Abnormality detection method and device for lane keeping assistance system

Abstract

The present invention relates to an abnormality detection method and apparatus for a lane keeping assist system, comprising: a step in which an electronic device confirms a first steering angle on the basis of image data for the front of a traveling vehicle; a step in which the electronic device confirms a second steering angle on the basis of sensing data acquired from a steering sensor provided in the vehicle; and a step in which the electronic device performs abnormality detection of the lane keeping assist system using the first steering angle and the second steering angle, and can also be applied to other embodiments.

Description

Technical Field

[0001] The invention relates to an abnormality detection method and device for a lane keeping assist system. Background Art

[0002] Recently, vehicles equipped with advanced driver assist systems (ADAS) that can actively provide information on vehicle status, driver status, and surrounding environment are being actively studied to reduce the burden on drivers and improve convenience.

[0003] Among these advanced driver assistance systems, lane following assist (LFA) and lane keeping assist (LKA) are also used in self-driving cars and are functions that help drivers drive more safely.

[0004] At this time, the lane keeping assist system refers to a driving convenience system that controls the steering so that the vehicle can keep in the center of the road while driving. However, when the lane keeping assist system functions abnormally, the vehicle cannot accurately keep in the center of the lane, and the vehicle may sway from side to side, causing the driver to feel dizzy and other inconveniences.

[0005] As described above, in the event of inconvenience caused by abnormality of the lane keeping assist system while driving, the driver will visit a repair station to resolve the problem using methods such as correcting the front camera, wheel alignment, checking air pressure, etc. However, to date, no instructions have been provided on when to maintain the lane keeping assist system, so that drivers usually visit the repair station only when they feel inconvenience. Summary of the invention

[0006] Problem that the invention aims to solve

[0007] An embodiment of the present invention for solving the above-mentioned previous problems provides a lane keeping assist system abnormality detection method and device, which uses the steering angle inferred from the image data in front of the moving vehicle and the actual measured steering angle to confirm whether the lane keeping assist system is abnormal, thereby providing a maintenance alert before the user experiences the inconvenience caused by the abnormality of the lane keeping assist system.

[0008] Means used to solve problems

[0009] According to an embodiment of the present invention, a method for detecting an abnormality of a lane keeping assist system includes the following steps: a step in which an electronic device confirms a first steering angle based on image data in front of a moving vehicle; a step in which the electronic device confirms a second steering angle based on sensor data obtained from a steering sensor provided on the vehicle; and a step in which the electronic device uses the first steering angle and the second steering angle to perform abnormality detection of a lane keeping assist system.

[0010] In addition, the step of performing abnormality detection of the lane keeping assist system includes the step of confirming whether the road on which the vehicle is traveling is straight based on the image data.

[0011] In addition, after the step of confirming whether the road is straight or not, the method further includes: when the traveling road is a straight road, the electronic device confirms the residual difference between the first steering angle and the second steering angle during a first critical time.

[0012] In addition, after the step of confirming whether the road is straight or not, the method further includes: when the traveling road is a curved road, the electronic device confirms the residual difference between the first steering angle and the second steering angle during a second critical time.

[0013] In addition, the step of performing abnormality detection of the lane keeping assist system includes: a step of calculating a root mean square difference using a residual of the first steering angle and the second steering angle.

[0014] In addition, the step of performing abnormality detection of the lane keeping assist system is a step of confirming that an abnormality has occurred in the lane keeping assist system when the RMS difference is greater than a critical value.

[0015] In addition, after executing the step of detecting abnormality of the lane keeping assist system, the method further includes: displaying abnormality occurrence information of the lane keeping assist system.

[0016] In addition, the step of confirming the first steering angle includes: the step of inferring the steering angle by analyzing the image data using a deep learning algorithm; the step of generating a steering control signal for controlling the steering of the vehicle by analyzing the image data; the step of performing linear interpolation of the steering angle based on the steering control signal; and the step of generating the first steering angle by comparing the value of the linear interpolation with the steering control signal.

[0017] In addition, the step of calculating the root mean square difference includes: the step of generating a curve graph of the first steering angle and the second steering angle confirmed during the first critical time or the second critical time according to whether the road is straight or not; the step of dividing the critical time of vehicle travel into at least one interval based on the intersection position of the values ​​of the first steering angle and the second steering angle; and the step of confirming the maximum difference among the differences confirmed in the at least one interval as the residual of the corresponding interval.

