WO2017154456A1

WO2017154456A1 - Self-driving control device and self-driving system

Info

Publication number: WO2017154456A1
Application number: PCT/JP2017/004855
Authority: WO
Inventors: 将広内藤; 袴田　浩司; 毅小形; 谷　則幸
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2016-03-09
Filing date: 2017-02-10
Publication date: 2017-09-14
Also published as: JP2017159801A

Abstract

The present invention assists obstacle avoidance in self-driving of a vehicle. An obstacle detection unit of an obstacle detection device detects an obstacle on the basis of output data from a TOF camera. A reliability estimation unit of the obstacle detection device estimates the reliability of an obstacle detection result from the obstacle detection unit on the basis of the intensity of background light. An obstacle information acquisition unit of the automatic drive control device acquires the obstacle detection result and the reliability thereof from the obstacle detection device. In accordance with the obstacle detection result, a behavior determination unit of the automatic drive control device determines a vehicle behavior so as to avoid collision with the obstacle, and determines a different behavior in accordance with the reliability of the obstacle detection result. A drive control unit of the self-driving control device causes the vehicle to execute a behavior determined by the behavior determination unit.

Description

Automatic operation control device and automatic operation system

The present disclosure relates to data processing technology, and more particularly to an automatic operation control device and an automatic operation system.

Using a device (such as a TOF (Time 時間 of まで flight) camera) that measures the distance to the object based on the time it takes for the pulsed light to be reflected and returned to the object, it is present on the travel route of the vehicle. Techniques for detecting obstacles have been proposed.

JP 2010-256179 A

When the TOF camera is used outdoors, the light of the same wavelength included in the outdoor light becomes noise, and the accuracy of distance measurement by the TOF camera decreases. Therefore, for example, in an automatic driving system using a TOF camera, even if an obstacle is detected, the prediction of the distance from the vehicle to the obstacle may be inaccurate. On the other hand, in the automatic driving of the vehicle, since it is necessary to control the vehicle so as to avoid the obstacle, it is preferable that the prediction of the distance from the vehicle to the obstacle is accurate.

The present disclosure has been made in view of such a situation, and a main object is to provide a technology that supports obstacle avoidance in automatic driving of a vehicle.

In order to solve the above problem, an automatic operation control device according to an aspect of the present disclosure includes an acquisition unit that acquires an obstacle detection result, and a collision with an obstacle according to the detection result acquired by the acquisition unit. A determination unit that determines the behavior of the vehicle so as to avoid, and an operation control unit that causes the vehicle to execute the behavior determined by the determination unit. The acquisition unit further acquires information indicating the reliability of the obstacle detection result, and the determination unit determines different behaviors according to the information indicating the reliability.

Another aspect of the present disclosure is an automatic driving system. The automatic driving system includes an obstacle detection device and an automatic driving control device. The obstacle detection device includes a light emitting unit that emits light toward the outside of the vehicle, a detection unit that detects an obstacle according to the time until the light emitted from the light emitting unit is reflected back to the object, and the light emission. An estimation unit for estimating the reliability of the detection result by the detection unit according to the intensity of the background light received when the unit is not emitting light, information on the obstacle detected by the detection unit, and the reliability estimated by the estimation unit And a transmitter for transmitting the information to the automatic operation control device. The automatic driving control device determines the behavior of the vehicle according to the obstacle information transmitted from the obstacle detection device so as to avoid a collision with the obstacle, and the behavior determined by the decision unit. An operation control unit to be executed by the vehicle. The determination unit determines different behaviors according to the reliability information transmitted from the obstacle detection device.

Note that any combination of the above components, the expression of the present disclosure converted between a method, a computer program, a recording medium on which the computer program is recorded, a vehicle equipped with the apparatus, and the like are also included in the aspects of the present disclosure It is effective as

According to the present disclosure, obstacle avoidance in automatic driving of a vehicle can be supported.

It is a figure which shows the principle of a TOF camera. It is a figure which shows the structure of the vehicle which concerns on 1st Example. It is a figure which shows the light emission pulse and light reception pulse in a TOF camera. It is a figure which shows the example of the obstacle avoidance behavior of a vehicle. It is a flowchart which shows operation | movement of the automatic driving system which concerns on 1st Example. It is a flowchart which shows operation | movement of the automatic driving | operation system which concerns on 2nd Example. It is a figure which shows the structure of the obstruction detection apparatus which concerns on 3rd Example. It is a flowchart which shows operation | movement of the automatic driving | operation system which concerns on 3rd Example.

DETAILED DESCRIPTION An outline of an embodiment will be described before specifically describing an embodiment of the present disclosure.

Fig. 1 shows the principle of the TOF camera. The TOF camera 10 is also referred to as a TOF type distance image sensor, and the light receiving element 14 receives the reflected light returned from the measurement light emitted from the light emitting element 12 after being reflected by the object 16. The TOF camera 10 measures the distance Z to the object 16 based on the time from the emission of the measurement light to the reception of the reflected light. The distance Z is obtained by (light velocity c × delay time Δt / 2 from the measurement light emission to the reflection light reception).

