (S510) may be omitted if the mobility is shared mobility operated by the mobility operation server 300.

In the connection step (S520), the transceiver 210 may be connected to the mobility controller 100 for communication. The transceiver unit 210 may be connected to the mobility controller 100 via Bluetooth or NFC.

In the driving environment determination step (S530), when the modeling unit 230 is connected to the mobility controller 100, it receives the surrounding ground image of the mobility from the camera unit 220 and creates a class for the surrounding environment of the mobility based on this. Classify. The determination unit 250 determines whether the mobility is to be driven based on the class classification result of the modeling unit 230. This driving environment determination process will be described in detail later with reference to FIG. 6. On the other hand, if the mobility is shared mobility, a step of allowing mobility start only when communication is connected between the connection step (S520) and the driving environment determination step (S530) between the mobility controller 100 and the terminal 200 may be additionally applied. .

In the reset step (S540), when the driving environment of the mobility is road driving, the determination unit 250 resets the state related to the guided driving of the mobility. The guided driving state may include a guided driving flag and a guided driving bit. Each of the delivery driving flag and delivery driving bit may be set to 0 upon reset. The determination unit 250 may inform the mobility controller 100 of the road driving status of the mobility.

In the continuous determination step (S550), when the driving environment of the mobility is sidewalk driving, the determination unit 250 determines whether the driving environment of the previous frame is road driving. This is to determine whether the mobility is continuously driving on the sidewalk.

In the first state setting step (S560), when the driving environment of the previous frame is road driving and the driving environment of the current frame is road driving, the decision unit 250 sets the road driving flag to the on state. The guided driving flag in the on state may have a value of 1.

In the first driving distance calculation step (S570), if the delivery driving flag is on, the calculation unit 260 may accumulate and calculate the delivery driving distance of the mobility based on the GPS value of the sensor unit 240.

Meanwhile, in the second driving distance calculation step (S580) after the continuous determination step (S550), if the driving environment of the previous frame is Indian driving, the calculation unit 260 moves based on the GPS value of the sensor unit 240. It can be calculated by accumulating the driving distance of the frame and the current frame.

In the distance determination step (S590), the determination unit 250 may determine whether the delivery driving distance exceeds a preset reference distance. The reference distance can be set appropriately according to user needs.

In the second state setting step (S600), the determination unit 250 sets the delivery driving bit to the on state when the calculated delivery driving distance exceeds the reference distance. The guided driving bit in the on state may have a value of 1. The determination unit 250 may request a maximum speed limit for mobility by transmitting the delivery driving bit to the mobility controller 100.

Meanwhile, in the starting determination step (S610), the determination unit 250 may receive information related to starting the mobility from the mobility controller 100 and check whether the starting is off. If the mobility is starting, the driving environment determination step (S530) and subsequent steps may be continuously performed.

FIG. 6 is a flowchart showing the detailed configuration of the driving environment determination step of FIG. 5.

Referring to FIGS. 1 and 6 , the driving environment determination step (S530) may include an input step (S531) to a storage step (S539).

In the input step (S531), the modeling unit 230 may receive a ground image captured from the camera unit 220 of the ground surrounding the mobility.

In the learning step (S532), the modeling unit 230 may learn the ground image using a deep learning model.

In the output step (S533), the modeling unit 230 may output scores for two classes representing sidewalks and roads according to learning of the ground image. The modeling unit 230 may calculate scores for each of the first class representing a sidewalk and the second class representing a road. In one embodiment, the modeling unit 230 may determine that the road in the ground image is India with the highest prediction probability when the probability that it is India is 0.98 and the probability that it is a province is 0.02 according to the ground learning result.

In the threshold comparison step (S534), the modeling unit 230 may compare the maximum score among the scores for each class with a preset threshold. The threshold can be appropriately determined through experimentation. In one embodiment, the modeling unit 230 may determine that class discrimination is uncertain because the maximum score does not exceed the threshold when the threshold is 0.7, the probability of being on the sidewalk is 0.6, and the probability of being on the road is 0.4.

In the first class determination step (S535), if the maximum score among the scores for each class exceeds the threshold, the modeling unit 230 may determine the class corresponding to the maximum score as the surrounding environment of mobility.