[0018] In addition, the second critical time is a time longer than the first critical time.

[0019] At the same time, according to an embodiment of the present invention, the abnormality detection device of the lane keeping assist system includes: a camera, a camera for obtaining image data for the front of a moving vehicle; and a control unit, which confirms a first steering angle based on the image data, and confirms a second steering angle based on sensor data obtained from a steering sensor provided on the vehicle, and uses the first steering angle and the second steering angle to perform abnormality detection of the lane keeping assist system.

[0020] Furthermore, the control unit confirms whether the road on which the vehicle is traveling is straight based on the image data.

[0021] Furthermore, when the traveling road is a straight road, the control unit checks a residual difference between the first steering angle and the second steering angle during a first critical time.

[0022] In addition, when the traveling road is a curved road, the control unit checks a residual difference between the first steering angle and the second steering angle during a second critical time.

[0023] In addition, the control unit calculates a root mean square difference using a residual between the first steering angle and the second steering angle.

[0024] In addition, when the RMS difference is greater than a critical value, the control unit confirms that an abnormality has occurred in the lane keeping assist system.

[0025] In addition, a display unit is included, and the control unit displays abnormality occurrence information of the lane keeping assist system through the display unit.

[0026] In addition, the control unit performs linear interpolation of the steering angle based on the steering angle estimated by analyzing the image data using a deep learning algorithm and a steering control signal generated to control the steering of the vehicle, and generates the first steering angle by comparing the value of the linear interpolation with the steering control signal.

[0027] In addition, the control unit generates a curve graph for the first steering angle and the second steering angle confirmed during the first critical time period or the second critical time period according to whether the road is straight or not, and divides the critical time of vehicle travel into at least one interval based on the position where the values ​​of the first steering angle and the second steering angle intersect, and confirms the maximum difference among the differences confirmed in the at least one interval as the residual of the corresponding interval.

[0028] In addition, the second critical time is a time longer than the first critical time.

[0029] Effects of the Invention

[0030] As described above, the abnormality detection method and device of the lane keeping assist system according to the present invention have the following effects: it can use the steering angle inferred from the image data in front of the moving vehicle and the actually measured steering angle to confirm whether the lane keeping assist system is abnormal, thereby providing a maintenance alert before the user experiences the inconvenience caused by the abnormality of the lane keeping assist system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 FIG. 1 is a diagram showing a main structure of an electronic device that performs abnormality detection of a lane keeping assist system according to an embodiment of the present invention.

[0032] Figure 2 is a flowchart for illustrating an abnormality detection method of a lane keeping assist system according to an embodiment of the present invention.

[0033] Figure 3 It is a detailed flow chart for explaining a method for confirming a first steering angle through image data analysis according to an embodiment of the present invention.

[0034] Figure 4 is a detailed flowchart for explaining a method for detecting abnormality of a lane keeping assist system using a steering angle according to an embodiment of the present invention.

[0035] Figure 5 is a graph for explaining an abnormality detection method of a lane keeping assist system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Figure 1 The detailed description disclosed below is intended to describe exemplary embodiments of the present invention, but is not intended to represent the only implementation method in which the present invention can be implemented. In order to clearly illustrate the present invention, parts not related to the description may be omitted from the drawings, and the same reference numerals may be used for the same or similar components throughout the specification.

[0037] Figure 1FIG. 1 is a diagram showing a main structure of an electronic device that performs abnormality detection of a lane keeping assist system according to an embodiment of the present invention.

[0038] Reference Figure 1 The electronic device 100 according to the present invention may include: a communication unit 110 , an input unit 120 , a sensor unit 130 , a camera 140 , a display unit 150 , a memory 160 and a control unit 170 .

[0039] The communication unit 110 can provide the control unit 170 with the sensing data obtained from the sensor unit 130 and the image data obtained from the camera 140 by communicating with the sensor unit 130 and the camera 140. To this end, the communication unit 110 can perform communications such as controller area network (CAN), near field communication (NFC), Zigbee, etc., and can perform serial communications such as RS-232.

[0040] The input unit 120 generates input data in response to a user's input to the vehicle. To this end, the input unit 120 may include a key pad, a dome switch, a touch panel, a touch key, a button, and the like.

