Autonomous vehicle

Abstract

To reduce power consumption.SOLUTION: The autonomous vehicle includes a camera arranged to capture an image of a road surface, a light emitting diode that irradiates the road surface with light, an auxiliary storage device that stores map data in which map image data obtained by capturing an image of the road surface in advance and position information are associated with each other, an internal sensor, and a control unit. The control unit estimates a self-position by odometry from a detection result of the internal sensor, causes the light emitting diode to emit light and causes the camera to capture an image every time the autonomous traveling vehicle moves a certain distance, and corrects the self-position by collating image data acquired from the camera with map image data.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to autonomous vehicles. [Background technology]

[0002] The autonomous vehicle disclosed in Patent Document 1 includes a camera, a light-emitting diode, a storage device, and a control unit. The camera is positioned to capture an image of the road surface. The light-emitting diode irradiates the road surface with light. The storage device stores map data. The map data is data that links map image data, which is a pre-image of the road surface, with location information. The control unit acquires the image data from the camera. The control unit estimates the vehicle's own location by matching the image data with the map image data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-166853 Summary of the Invention [Problem to be solved by the invention]

[0004] Estimating the self-location requires the light emitting diode to emit light and the camera to capture an image, and the more frequently the light emitting diode emits light and the camera captures an image, the greater the power consumption. [Means for solving the problem]

[0005] An autonomous vehicle that solves the above problem is an autonomous vehicle that includes a camera positioned to capture images of the road surface, a light-emitting diode that irradiates light onto the road surface, a storage device that stores map data that links map image data of the road surface that has been previously captured with location information, an internal sensor, and a control unit, wherein the control unit estimates its own position using odometry from the detection results of the internal sensor, causes the light-emitting diode to emit light and the camera to take images every time the autonomous vehicle moves a certain distance, and corrects its own position by comparing the image data obtained from the camera with the map image data.

[0006] The control unit corrects the autonomous vehicle's position by comparing the image data with map image data every time the autonomous vehicle travels a certain distance. If the autonomous vehicle is traveling at a low speed, it takes a long time for the autonomous vehicle to travel a certain distance, so the frequency with which the light-emitting diodes emit light and the cameras capture images decreases. Therefore, power consumption can be reduced compared to when image data is compared with map image data every certain time. [Effects of the Invention]

[0007] According to the present invention, power consumption can be reduced. [Brief explanation of the drawings]

[0008] [Figure 1] Figure 1 is a side view of an autonomous vehicle. [Figure 2] Figure 2 is a schematic diagram of an autonomous vehicle. [Figure 3] FIG. 3 is a flowchart showing the self-position estimation control. DETAILED DESCRIPTION OF THE INVENTION

[0009] An embodiment of an autonomous vehicle will now be described. <Autonomous Vehicles> As shown in FIG. 1, autonomous vehicle 10 includes a vehicle body 11, drive wheels 21, and steering wheels 31. Autonomous vehicle 10 may be a passenger vehicle or a transport vehicle. Transport vehicles include towing tractors and forklifts. Autonomous vehicle 10 may be capable of only autonomous driving, or may be capable of switching between autonomous driving and manual driving.

[0010] As shown in FIG. 2, autonomous vehicle 10 includes traction motor 22, traction motor driver 23, steering motor 32, and steering motor driver 33. Traction motor 22 is a motor for rotating drive wheels 21. Traction motor driver 23 drives traction motor 22. Drive of traction motor 22 rotates drive wheels 21, causing autonomous vehicle 10 to travel. Steering motor 32 is a motor for steering steering wheels 31. Steering motor driver 33 drives steering motor 32. Drive of steering motor 32 steers steering wheels 31, causing autonomous vehicle 10 to turn.

