KR102475878B1 - Compensation Device and Method for Vehicle Motion Sensor Capable of Frequent Correction for Each Lane Keeping Section Extracted Based on Image Information, and Medium Being Recorded with Program for Executing the Method - Google Patents

Abstract

Disclosed are a correction device and method for correcting a vehicle motion sensor capable of performing frequent correction for each lane keeping section extracted based on image information, and a recording medium on which a program for executing the same is recorded. The present invention includes a calculation unit for extracting a lane keeping time interval of the vehicle based on a photographed image of a camera unit capturing a lane of a road on which the vehicle is traveling, and the calculation unit is applied to a motion sensor for detecting motion of the vehicle. Calculate the correction value during the lane keeping time section for According to the present invention, a lane keeping time interval is calculated by analyzing a photographed image of a road captured in real time while a vehicle is driving, and a correction value of a motion sensor is calculated for each calculated lane keeping time interval for a relatively short driving interval. As the correction value of the motion sensor can be repeatedly calculated, the accuracy of the motion sensor can be greatly improved. In addition, according to the present invention, the correction value of the gyro sensor can be accurately calculated even in a curved driving section of the vehicle.

Description

Compensation Device and Method for Vehicle Motion Sensor Capable of Frequent Correction for each Lane Keeping Section Extracted Based on Image Information, and Medium Being Recorded with Program for Executing the Method}

The present invention relates to a correction device and correction method for a vehicle motion sensor capable of performing frequent correction for each lane keeping section extracted based on image information, and a recording medium on which a program for executing the correction is recorded, and more particularly, in the process of driving a vehicle. By analyzing the captured image of the road captured in real time, lane keeping time intervals are calculated, and motion sensor correction values are calculated for each calculated lane keeping time interval to repeatedly calculate motion sensor correction values for relatively short driving sections. As this becomes possible, not only can the accuracy of the motion sensor be greatly improved, but also a vehicle motion sensor correction device and correction method that can accurately calculate the correction value of the gyro sensor even in the curved driving section of the vehicle, and a program for executing the same This is related to the recorded recording medium.

Motion sensors such as accelerometers and gyroscopes that measure the motion of a vehicle in motion generally have a bias. Accordingly, in order to improve the accuracy of the motion sensor measurement value, the bias is estimated and calculated. The process of elimination is essential.

To this end, conventionally, when the vehicle travels a straight line section identified through the output of the gyro sensor and the heading measurement value of GNSS for a predetermined time (dt) interval, integration of the output data of the gyro sensor installed in the vehicle A method of estimating, as a bias, a proportional constant that causes a change in the cumulative heading value calculated through this to be 0 is used.

However, such a conventional method has a problem in that it must depend on the accuracy of a heading measurement value of GNSS because it needs to use a gyro sensor that already has a bias in the process of finding a straight line.

In addition, since the conventional method requires a fairly large value for the driving time (dt) on a straight line, it takes a considerable amount of time to estimate the bias, and above all, the gyro sensor can be corrected only when the vehicle is traveling on a straight line. there is

In addition, since the bias of the motion sensor changes quickly depending on the surrounding environment and conditions, the bias estimated through the conventional method that takes a considerable amount of time as described above has a problem in that accuracy is limited.

Therefore, an object of the present invention is to calculate a lane keeping time interval by analyzing a photographed image of a road captured in real time during a vehicle driving process, and by calculating a correction value of a motion sensor for each calculated lane keeping time interval, relatively short driving. As the correction value of the motion sensor can be repeatedly calculated for the section, the accuracy of the motion sensor can be greatly improved, and the correction value of the gyro sensor can be accurately calculated even in the curved driving section of the vehicle. An object of the present invention is to provide a motion sensor calibration device and calibration method, and a recording medium on which a program for executing the calibration device is recorded.

An apparatus for correcting a vehicle motion sensor according to the present invention for achieving the above object includes a calculation unit for extracting a lane keeping time interval of the vehicle based on a photographed image of a camera unit capturing a lane of a road on which the vehicle is traveling, , The calculation unit may calculate a correction value for the motion sensor for detecting motion according to driving of the vehicle during the lane keeping time interval.