[0041] The sensor unit 130 may include a steering sensor for sensing a steering angle. Based on this, the sensor unit 130 may obtain the steering angle actually confirmed when the vehicle is traveling and provide it to the control unit 170. To this end, the sensor unit 130 may perform wireless communication such as Bluetooth, Bluetooth low energy (BLE), controller area network (CAN), near field communication (NFC), Zigbee, etc. with the control unit 170, and may perform serial communication such as RS-232.

[0042] The camera 140 is disposed on the vehicle to obtain image data of the vehicle's external environment. In particular, the camera 140 can obtain image data of the front of the moving vehicle and provide it to the control unit 170.

[0043] The display unit 150 displays display data associated with actions executed in the electronic device 100. The display unit 150 includes a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a micro electromechanical system (MEMS) display, and an electronic paper display. The display unit 150 can be combined with the input unit 120 to realize a touch screen capable of inputting.

[0044] The memory 160 stores an action program for the electronic device 100 to operate. In particular, the memory 160 may store a deep learning algorithm for confirming a first steering angle by analyzing image data in front of a moving vehicle. In addition, the memory 160 may store an algorithm for calculating a residual by comparing the first steering angle with a second steering angle included in the sensing data obtained from the sensor unit 130 and calculating a root mean square error (RMSE) value of the obtained residual. The memory 160 may store a critical value used as a reference for detecting an abnormality of the lane keeping assist system.

[0045] The control unit 170 confirms the first steering angle based on the image data in front of the moving vehicle, confirms the second steering angle based on the sensing data obtained from the steering sensor in the sensor unit 130, and performs abnormality detection of the lane keeping assist system using the first steering angle and the second steering angle. To this end, the control unit 170 may include a generating unit 171, a measuring unit 172, a comparing unit 173, and an abnormality detecting unit 174.

[0046] The generation unit 171 identifies the lanes marked on the road on which the vehicle is traveling by analyzing the image data of the moving vehicle obtained from the camera 140, thereby generating a steering control signal that can substantially control the steering angle of the vehicle. In addition, the generation unit 171 calls the deep learning algorithm stored in the memory 160, and applies the image data of the moving vehicle obtained from the camera 140 to the deep learning algorithm. The generation unit 171 can analyze the image data using a deep learning algorithm to infer the steering angle corresponding to the lane marked on the moving road. At this time, the deep learning algorithm can be Pilot-Net, FOLOLane (focus on local lane), CLRNet (cross layer refinement network), LaneNet, RFSA (recurrent feature-shift aggregator) and other algorithms.

[0047] The generator 171 matches the estimated steering angle with the generated steering control signal to perform linear interpolation. For example, when the period of estimating the steering angle using image data is 50 ms and the period of generating the steering control signal is 10 ms, the generator 171 can match the estimated steering angle with the period of acquiring the steering control signal to perform linear interpolation. To this end, the generator 171 can confirm the value corresponding to the steering control signal.

[0048] The generator 171 compares the linear interpolation value with the generated steering control signal. The generator 171 generates a first steering angle based on the comparison result of the stage. For example, the generator 171 can generate the first steering angle by calculating the intermediate value of the residual between the linear interpolation value and the steering control signal value.

[0049] In addition, the generating unit 171 may confirm whether the road on which the vehicle is traveling is a straight road based on the analysis result of the image data and provide the result to the comparing unit 173 .

[0050] The measuring unit 172 confirms the second steering angle based on the sensing data acquired from the sensor unit 130 .

[0051] The comparison unit 173 compares the first steering angle with the second steering angle. More specifically, the comparison unit 173 confirms whether the traveling road provided by the generation unit 171 is straight or not. When the traveling road is a straight road, the comparison unit 173 confirms the residual of the first steering angle and the second steering angle during the first critical time. On the contrary, when the traveling road is a curved road, the comparison unit 173 confirms the residual of the first steering angle and the second steering angle during the second critical time. At this time, the first critical time may be 20 seconds, and the second critical time may be 30 seconds.

[0052] The comparison unit 173 calculates the root mean square error (RMSE) value of the residual. More specifically, the comparison unit 173 can generate a graph of changes in the first steering angle and the second steering angle during the first critical time or the second critical time. The comparison unit 173 can divide the critical time of vehicle travel into at least one interval based on the position where the graphs for the first steering angle and the second steering angle intersect. The comparison unit 173 confirms the maximum difference among the differences confirmed in at least one interval as the residual of the corresponding interval, and can calculate the root mean square difference using the residuals confirmed in each interval.