[0011] The autonomous vehicle 10 is equipped with an internal sensor 41. The internal sensor 41 includes at least one of an encoder and an inertial measurement unit. The encoder detects, for example, the rotation speed and rotation direction of the drive wheels 21. An encoder is provided for each of the two drive wheels 21. The inertial measurement unit includes, for example, a gyro sensor that detects the movement of the autonomous vehicle 10 as an angular velocity, and an acceleration sensor that detects the acceleration acting on the autonomous vehicle 10.

[0012] The autonomous vehicle 10 includes a cruise control device 51. The cruise control device 51 includes a processor 52 and a storage unit 53. The processor 52 is, for example, a central processing unit (CPU), a graphics processing unit (GPU), or a digital signal processor (DSP). The storage unit 53 includes a random access memory (RAM) and a read-only memory (ROM). The storage unit 53 stores program code or instructions configured to cause the processor 52 to execute processing. The storage unit 53, i.e., a computer-readable medium, includes any available medium accessible by a general-purpose or special-purpose computer. The cruise control device 51 may be configured with a hardware circuit such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). The cruise control device 51, which is a processing circuit, may include one or more processors operating according to a computer program, one or more hardware circuits such as ASICs or FPGAs, or a combination thereof.

[0013] Cruise control device 51 acquires the detection results of internal sensor 41. Cruise control device 51 gives commands to travel motor driver 23 and steering motor driver 33 to cause autonomous vehicle 10 to travel.

[0014] The autonomous vehicle 10 includes a light-emitting diode 61 and an LED driver 62 that controls the light-emitting diode 61. As shown in FIG. 1 , the light-emitting diode 61 emits light onto a road surface Sr. The light-emitting diode 61 is provided at the bottom of the vehicle body 11 while facing vertically. The LED driver 62, for example, passes a current through the light-emitting diode 61 to energize and cut off the current flow to the light-emitting diode 61. When energized, the light-emitting diode 61 emits light.

[0015] The autonomous vehicle 10 includes a camera 71. The camera 71 is a monocular camera. The camera 71 is a digital camera. The camera 71 includes an image sensor. The image sensor is, for example, a charge coupled device (CCD) image sensor or a complementary metal oxide semiconductor (CMOS) image sensor. The camera 71 is, for example, an RGB camera, an infrared camera, a grayscale camera, or a visible light camera.

[0016] The camera 71 captures an image and generates image data. This image data is digital data of the image captured by the camera 71. The camera 71 is positioned to capture an image of the road surface Sr. More specifically, the camera 71 is positioned to capture an image of the road surface Sr that is illuminated by the light-emitting diodes 61. The camera 71 generates image data that indicates the image of the road surface Sr. The camera 71 is mounted on the bottom of the vehicle body 11 facing vertically.

[0017] Autonomous vehicle 10 includes secondary storage device 81. Secondary storage device 81 is, for example, a hard disk drive, a solid state drive, or a flash memory. The auxiliary storage device 81 stores map data M1. The auxiliary storage device 81 is an example of a storage device that stores map data M1. The map data M1 associates map image data obtained by capturing a road surface Sr in advance with position information. The range in which autonomous vehicle 10 travels is determined in advance. The position information includes coordinates and attitude. The coordinates are coordinates in a map coordinate system, which is a coordinate system that represents absolute positions. The map coordinate system may be a Cartesian coordinate system or a geographic coordinate system. The map coordinate system has an X axis and a Y axis. The X axis and the Y axis are orthogonal to each other. The X axis and the Y axis are coordinate systems that represent the horizontal direction. The attitude is information that indicates the inclination of autonomous vehicle 10 with respect to the coordinate axes of the map coordinate system. Map data M1 is data that represents the coordinates in the map coordinate system of the range in which autonomous vehicle 10 travels and the attitude.

[0018] The autonomous vehicle 10 includes a self-position estimation control device 91. The self-position estimation control device 91 includes, for example, the same hardware configuration as the cruise control device 51. The self-position estimation control device 91 includes, for example, a processor 92 and a storage unit 93.