Preferably, the calculation unit calculates the correction value based on a cumulative measurement value of the motion sensor during the lane keeping time interval.

The calculation unit may calculate a curvature value of a driving section corresponding to the lane keeping time section based on the location information of the vehicle in the lane keeping time section.

The calculation unit may calculate the correction value based on a curvature value of a driving section corresponding to the lane keeping time section.

Meanwhile, a method for correcting a vehicle motion sensor according to the present invention includes: (a) a correction device for a vehicle motion sensor, based on a captured image of a camera unit capturing a lane of a road on which a vehicle is traveling, during a lane keeping time period of the vehicle; Extracting; and (b) calculating, by the correction device, a correction value for a motion sensor for detecting a motion according to driving of the vehicle during the lane keeping time interval.

Preferably, in the step (b), the correction device calculates the correction value based on an accumulated measurement value of the motion sensor during the lane keeping time interval.

In addition, after the step (a) and before the step (b), the correction device determines the curvature value of the driving section corresponding to the lane keeping time section based on the location information of the vehicle in the lane keeping time section. The step of calculating is further included.

In the step (b), the correction device may calculate the correction value based on the curvature value of the driving section corresponding to the lane keeping time section.

Meanwhile, the recording medium according to the present invention is characterized in that a program for executing the above method is recorded.

According to the present invention, a lane keeping time interval is calculated by analyzing a photographed image of a road captured in real time while a vehicle is driving, and a correction value of a motion sensor is calculated for each calculated lane keeping time interval for a relatively short driving interval. As the correction value of the motion sensor can be repeatedly calculated, the accuracy of the motion sensor can be greatly improved.

In addition, according to the present invention, the correction value of the gyro sensor can be accurately calculated even in a curved driving section of the vehicle.

1 is a functional block diagram showing the structure of a correction device for a vehicle motion sensor according to an embodiment of the present invention;
2 is a flow chart illustrating an execution process of a method for correcting a vehicle motion sensor according to an embodiment of the present invention;
3 is a view showing a driving road photographed image during the execution of a method for correcting a vehicle motion sensor according to an embodiment of the present invention;
4 is a diagram explaining a method of analyzing a photographed image of a driving road in a process of executing a method of correcting a vehicle motion sensor according to an embodiment of the present invention; and
5 is a diagram illustrating a heading value, which is a driving direction value measured by a gyro sensor module according to a driving time of a vehicle.

Hereinafter, the present invention will be described in more detail with reference to the drawings. It should be noted that like elements in the drawings are indicated by like numerals wherever possible. In addition, detailed descriptions of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.

1 is a functional block diagram showing the structure of an apparatus for compensating a vehicle motion sensor according to an embodiment of the present invention. Referring to FIG. 1 , an apparatus 100 for compensating a vehicle motion sensor according to an embodiment of the present invention includes a motion sensor 110, a communication unit 130, a camera unit 150, and a calculation unit 170.

The motion sensor 110 is installed in a running vehicle to measure motion of the vehicle, and the motion sensor 110 may include an accelerometer module and a gyroscope module.

While the vehicle is driving, the acceleration value of the vehicle measured through the acceleration sensor module and the angular velocity value of the vehicle measured through the gyro sensor module are transmitted to the calculation unit 170 in real time.

In implementing the present invention, the motion sensor 110 is not provided in the vehicle motion sensor calibration device 100 according to the present invention, but is installed separately in an autonomous vehicle or operated in a state in which it is provided in a vehicle navigation device such as a smart phone Maybe.

On the other hand, the communication unit 130 is provided with a GNSS (Global Navigation Satellite System) module, the GNSS module calculates the location information of the driving vehicle in real time based on the radio signal received from the satellite, and the calculated location information is calculated by the calculation unit ( 170).