[0053] When the calculated RMS difference is greater than the critical value, the abnormality detection unit 174 confirms that an abnormality is detected in the lane keeping assist system, and then generates information notifying the abnormality and displays it on the display unit 150. On the contrary, when the calculated RMS difference is less than the critical value, the abnormality detection unit 174 confirms that no abnormality is detected in the lane keeping assist system, and then confirms in real time or periodically whether an abnormality is detected in the lane keeping assist system until the vehicle stops traveling.

[0054] Figure 2 is a flowchart for illustrating an abnormality detection method of a lane keeping assist system according to an embodiment of the present invention.

[0055] Reference Figure 2 In step 201, the control unit 170 confirms whether the vehicle is running. According to the confirmation result of step 201, if the vehicle is running, the control unit 170 executes step 203, and if the vehicle is not running, the control unit 170 continues to confirm whether the vehicle is running.

[0056] In step 203, the control unit 170 analyzes the image data acquired from the camera 140. In this case, the image data may be image data of the road ahead of the moving vehicle.

[0057] In step 205, the control unit 170 determines the first steering angle by analyzing the image data. Figure 3 Let's explain it in more detail. Figure 3 It is a detailed flow chart for explaining a method for confirming a first steering angle through image data analysis according to an embodiment of the present invention.

[0058] Reference Figure 3In step 301, the control unit 170 analyzes the image data by applying the deep learning algorithm to the image data, so that the steering angle corresponding to the lane marked on the road can be estimated. In step 303, the control unit 170 can generate a steering control signal that can substantially control the steering angle of the vehicle by confirming the lane marked on the road. At this time, the deep learning algorithm can be Pilot-Net, FOLOLane (focus on local lane), CLRNet (cross layer refinement network), LaneNet, RFSA (recurrent feature-shift aggregator) and other algorithms.

[0059] In step 305, the control unit 170 performs linear interpolation by matching the steering angle estimated in step 301 with the steering control signal generated in step 303. For example, when the period of estimating the steering angle using image data is 50 ms and the period of generating the steering control signal is 10 ms, the control unit 170 may match the estimated steering angle with the period of acquiring the steering control signal to perform linear interpolation. To this end, the control unit 170 may confirm the value corresponding to the steering control signal.

[0060] In step 307, the control unit 170 compares the value of the linear interpolation with the steering control signal generated in step 303. In step 309, the control unit 170 generates a first steering angle based on the comparison result of step 307 and returns the first steering angle. Figure 2 Step 207. For example, the control portion 170 may generate the first steering angle by calculating an intermediate value of a residual between a value performing linear interpolation and a value of the steering control signal.

[0061] In step 207, the control unit 170 may confirm the second steering angle based on the sensing data acquired from the sensor unit 130. At this time, the sensor unit 130 may include a steering sensor for sensing a steering angle, and the sensing data may be sensing data related to the steering angle.

[0062] In step 209, the control unit 170 uses the first steering angle and the second steering angle to detect whether the lane keeping assist system is abnormal. Figure 4 Let's explain it in more detail. Figure 4 is a detailed flowchart for explaining a method for detecting abnormality of a lane keeping assist system using a steering angle according to an embodiment of the present invention.

[0063] Reference Figure 4In step 401, the control unit 170 confirms whether the road on which the vehicle is traveling is a straight road. At this time, the control unit 170 can use the lane confirmed from the image data to confirm whether the road on which the vehicle is traveling is straight. According to the confirmation result of step 401, if the road on which the vehicle is traveling is a straight road, the control unit 170 executes step 403, and if it is a curved road, it executes step 405.

[0064] When the traveling road is a straight road, in step 403, the control unit 170 confirms the residual of the first steering angle and the second steering angle during the first critical time and executes step 407. On the contrary, when the traveling road is a curved road, in step 405, the control unit 170 confirms the residual of the first steering angle and the second steering angle during the second critical time and executes step 407. At this time, the first critical time may be 20 seconds, and the second critical time may be 30 seconds.