In the initial value retention determination step (S536), the modeling unit 230 may determine whether the initial score for each class is retained when the maximum score among the scores for each class is less than or equal to the threshold.

In the holding step (S537), if there is no initial score for each class, the modeling unit 230 may suspend class classification and determine the mobility on the road. At this time, the driving environment bit signal may have 0.

In the rate of change calculation step (S538), if there is an initial score for each class, the modeling unit 230 may calculate the rate of change in the score based on the score at the previous time and the score at the current time. In one embodiment, at time t (current time), the score of the first class may be 0.65 and the score of the second class may be 0.45. At time t-1 (previous time), the score of the first class may be 0.6 and the score of the second class may be 0.4. At this time, the score change rate of the first class is 0.05, and the score change rate of the second class is -0.05. Since the score change rate of the first class is relatively large, the delivery driving corresponding to the first class can be determined.

In the second class determination step (S539), the modeling unit 230 compares the score change rates of each of the two classes, and according to the comparison result, the class with a relatively large score change rate can be determined as the surrounding environment of the mobility. there is.

In the storage step (S539a), the modeling unit 230 may store the class and score. The modeling unit 230 may transmit the class classification result for the surrounding mobility environment classified as a sidewalk or a roadway according to the class and score to the determination unit 250.

Figure 7 is a flowchart of a process for allowing starting of shared mobility in the method of determining the driving environment of mobility using a terminal according to a preferred embodiment of the present invention.

Referring to FIGS. 1, 5, and 7, the method of determining the driving environment of mobility using a terminal according to a preferred embodiment of the present invention includes a start-up permit process in the case of shared mobility operated by the mobility operation server 300. Further considerations can be made.

In the pairing step (S710), the transceiver 210 pairs with the mobility controller 100.

In the first notification step (S720), when the determination unit 250 determines that the transceiver unit 210 and the mobility controller 100 are in an unconnected state, the communication unconnected state is notified to the user through a separate notification device (not shown). can be notified to

In the helmet wearing determination step (S730), when the determination unit 250 determines that the transceiver unit 210 and the mobility controller 100 are connected, it receives a captured image of the user's body from the camera unit 220 and based on this, You can determine whether or not you are wearing a helmet.

In the second notification step (S740), if the determination unit 250 determines that the user is not wearing a helmet, it may notify the user of the status of the helmet not being worn through a separate notification device.

In the start-up permission step (S750), the determination unit 250 may transmit a mobility start-permission related signal to the mobility controller 100 when it is determined that the user is wearing a helmet. After the start-up permission step (S750), the driving environment determination step (S530) may be performed.

Figure 8 is a flowchart of a method for limiting the maximum speed of mobility during sidewalk driving.

Referring to FIGS. 1 and 8 , when the mobility controller 100 receives a signal related to guided driving of the mobility from the terminal 200, it is characterized in that it limits the maximum speed of the mobility in order to guide the mobility to travel on the road.

In the delivery driving determination step (S810), the mobility controller 100 receives a signal related to the delivery driving or road driving of the mobility from the terminal 200 and determines whether the mobility will be guided driving based on this.

In the regional option determination step (S820), when it is determined that the mobility is guided, the mobility controller 100 determines whether to use the regional speed limit option. The regional speed limit option limits the speed based on the current location of the vehicle and can be set appropriately according to the user's needs or vehicle specifications.

In the first option reflection step (S830), when using the regional speed limit option, the mobility controller 100 may reflect the preset regional speed limit to the maximum speed of mobility.

In the object option determination step (S840), if the regional speed limit option is not used, the mobility controller 100 determines whether to use the object number speed limit option. The object speed limit option limits the speed of mobility according to the number of people in the sidewalk and can be set appropriately according to the user's needs.

In the second option reflection step (S850), when the object speed limit option is used, the mobility controller 100 may reflect the preset speed limit for each region to the maximum speed of mobility.

In the manual option determination step (S860), if the object number speed limit option is not used, the mobility controller 100 determines whether to use the manual speed limit option. The manual speed limit option may be to manually set the speed limit according to user operation.