[0019] The self-location estimation control device 91 causes the camera 71 to capture an image. The self-location estimation control device 91 causes the light-emitting diode 61 to emit light. For example, the self-location estimation control device 91 causes the light-emitting diode 61 to emit light by issuing a command to the LED driver 62. The self-location estimation control device 91 is capable of reading information stored in the auxiliary storage device 81. The self-location estimation control device 91 is configured to be able to communicate with the driving control device 51. This allows the self-location estimation control device 91 to acquire information from the driving control device 51.

[0020] Autonomous vehicle 10 includes battery 100. Battery 100 is a power source for electrical components included in autonomous vehicle 10. The electrical components include light-emitting diodes 61 and cameras 71. <Self-position estimation control> The self-position estimation control device 91 performs self-position estimation control. The self-position estimation control is repeatedly performed at a predetermined control period. The self-position estimation control device 91 is an example of a control unit.

[0021] 3, in step S1, the self-position estimation control device 91 acquires odometry information from the driving control device 51. The odometry information is information required to estimate the self-position by odometry. The odometry information is the detection result of the internal sensor 41.

[0022] Next, in step S2, the self-location estimation control device 91 estimates its own location using odometry. The self-location includes the coordinates of the autonomous vehicle 10 in a map coordinate system and the attitude of the autonomous vehicle 10. If the internal sensor 41 is an encoder, the self-location estimation control device 91 calculates a movement vector defined by the movement speed and movement direction from the output of the encoder. For example, the self-location estimation control device 91 calculates the movement speed from the rotation speed of the encoder and calculates the movement direction from the difference in rotation speed of the two drive wheels 21. The self-location estimation control device 91 can calculate the amount of change in position from the previous self-location by integrating the movement vector from the time when the previous self-location was determined to the present. If the internal sensor 41 is an inertial measurement unit, the amount of change in position from the previous self-location can be calculated by performing a second-order integration on the output of the inertial measurement unit. The self-location estimation control device 91 estimates its own location by adding the amount of change to the previous self-location. In this way, in estimating the self-position using odometry, the current self-position is estimated based on the change in position relative to the previous self-position.

[0023] Next, in step S3, the self-position estimation control device 91 determines whether the autonomous vehicle 10 has moved a certain distance. Whether the autonomous vehicle 10 has moved a certain distance can be determined, for example, by calculating the distance traveled by the autonomous vehicle 10 from the point in time when the determination result in step S3 is positive, and determining whether the distance traveled has reached a certain distance. If the determination in step S3 is positive, the self-position is estimated by comparing the image data with map image data. Step S3 is a determination of whether the autonomous vehicle 10 has moved a certain distance since the previous estimation of its self-position by comparing the image data with map image data. The travel distance of the autonomous vehicle 10 can be calculated by accumulating the amount of change in its self-position estimated by odometry.

[0024] The fixed distance can be set arbitrarily. Odometry-based self-location estimation estimates the current self-location based on the change in relative position from the previous self-location. Therefore, errors accumulate as the moving distance increases. The fixed distance is set within an acceptable range for the accumulated errors due to the self-location estimation based on odometry.

[0025] If the determination result in step S3 is negative, the self-position estimation control device 91 returns to step S1 If the determination result in step S3 is positive, the self-position estimation control device 91 proceeds to step S4.

[0026] In step S4, the self-position estimation control device 91 causes the light-emitting diode 61 to emit light and captures an image with the camera 71. As a result, the self-position estimation control device 91 acquires image data of the captured road surface Sr.

[0027] Next, in step S5, the self-location estimation control device 91 estimates its own location by comparing the image data with map image data. First, the self-location estimation control device 91 matches the image data with the map image data. The self-location estimation control device 91 extracts feature points from the image data. The self-location estimation control device 91 describes the feature amounts of the feature points. The feature amounts are, for example, feature amount vectors or brightness values. The self-location estimation control device 91 also extracts feature points and describes the feature amounts using the map image data. The self-location estimation control device 91 compares the feature points and feature amounts obtained from the image data with the feature points and feature amounts obtained from the map image data, and searches for pairs of feature points with similar feature amounts. The self-location estimation control device 91 identifies map image data corresponding to the image data based on the feature point pairs. For example, the self-location estimation control device 91 identifies map image data in which feature point pairs are concentrated as map image data corresponding to the image data. The above-mentioned matching can be performed using feature amount descriptors. The feature descriptor is, for example, Oriented Fast and Rotated Brief (ORB), Scale-Invariant Feature Transform (SIFT), or Speeded Up Robust Features (SURF).