[0065] In step 407, the control unit 170 calculates the root mean square error (RMSE) value of the residual. More specifically, the control unit 170 may generate a graph of changes in the first steering angle and the second steering angle during the first critical time or the second critical time. The control unit 170 may divide the critical time of vehicle travel into at least one interval based on the position where the graphs for the first steering angle and the second steering angle intersect. The control unit 170 confirms the maximum difference among the differences confirmed in at least one interval as the residual of the corresponding interval, and may calculate the root mean square difference using the residuals confirmed in each interval.

[0066] In step 409, if the RMS value of the residuals is greater than the critical value, the control unit 170 executes step 411. If the RMS value of the residuals is less than the critical value, the control unit 170 returns to step 412. Figure 2 In step 409, when the RMS value of the residual is greater than the critical value, in step 411, the control unit 170 confirms that an abnormality is detected in the lane keeping assist system and executes step 413. In step 413, the control unit 170 generates information notifying that an abnormality is detected in the lane keeping assist system and displays it on the display unit 150 and returns Figure 2 Step 211.

[0067] In step 211, if the termination of the driving of the vehicle is confirmed, the control unit 170 terminates the corresponding process. If the termination of the driving is not confirmed, the control unit 170 returns to step 203 and re-executes steps 203 to 209. Thus, the present invention has the following effects: it can confirm whether the lane keeping assist system is abnormal by using the steering angle estimated from the image data in front of the driving vehicle and the steering angle actually measured, so that a maintenance alarm can be provided before the user experiences the inconvenience caused by the abnormality of the lane keeping assist system.

[0068] Figure 5 is a graph for explaining an abnormality detection method of a lane keeping assist system according to an embodiment of the present invention.

[0069] Reference Figure 5 The control unit 170 confirms a graph of a first steering angle 501 confirmed based on the image data and a second steering angle 503 confirmed in the sensing data obtained from the sensor unit 130. In the graph, the x-axis may represent time and the y-axis may represent the steering angle. For example, Figure 5 It may be a graph representing a graph for the first steering angle and the second steering angle confirmed during a first critical time (for example, 20 seconds) in a state when the vehicle is traveling on a straight road.

[0070] The control unit 170 calculates the residual between the two curves. The control unit 170 sets the curves for the two steering angles to the first interval 511 immediately after (0, 0), the second interval 512 immediately after the first interval 511, and the third interval 513 immediately after the second interval 512, and calculates the residual in each interval. At this time, among the residuals confirmed in each interval, the control unit 170 can calculate the residual with the maximum value as the first residual a, the second residual b, and the third residual c.

[0071] The control unit 170 calculates the root mean square error (RMSE) values ​​of the first residual a, the second residual b, and the third residual c. When at least one of the root mean square error values ​​of the three residuals is greater than a critical value (e.g., 2), the control unit 170 may confirm that an abnormality is detected in the lane keeping assist system. At the same time, the critical value as the abnormality detection standard of the lane keeping assist system can be changed and applied.

[0072] The embodiments of the present invention disclosed in this specification and the drawings are only presented as specific examples to facilitate easy explanation of the description of the present invention and help understand the present invention, and are not intended to limit the scope of the present invention. Therefore, the scope of the present invention should be interpreted as all changes or deformations derived from the technical ideas of the present invention in addition to the embodiments disclosed herein should be included in the scope of the present invention.

Claims

1. A method for detecting abnormality of a lane keeping assist system, characterized in that: include: The electronic device determines a first steering angle based on image data in front of the moving vehicle; The electronic device confirms a second steering angle based on sensing data obtained from a steering sensor provided on the vehicle; as well as The electronic device uses the first steering angle and the second steering angle to perform a step of abnormality detection of a lane keeping assist system.

2. The abnormality detection method of the lane keeping assist system according to claim 1, characterized in that: The steps of performing abnormality detection of the lane keeping assist system include: Based on the image data, it is confirmed whether the road on which the vehicle is traveling is a straight line.

3. The abnormality detection method of the lane keeping assist system according to claim 2, characterized in that: After the step of confirming whether the road is a straight line, the method further includes: When the traveling road is a straight road, the electronic device confirms a residual difference between the first steering angle and the second steering angle during a first critical time.

4. The abnormality detection method of the lane keeping assist system according to claim 2, characterized in that: After the step of confirming whether the road is a straight line, the method further includes: When the traveling road is a curved road, the electronic device confirms a residual difference between the first steering angle and the second steering angle during a second critical time.