In the third option reflection step (S870), when using the manual speed limit option, the mobility controller 100 may reflect the speed limit according to the user's manual operation to the maximum speed of the mobility.

In the manual option determination step (S880), the mobility controller 100 determines whether to use the parking position option when the manual speed limit option is not used. The parking location option may be to set an insurance premium based on whether there is a parking spot for the mobility within a reference distance.

In the fourth option reflection step (S890), when the parking location option is used, the mobility controller 100 may exclude the selection of an insurance premium surcharge according to the guided driving of the mobility. In the case of parking locations, various locations can be set, but if additional locations other than those provided by default are selected, an insurance premium surcharge may be included.

In the limit information confirmation step (S900), the mobility controller 100 checks speed limit information according to various options.

In the vehicle speed limit reflection step (S910), the mobility controller 100 reflects the speed limit information according to the guided driving of the mobility to the maximum speed of the mobility.

In the speed control step (S920), the mobility controller 100 performs speed limit or torque limit control according to the vehicle system by considering the maximum speed prepared in step S910.

Meanwhile, in the case of road driving of the mobility in step S810, the mobility controller 100 normally controls the road driving of the mobility according to the user's request in the normal control step (S930).

Figure 9 is a flowchart of a method of generating a warning sound according to guided driving of mobility.

Referring to FIG. 9, a method of determining the driving environment of mobility using a terminal according to another embodiment of the present invention is characterized in that it induces driving on the road by generating a warning sound when the mobility vehicle enters the sidewalk.

In the delivery driving determination step (S1000), the mobility controller 100 receives a signal related to the delivery driving or road driving of the mobility from the terminal 200 and determines whether the mobility will be guided driving based on this.

In the normal control step (S1010), when driving on a mobility road, the mobility controller 100 controls normal driving without a separate speed limit.

In the speed determination step (S1020), the mobility controller 100 compares the current speed with the preset speed limit in the case of guided mobility. The speed limit can be set appropriately according to user needs.

In the warning sound generation step (S1030), if the current speed exceeds the speed limit, the mobility controller 100 generates a warning sound through a separate notification device (not shown).

In the time calculation step (S1040), the mobility controller 100 accumulates and calculates the warning sound generation time to generate the cumulative warning sound generation time. The cumulative occurrence time of the warning sound can be applied to insurance premium surcharges and discounts.

Meanwhile, in the warning sound cancellation step (S1050), if the current speed is below the speed limit, the mobility controller 100 determines normal driving and cancels the warning sound.

The above description is merely an illustrative explanation of the technical idea of the present invention, and various modifications, changes, and substitutions can be made by those skilled in the art without departing from the essential characteristics of the present invention. will be. Accordingly, the embodiments disclosed in the present invention and the attached drawings are not intended to limit the technical idea of the present invention, but are for illustrative purposes, and the scope of the technical idea of the present invention is not limited by these embodiments and the attached drawings. .

Steps and/or operations according to the invention may occur simultaneously in different embodiments, in different orders, in parallel, for different epochs, etc., as would be understood by those skilled in the art. You can.

Depending on the embodiment, some or all of the steps and/or operations may include executing instructions, programs, interactive data structures, clients and/or servers stored on one or more non-transitory computer-readable media. At least a portion may be implemented or performed using one or more processors. The one or more non-transitory computer-readable media may illustratively be software, firmware, hardware, and/or any combination thereof. Additionally, the functionality of the “modules” discussed herein may be implemented in software, firmware, hardware, and/or any combination thereof.

100: Mobility controller
200: terminal
210: Transmitter and receiver
220: Camera unit
230: Modeling department
240: sensor unit
250: Judgment unit
260: Calculation unit

Claims (18)