[0028] The self-location estimation control device 91 estimates its own location based on map image data. The self-location estimation control device 91 calculates the relative position between the map image data and the image data, and the relative angle between the map image data and the image data. The relative position between the map image data and the image data is the amount of deviation between the image data and the map image data. The relative angle between the image data and the map image data is the angle of deviation between the image data and the map image data. The image data and the map image data often do not match perfectly. This is because the position and posture of the autonomous vehicle 10 rarely match perfectly between the time the map image data is acquired and the time the image data is acquired. For this reason, the image data often only matches part of the map image data. If the position of the autonomous vehicle 10 is different between the time the map image data is acquired and the time the image data is acquired, the difference in the position of the autonomous vehicle 10 causes a deviation between the position of the road surface Sr shown in the map image data and the position of the road surface Sr shown in the image data. This amount of deviation is the relative position between the map image data and the image data. The amount of deviation can be determined from the positional relationship between the feature points of the map image data and the feature points of the image data. Similarly, the image data is a rotated version of the map image data due to the difference in the attitude of the autonomous vehicle 10 between the time the map image data was acquired and the time the image data was acquired. The angle of deviation resulting from this rotation is the relative angle between the image data and the map image data. The self-location estimation control device 91 estimates the self-location based on the position information, relative position, and relative angle associated with the map image data. The self-location estimation control device 91 shifts the coordinates associated with the map image data by the coordinates corresponding to the relative position. The self-location estimation control device 91 shifts the attitude associated with the map image data by the relative angle. The self-location estimation control device 91 regards the coordinates and attitude in the map coordinate system obtained as the self-location.

[0029] If the determination result in step S3 is negative, the self-location estimation control device 91 adopts the self-location obtained by estimating the self-location using odometry as the current self-location. If the determination result in step S3 is positive, the self-location estimation control device 91 adopts the self-location obtained by estimating the self-location by comparing the image data with the map image data as the current self-location. This corrects the self-location estimated by odometry.

[0030] Next, in step S6, the self-position estimation control device 91 records the time when the self-position estimation is performed by comparing the image data with the map image data. Next, in step S7, the self-position estimation control device 91 determines whether the power supply to the autonomous vehicle 10 is OFF. When the power supply to the autonomous vehicle 10 is OFF, the autonomous vehicle 10 is unable to travel. When the power supply to the autonomous vehicle 10 is ON, the autonomous vehicle 10 is able to travel. If the determination result in step S7 is negative, the self-position estimation control device 91 returns to step S1. If the determination result in step S7 is positive, the self-position estimation control device 91 ends the self-position estimation control.

[0031] [Operation of this embodiment] The self-position estimation control device 91 corrects the self-position by comparing the image data with map image data each time the autonomous vehicle 10 moves a certain distance. After correcting the self-position by comparing the image data with map image data, the self-position estimation control device 91 estimates the self-position by odometry until the autonomous vehicle 10 has traveled a certain distance. Errors in the self-position estimation by odometry accumulate as the travel distance increases. For this reason, the self-position is corrected by comparing the image data with map image data each time the autonomous vehicle 10 moves a certain distance. This makes it possible to reset errors that occur when the autonomous vehicle 10 estimates its self-position by odometry.