In implementing the present invention, the communication unit 130 is not provided in the vehicle motion sensor calibration device 100 according to the present invention, but is separately installed in an autonomous vehicle or operated in a state in which it is provided in a vehicle navigation device such as a smart phone. It could be.

The camera unit 150 captures a lane of the road on which the vehicle is traveling in real time, and transmits the captured image of the driving road to the calculation unit 170 in real time.

In implementing the present invention, the camera unit 150 may also be installed separately on the front of an autonomous vehicle or operated in a vehicle black box without being provided in the vehicle motion sensor calibration device 100 according to the present invention. will be.

Meanwhile, the calculation unit 170 analyzes the captured image of the road received in real time from the camera unit 150 during the driving process of the vehicle to determine the time of the vehicle's lane change, and determines the vehicle's lane keeping time interval until the time of the lane change. and a correction value for the motion sensor 110 is calculated for each calculated lane keeping time section.

The vehicle motion sensor calibration device 100 according to the present invention may be directly installed in an autonomous vehicle or may be additionally installed in a navigation device installed in a vehicle.

2 is a flowchart illustrating an execution process of a method for correcting a vehicle motion sensor according to an embodiment of the present invention. Hereinafter, with reference to FIGS. 1 and 2 , an execution process of a method for correcting a vehicle motion sensor according to an embodiment of the present invention will be described.

First, as the driving of the vehicle starts, the camera unit 150 captures a lane of the road on which the vehicle is driving, as shown in FIG. 3 , and transmits the captured image to the calculation unit 170 in real time.

Meanwhile, the GNSS module provided in the communication unit 130 calculates the location information of the driving vehicle in real time based on the radio signal received from the satellite, and transmits the calculated location information to the calculation unit 170 (S210).

The calculation unit 170 analyzes the photographed image of the road received from the camera unit 150 to determine a time point for the vehicle to change lanes (S230). In implementing the present invention, it is preferable that the camera unit 150 captures a road image in the direction of the line of sight looking at the road from the center of the front of the driving vehicle as shown in FIG. 3, and in this case, the calculation unit ( 170), as shown in FIG. 4, calculates the lane angle R, which is an angle between the vertical line L passing through the center point of the captured image and the lane adjacent to the vertical line L, and the calculated lane angle R and the following arithmetic Based on Equation 1, the lane change timing of the vehicle may be determined.

Specifically, the calculation unit 170 determines that the vehicle maintains its lane when the calculated lane angle R is within a predetermined reference angle (eg, ±10°) and thus satisfies Equation 1. , and the time point at which Equation 1 is not satisfied because the lane angle R is out of the predetermined reference angle range is determined as the time point at which the vehicle starts to change lanes, and thereafter, the lane angle R is determined to be The time point at which Equation 1 is satisfied again by coming within the reference angle range may be determined as the time point at which the vehicle has completed the lane change.

As such, the calculation unit 170 can continuously determine the start time of the lane change and the time of completion of the lane change during the driving process of the vehicle, and the time interval from the completion of the first lane change to the start of the second lane change. It is possible to extract as a lane keeping time interval (S250).

In addition, in implementing the present invention, the calculation unit 170 may extract a time interval in which the lane angle R continuously satisfies Equation 1 above as the lane keeping time interval.

As described above, according to the present invention, the calculation unit 170 may continuously extract lane keeping time sections in the driving process of the vehicle based on the road-captured image, and as a result, the motion sensor 110 detects the motion sensor 110 for each lane keeping time section. The correction value of can be calculated individually.

Meanwhile, in implementing the present invention, the calculation unit 170 calculates the curvature value of the vehicle driving section corresponding to the lane keeping time section extracted in the above-described step S250 in order to more accurately calculate the correction value of the motion sensor 110. It will be possible (S270).

Specifically, the calculation unit 170 determines the curvature value of the driving section in which the vehicle traveled during the lane keeping time section extracted in step S250 based on the location information of the driving vehicle calculated by the GNSS module in step S210 described above. and a correction value of the motion sensor 110 for the corresponding lane keeping time section may be calculated based on the curvature value of the driving section calculated as described above (S290).