5. The abnormality detection method of the lane keeping assist system according to claim 2, characterized in that: The steps of performing abnormality detection of the lane keeping assist system include: The step of calculating a root mean square difference using a residual of the first steering angle and the second steering angle.

6. The abnormality detection method of the lane keeping assist system according to claim 5, characterized in that: The steps of performing abnormality detection of the lane keeping assist system are: When the RMS difference is greater than a critical value, confirming that an abnormality has occurred in the lane keeping assist system.

7. The abnormality detection method of the lane keeping assist system according to claim 6, characterized in that: After executing the abnormality detection step of the lane keeping assist system, the method further includes: The step of displaying abnormality occurrence information of the lane keeping assist system.

8. The abnormality detection method of the lane keeping assist system according to claim 1, characterized in that: The step of confirming the first steering angle comprises: The step of inferring the steering angle by analyzing the image data using a deep learning algorithm; A step of generating a steering control signal for controlling the steering of the vehicle by analyzing the image data; Based on the steering control signal, performing a step of linear interpolation of the steering angle; and The step of generating the first steering angle by comparing the value performing the linear interpolation with the steering control signal.

9. The abnormality detection method of the lane keeping assist system according to claim 5, characterized in that: The step of calculating the root mean square difference comprises: generating a graph of the first steering angle and the second steering angle confirmed during a first critical time or during a second critical time according to whether the road is a straight line; A step of dividing a critical time of vehicle travel into at least one interval based on a position where the values ​​of the first steering angle and the second steering angle intersect; and The step of confirming the maximum difference among the differences confirmed in the at least one interval as the residual of the corresponding interval.

10. The abnormality detection method of the lane keeping assist system according to claim 9, characterized in that: The second critical time is a time longer than the first critical time.

11. An electronic device for abnormality detection of a lane keeping assist system, characterized in that: include: A camera, which obtains image data in front of a moving vehicle; as well as The control unit confirms a first steering angle based on the image data, confirms a second steering angle based on sensor data obtained from a steering sensor provided on the vehicle, and performs abnormality detection of the lane keeping assist system using the first steering angle and the second steering angle.

12. The electronic device for abnormality detection of a lane keeping assist system according to claim 11, characterized in that: The control unit confirms whether the road on which the vehicle is traveling is a straight line based on the image data.

13. The electronic device for abnormality detection of a lane keeping assist system according to claim 12, characterized in that: When the traveling road is a straight road, the control unit confirms a residual difference between the first steering angle and the second steering angle during a first critical time.

14. The electronic device for abnormality detection of a lane keeping assist system according to claim 12, characterized in that: When the traveling road is a curved road, the control unit confirms a residual difference between the first steering angle and the second steering angle during a second critical time.

15. The electronic device for abnormality detection of a lane keeping assist system according to claim 12, characterized in that: The control unit calculates a root mean square difference using a residual between the first steering angle and the second steering angle.

16. The electronic device for abnormality detection of a lane keeping assist system according to claim 15, characterized in that: When the RMS difference is greater than a critical value, the control unit confirms that an abnormality has occurred in the lane keeping assist system.

17. The electronic device for abnormality detection of a lane keeping assist system according to claim 16, characterized in that: Also includes: Display unit; The control unit displays abnormality occurrence information of the lane keeping assist system through the display unit.

18. The electronic device for abnormality detection of a lane keeping assist system according to claim 11, characterized in that: The control unit performs linear interpolation of the steering angle based on the steering angle and the steering control signal, and generates the first steering angle by comparing the value of the linear interpolation with the steering control signal. The steering angle is estimated by analyzing the image data using a deep learning algorithm, and the steering control signal is generated to control the steering control of the vehicle.

19. The electronic device for abnormality detection of a lane keeping assist system according to claim 12, characterized in that: The control unit generates a curve graph for the first steering angle and the second steering angle confirmed during a first critical time period or a second critical time period according to whether the road is a straight line, divides the critical time of vehicle travel into at least one interval based on the position where the values ​​of the first steering angle and the second steering angle intersect, and confirms the maximum difference among the differences confirmed in the at least one interval as the residual of the corresponding interval.

20. The electronic device for abnormality detection of a lane keeping assist system according to claim 19, characterized in that: The second critical time is a time longer than the first critical time.

Images