A transceiver unit for communication connection to a mobility controller that controls mobility;
a modeling unit that learns images of the surrounding ground of the mobility and classifies the surrounding environment of the mobility into at least one class; and
a determination unit that determines whether the mobility is to be driven based on the classification result of the modeling unit and transmits the determination result to the mobility controller;
A terminal equipped with a mobility driving environment judgment function including. According to claim 1,
a camera unit that photographs the ground surrounding the mobility and transmits the captured ground image to the modeling unit;
A terminal equipped with a mobility driving environment determination function, further comprising: According to claim 1,
A sensor unit that detects the current location of the mobility and transmits the detected location signal to the determination unit;
A terminal equipped with a mobility driving environment determination function, further comprising: According to claim 3,
The judgment department,
A terminal equipped with a mobility driving environment determination function, characterized in that it determines whether the mobility is guided or not based on the classification result of the modeling unit and the position signal of the sensor unit. According to claim 4,
The judgment department,
Calculate the delivery driving distance of the mobility, compare the calculated delivery distance with a preset reference distance, and determine delivery of the mobility if the delivery distance exceeds the reference distance according to the comparison result. A terminal equipped with a function to determine the driving environment of mobility. According to claim 5,
The judgment department,
A terminal equipped with a mobility driving environment determination function, characterized in that the maximum speed of the mobility is limited by transmitting a driving environment bit signal related to the guided driving of the mobility to the mobility controller. According to claim 1,
The modeling department,
A terminal equipped with a mobility driving environment judgment function, characterized by learning the ground image using a pre-prepared deep learning model and outputting scores for two classes representing sidewalks and roads. According to claim 7,
The modeling department,
A terminal equipped with a mobility driving environment judgment function, characterized in that the maximum score among the scores of each of the two classes is compared with a preset threshold. According to claim 8,
The modeling department,
If the maximum score exceeds the threshold, classifying the surrounding environment of the mobility into a class corresponding to the maximum score,
A terminal equipped with a mobility driving environment determination function, characterized in that it determines whether the two classes have initial scores when the maximum score is less than the threshold. According to clause 9,
The modeling department,
If you do not have the initial score, suspend class classification for the surrounding environment of the mobility,
A terminal equipped with a mobility driving environment determination function, characterized in that, when holding the initial score, the score change rate for each of the two classes is calculated. According to claim 10,
The modeling department,
A terminal equipped with a mobility driving environment judgment function, characterized in that the score change rates of each of the two classes are compared with each other, and according to the comparison result, the class having a relatively large score change rate is classified as the surrounding environment of the mobility. A connection step in which the transceiver communicates with a mobility controller that controls mobility;
A classification step in which a modeling unit learns ground images surrounding the mobility and classifies the surrounding environment of the mobility into at least one class; and
A determination step in which a determination unit determines whether the mobility is to be driven based on the classification result of the modeling unit;
A method of determining the driving environment of mobility using a terminal including. According to claim 12,
The classification step is,
After the connection step, an input step in which a camera unit photographs the ground surrounding the mobility and inputs the captured ground image to the modeling unit;
A method of determining the driving environment of mobility using a terminal, comprising: According to claim 13,
The classification step is,
After the input step, a learning step in which the modeling unit learns the ground image using a pre-prepared deep learning model; and
An output step in which the modeling unit outputs scores for two classes representing a sidewalk and a road according to the learning results of the learning step;
A method of determining the driving environment of mobility using a terminal, comprising: According to claim 14,
The classification step is,
a threshold comparison step in which the modeling unit compares a maximum score among the scores of each of the two classes with a preset threshold;
A method of determining the driving environment of mobility using a terminal, comprising: According to claim 15,
The classification step is,
A first class determination step in which the modeling unit classifies the surrounding environment of the mobility into a class corresponding to the maximum score when the maximum score exceeds the threshold; and
an initial value retention determination step in which the modeling unit determines whether or not an initial score is retained for each of the two classes when the maximum score is less than the threshold value;
A method of determining the driving environment of mobility using a terminal, comprising: According to claim 16,
The classification step is,
a holding step of withholding class classification for the surrounding environment of the mobility when the modeling unit does not hold the initial score; and
a change rate calculation step of calculating a score change rate for each of the two classes when the modeling unit holds the initial score;
A method of determining the driving environment of mobility using a terminal, comprising: According to claim 17,
The classification step is,
a second class determination step in which the modeling unit compares the score change rates of each of the two classes, and classifies a class with a relatively large score change rate as the surrounding environment of the mobility according to the comparison result;
A method of determining the driving environment of mobility using a terminal, comprising:

Images