Specifically, when the curvature value of the vehicle driving section corresponding to the lane keeping time section is calculated to be 0, the calculation unit 170 determines that the vehicle traveled in a straight section during the lane keeping time section.

At this time, the calculation unit 170 may calculate the cumulative measurement value of the gyro sensor module in the corresponding lane keeping time interval, and calculate a value that makes the calculated cumulative measurement value equal to 0 as a correction value of the gyro sensor module.

In addition, in implementing the present invention, when the curvature value K of the vehicle driving section corresponding to the lane keeping time section is calculated as a value other than 0, the calculation unit 170 determines that the vehicle curves the curved section during the lane keeping time section. judged to be driving.

At this time, the calculation unit 170 calculates the driving distance (L) of the vehicle during the lane keeping time interval based on the location information of the driving vehicle calculated by the GNSS module, and calculates the elapsed time (T) in the lane keeping time interval By calculating, the estimated angular velocity W of the vehicle in the corresponding lane keeping time section may be calculated based on Equation 2 below.

In addition, the calculation unit 170 calculates an average value of the angular velocities measured by the gyro sensor module in the corresponding lane keeping time interval, and a value for making the calculated average angular velocity equal to the estimated angular velocity (W) calculated by Equation 2 above. may be calculated as a correction value of the gyro sensor module.

In addition, in implementing the present invention, based on the location information of the driving vehicle calculated by the GNSS module, both the curvature value of the vehicle driving section corresponding to the lane keeping time section and the vehicle's driving distance in the lane keeping time section are When the calculation is 0, the calculation unit 170 may determine that the vehicle was in a stopped state during the lane keeping time section.

At this time, the calculation unit 170 may calculate the cumulative measurement value of the acceleration sensor module in the corresponding lane keeping time section, and calculate a value that makes the calculated cumulative measurement value become 0 as a correction value of the acceleration sensor module.

5 is a diagram illustrating a heading value h, which is a driving direction value measured by a gyro sensor module according to a driving time t of a vehicle. When the vehicle drives while maintaining its lane, the time_heading value graph is displayed as a straight line graph section as shown in FIG. It becomes a sum.

For example, when the vehicle driving section is a straight section and the curvature value is 0 and the deviation value of the gyro sensor module is 0, the slope of the straight line graph becomes 0 as indicated by the blue dotted line in FIG. 5 .

In addition, when at least one of the bias value of the gyro sensor module and the curvature value of the vehicle driving section is not 0, the time_heading value graph in the time section in which the vehicle travels while maintaining its lane is red in FIG. 5 . A time_heading value graph in a time section in which the vehicle changes lanes while driving is displayed as a dotted line portion and is displayed as a convex curve portion in FIG. 5 .

In addition, when the vehicle drives in a turn, such as when turning left, turning right, or driving at an intersection, the time_heading value graph is displayed in the form of a step-like graph for the turn driving time section.

Meanwhile, in implementing the present invention, as shown in FIG. 5 , the calculation unit 170 analyzes the shape of a graph output by the heading value continuously measured by the gyro sensor module according to the driving of the vehicle, thereby analyzing the lane keeping time of the driving vehicle. A section, a lane change time section, and a turn driving time section may be calculated.

In addition, in implementing the present invention, the calculation unit 170 may analyze the captured image of the road received from the camera unit 150 to calculate the lane keeping time interval, the lane changing time interval, and the turning driving time interval. .

Specifically, for example, the calculation unit 170 calculates a lane angle R, which is an angle formed between a vertical line L passing through the center point of the captured image and a lane adjacent to the vertical line, and the calculated lane angle R is obtained from the equation above. A time interval satisfying 1 may be calculated as a lane keeping time interval, and a time interval in which the lane angle R does not satisfy Equation 1 may be calculated as a lane changing time interval.

In addition, the calculation unit 170 calculates the level of the captured image from the time when the vertical line L passing through the center point of the captured image and an adjacent lane forming an angle within a predetermined reference angle (eg, ±10°) disappear from the captured image. The time interval until the vertical line (L) passing through the center point and the adjacent lane forming an angle within a predetermined reference angle (eg, ± 10 °) reappears in the captured image can be calculated as the turning driving time interval. will be.