[0032] When the speed of the autonomous vehicle 10 is low, it takes a long time for the autonomous vehicle 10 to travel a certain distance, so the frequency with which the light-emitting diode 61 emits light and the camera 71 captures images decreases. Therefore, power consumption can be reduced compared to when image data is compared with map image data at regular intervals. In particular, when comparing image data with map image data at regular intervals, the regular interval needs to be set based on the maximum error in the autonomous vehicle's position that can occur within the regular interval. The error in the autonomous vehicle's position increases as the distance traveled by the autonomous vehicle 10 increases. Therefore, the regular interval is set based on the error in the autonomous vehicle's position that occurs when the autonomous vehicle 10 is traveling at its maximum speed. In this case, errors accumulate over a short period of time, so the regular interval needs to be set short. Therefore, when comparing image data with map image data at regular intervals, the frequency with which the light-emitting diode 61 emits light and the camera 71 captures images increases compared to when image data is compared with map image data every time the autonomous vehicle 10 travels a certain distance.

[0033] When autonomous vehicle 10 compares image data with map image data every time it travels a certain distance, the slower the speed of autonomous vehicle 10, the less frequently its own position is corrected. Since the error in the own position due to odometry increases as the distance traveled by autonomous vehicle 10 increases, the error is less likely to increase when the speed of autonomous vehicle 10 is low. For this reason, when the speed of autonomous vehicle 10 is low, reducing the frequency at which image data is compared with map image data is less likely to cause practical problems.

[0034] When the speed of autonomous vehicle 10 is high, the time it takes for autonomous vehicle 10 to travel a certain distance becomes shorter, so the frequency with which light-emitting diode 61 emits light and camera 71 captures images increases. The higher the speed of autonomous vehicle 10, the longer the distance traveled by autonomous vehicle 10 in a short period of time. Therefore, the higher the speed of autonomous vehicle 10, the more errors accumulate in a short period of time. By increasing the frequency with which image data and map image data are compared as the speed of autonomous vehicle 10 increases, errors in the self-position are less likely to become large even when the speed of autonomous vehicle 10 is high.

[0035] [Effects of this embodiment] (1) The self-position estimation control device 91 corrects the self-position by comparing the image data with the map image data each time the autonomous vehicle 10 moves a certain distance. When the speed of the autonomous vehicle 10 is low, it takes a long time for the autonomous vehicle 10 to move a certain distance, so the frequency with which the light-emitting diode 61 emits light and the camera 71 captures images decreases. Therefore, power consumption can be reduced compared to when the image data is compared with the map image data at regular intervals.

[0036] [Example of change] The embodiment can be modified as follows: The embodiment and the following modifications can be combined with each other to the extent that they are not technically inconsistent.

[0037] The self-position estimation control device 91 may directly acquire the detection result of the internal sensor 41 from the internal sensor 41 . The self-position estimation using odometry may be performed by the driving control device 51. In this case, the self-position estimation control device 91 and the driving control device 51 are the control unit.

[0038] The storage device that stores the map data M1 may be the storage unit 93. The control unit may be a single control device that has the functions of the driving control device 51 and the self-position estimation control device 91.

[0039] Whether autonomous vehicle 10 has moved a certain distance may be determined by measuring the amount of change in the position of autonomous vehicle 10 using a Global Navigation Satellite System (GNSS). [Explanation of symbols]

[0040] M1...map data, Sr...road surface, 10...autonomous vehicle, 41...internal sensor, 61...light-emitting diode, 71...camera, 81...auxiliary memory device which is an example of a memory device, 91...self-position estimation control device which is an example of a control unit.

Claims

[Claim 1] a camera positioned to capture an image of the road surface; a light-emitting diode that irradiates light onto the road surface; a storage device that stores map data in which map image data of the road surface captured in advance is linked to location information; An internal sensor; A control unit and an autonomous vehicle comprising: The control unit Estimating a self-position by odometry based on the detection results of the internal sensor; causing the light-emitting diode to emit light and the camera to capture an image each time the autonomous vehicle moves a certain distance; The autonomous vehicle corrects its own position by comparing the image data acquired from the camera with the map image data.

Images