As described above, according to the present invention, the operation unit 170 can classify and calculate lane keeping time intervals, lane changing time intervals, and turning driving time intervals of the vehicle based on the photographed image of the road received from the camera as follows.

- Lane Keeping Time Section -

Meanwhile, in implementing the present invention, the calculation unit 170 may calculate the deviation B1 of the gyro sensor module in the lane keeping time interval through Equation 3 below.

Here, B1 is the deviation of the gyro sensor module in the lane keeping time interval, HS is the amount of change in the heading value of the gyro sensor module in the lane keeping time interval, and _{H M_E} is the end point based on the map information at the end point of the lane keeping driving interval. The heading value of the road, H _{M_S} , is the heading value of the road at the point of view based on the map information at the time of the lane keeping driving section, and T is the elapsed time in the lane keeping time section.

In carrying out the present invention, it is preferable that the calculation unit 170 stores map information in which heading value information for each road on which a vehicle travels is recorded.

Meanwhile, the calculation unit 170 may calculate a value that causes the deviation B1 of the gyro sensor module calculated by Equation 3 to be zero as a correction value of the gyro sensor module.

- Lane change time section -

In implementing the present invention, when the driving distance in the lane keeping time interval is not a sufficient distance (for example, 200 m), the operation unit 170 performs the lane keeping time interval prior to the specific lane change time interval (the first lane The deviation (B2) of the gyro sensor module in the time period in which the lane keeping time period (second lane keeping time period) is integrated with the lane keeping time period after the lane change time period and the lane keeping time period after the lane change time period may be calculated through Equation 4 below. .

Here, B2 is the deviation of the gyro sensor module in the time interval integrating the adjacent lane keeping time intervals, and H _S is the gyro sensor in the time interval from the start of the first lane keeping time interval to the end point of the second lane keeping time interval. Change amount of the heading value of the module, H _{M_E1} is the heading value of the end road based on the map information at the end of the first lane keeping driving section, H _{M_S1} is the starting point road based on the map information at the starting point of the first lane keeping driving section Heading value, H _{M_E2} is the heading value of the end road based on the map information at the end of the second lane maintaining driving section, H _{M_S2} is the heading value of the starting road based on the map information at the starting point of the second lane maintaining driving section, T is the sum of the elapsed time in the first lane keeping time interval and the elapsed time in the second lane keeping time interval.

Meanwhile, the calculation unit 170 may calculate a value that causes the deviation B2 of the gyro sensor module calculated by Equation 4 to be zero as a correction value of the gyro sensor module.

- Rotating Travel Time Section -

In implementing the present invention, the calculation unit 170 may calculate the deviation B3 of the gyro sensor module in the rotational driving time section through Equation 5 below.

Here, B3 is the deviation of the gyro sensor module in the rotational travel time interval, H _S is the amount of change in the heading value of the gyro sensor module in the rotational travel time interval, and H _{M_E'} is based on the map information at the end point of the rotation travel time interval. The heading value of the end road, H _{M_S'} , is the heading value of the starting road based on the map information at the starting point of the turn travel time section, T is the elapsed time in the turn travel time section.

Meanwhile, the calculation unit 170 may calculate a value that causes the deviation B3 of the gyro sensor module calculated by Equation 5 to be zero as a correction value of the gyro sensor module.

In implementing the present invention, a program for executing the vehicle motion sensor calibration method according to the present invention is installed in the vehicle motion sensor calibration device 100 or is a module constituting an autonomous driving program for an autonomous vehicle. It may be directly installed in the operation control device of the vehicle, and may also be directly installed in a smart phone or navigation device as a module constituting a navigation program.

In addition, in carrying out the present invention, the program for executing the method for correcting the vehicle motion sensor according to the present invention may be recorded in various computer-readable recording media or stored in a server that transmits the corresponding program through a network. .

Terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Although preferred embodiments and application examples of the present invention have been shown and described above, the present invention is not limited to the specific embodiments and application examples described above, and the present invention is not departing from the gist of the present invention claimed in the claims. Various modifications and implementations are possible by those skilled in the art, and these modifications should not be individually understood from the technical spirit or perspective of the present invention.

100: correction device, 110: motion sensor,
130: communication unit, 150: camera unit,
170: calculation unit.

Claims (7)

Calculating a lane angle, which is an angle formed by a vertical line passing through a center point of the captured image and a lane adjacent to the vertical line, based on a photographed image of a camera unit capturing a lane of a road on which a vehicle is traveling;
If the calculated lane angle is within a predetermined reference angle range, it is extracted as a lane keeping time section in which the lane of the vehicle is maintained;
Extracting a time interval from a point in time when the calculated lane angle deviates from the predetermined reference angle range to a point in time when the calculated lane angle enters the predetermined reference angle range as a lane change time interval;
A calculation unit for extracting a time interval from the point in time when the adjacent lane disappears from the captured image to the point in time when the adjacent lane reappears in the captured image as a lane turning driving time interval
Including,
The calculation unit calculates a deviation of the motion sensor based on a heading value of a motion sensor that senses motion according to driving of the vehicle, and based on the deviation, the lane keeping time section, the lane change time section, or the lane turning time section. Calculate a correction value during any one of the travel time sections,
The correction device of the vehicle motion sensor, wherein the calculation unit calculates the correction value using different methods according to which time period the driving of the vehicle belongs to. According to claim 1,
The calculation unit may include the amount of change in the heading value in the lane keeping time section, the heading value at the end of the lane keeping driving section, the heading value at the start of the lane keeping driving section, and the progress in the lane keeping driving section. and calculating a deviation of the motion sensor in the lane keeping time interval based on time. According to claim 1,
The calculation unit may include a change amount of the heading value in a time section from the start of the first lane keeping time section to the end of the second lane keeping time section, the heading value at the end point of the first lane keeping driving section, and the second lane keeping time section. The heading value at the start of the first lane keeping driving section, the heading value at the end of the second lane keeping driving section, the heading value at the start of the second lane keeping driving section, and the first lane keeping time section and calculating a deviation of the motion sensor in the lane-changing time interval based on a sum of an elapsed time in and an elapsed time in the second lane keeping time interval. According to claim 1,
The calculation unit may include the amount of change in the heading value in the rotational travel time interval, the heading value at the end of the rotational travel time interval, the heading value at the start of the rotational travel time interval, and the lapse in the rotational travel time interval. and calculating a deviation of the motion sensor in the lane turning driving time section based on time. According to claim 1,
The calculation unit calculates a curvature value of a driving section corresponding to a lane keeping time section, a lane change time section, or a lane turning driving time section;
and calculating the correction value based on the heading value and the curvature value. (a) The vehicle motion sensor correction device calculates a lane angle, which is an angle between a vertical line passing through the center point of the captured image and a lane adjacent to the vertical line, based on a captured image of a camera unit that captures a lane of the road on which the vehicle is traveling. calculate,
If the calculated lane angle is within a predetermined reference angle range, it is extracted as a lane keeping time section in which the lane of the vehicle is maintained;
Extracting a time interval from a point in time when the calculated lane angle deviates from the predetermined reference angle range to a point in time when the calculated lane angle enters the predetermined reference angle range as a lane change time interval;
extracting a time interval from a point in time when the adjacent lane disappears from the captured image to a point in time when the adjacent lane reappears in the captured image as a lane turning driving time interval; and
(b) the correction device calculates a deviation of the motion sensor based on a heading value of a motion sensor that senses motion according to driving of the vehicle, and based on the deviation, the lane keeping time interval and the lane changing time Calculating a correction value during a section or any one of the lane turning driving time sections
including,
The correction device calculates the correction value using different methods according to which time period the driving of the vehicle belongs to. A recording medium on which a program for executing the method of claim 6 is recorded.

Images