However, when the TOF camera 10 is used outdoors, the light of the same wavelength contained in the light from the noise light source 18 such as the sun becomes noise, and the measurement accuracy of the distance Z by the TOF camera 10 decreases. Therefore, for example, in an automatic driving system using a TOF camera, there may be a problem that an obstacle that does not exist is erroneously detected, or a problem that a distance to the obstacle is erroneously detected. As a result, there may be a problem that the operation for avoiding the obstacle cannot be executed at an appropriate timing. Further, when a separate sensor for measuring noise light is provided, there arises a problem that the number of parts and the cost increase.

Therefore, the obstacle detection apparatus of this embodiment detects an obstacle using a TOF camera and estimates the reliability of the detection result. The automatic driving control device of the present embodiment determines the behavior of the vehicle based on the obstacle detection result by the obstacle detection device so as to avoid collision with the obstacle, and causes the vehicle to execute the determined behavior. Moreover, even if the detection result of an obstruction is the same, the automatic driving | operation control apparatus of a present Example determines a different behavior according to the reliability of the detection result. That is, the automatic driving control device of the present embodiment changes the traveling control for avoiding the obstacle according to the reliability of the output value of the sensor. Note that the behavior of the vehicle for avoiding a collision with an obstacle includes stopping before the obstacle, and continuing to proceed while avoiding the obstacle, in other words, overtaking the obstacle.

(First embodiment)
FIG. 2 shows a configuration of the vehicle 20 according to the first embodiment. The vehicle 20 includes a driving operation unit 22, an obstacle detection device 24, and an automatic driving control device 26. In FIG. 2, elements constituting the automatic driving system 28 among various elements constituting the vehicle 20 are shown.

The driving operation unit 22 is an element that operates the behavior of the vehicle 20 and includes a steering wheel 30, a brake pedal 32, an accelerator pedal 34, and a winker switch 36. The steering 30, brake pedal 32, accelerator pedal 34, and blinker switch 36 can be electronically controlled by a steering ECU (ECU: Electronic Control Unit), brake ECU, engine ECU, and / or motor ECU. It is. In the automatic operation mode, the steering ECU, the brake ECU, the engine ECU, and the motor ECU drive the actuator in accordance with a control signal supplied from the automatic operation control device 26. The blinker controller turns on or off the blinker lamp according to a control signal supplied from the automatic operation control device 26.

The obstacle detection device 24 includes a TOF camera 40, an obstacle detection unit 48, a reliability estimation unit 50, and an obstacle information transmission unit 52. The TOF camera 40 includes a light emitting unit 42 corresponding to the light emitting element 12 in FIG. 1, a light receiving unit 44 corresponding to the light receiving element 14 in FIG. 1, and a distance measuring unit 46. The light emitting unit 42 emits predetermined measurement light (for example, infrared light) toward the outside of the vehicle 20 (for example, the front space of the vehicle 20).

The light receiving unit 44 receives light (hereinafter referred to as “reflected light”) when the light emitted from the light emitting unit 42 is reflected back to an object outside the vehicle 20 and returned. The light receiving unit 44 also receives light (hereinafter referred to as “background light”) input when the light emitting unit 42 does not emit light. The background light is light that does not include the reflected light accompanying the light emission by the light emitting unit 42, and is light that is received from the surrounding environment of the vehicle 20 regardless of whether the light emitting unit 42 emits light.

The distance measuring unit 46 measures the distance from the vehicle 20 to the object outside the vehicle 20 according to the time until the light emitted from the light emitting unit 42 is reflected by the object and returns. The distance measurement unit 46 generates image data (hereinafter referred to as “TOF image”) in which the distance from the vehicle 20 to the object is associated with each of the plurality of pixels, and outputs the image data to the obstacle detection unit 48.

FIG. 3 shows light emission pulses and light reception pulses in the TOF camera 40. The time from the rise of the light emission pulse to the rise of the light reception pulse corresponds to the delay time Δt from the emission of the measurement light to the reception of the reflected light. A0 in the figure is the exposure amount from the rising edge of the received light pulse to the falling edge of the light emission pulse. A1 in the figure is the exposure amount from the fall of the light emission pulse to the fall of the light reception pulse. The time from the fall of the light emission pulse to the fall of the light reception pulse also corresponds to the delay time Δt. BG in the figure is the exposure amount of the background light.

The distance measuring unit 46 may determine the delay time Δt based on the exposure amount of the received light pulse. Specifically, it may be calculated based on the area ratio of A0, A1, and BG shown in FIG. Alternatively, the distance measuring unit 46 may directly obtain the delay time Δt by time measurement, such as measuring the time from the rise of the light emission pulse to the rise of the light reception pulse. The distance measuring unit 46 may obtain the delay time Δt for each pixel of the TOF image, and associate the distance Z to the object calculated based on the light speed c and the delay time Δt to each pixel of the TOF image.

The reliability estimation unit 50 described later identifies the amplitude (Ha) of the light emission pulse shown in FIG. 3 as the light emission intensity, and uses the background light offset value Hb (for example, the distance from the reference value indicating the amount of received light 0) as the background. You may identify as light intensity. Or the reliability estimation part 50 mentioned later may identify the light emission amount (Aa) shown in FIG. 3 as light emission intensity, and may identify the exposure amount (BG) of background light as background light intensity. Furthermore, the reliability estimation unit 50 described later may identify the average value of Hb or BG measured for each pixel of the TOF image as the background light intensity.

2, the obstacle detection unit 48 detects an obstacle present on the travel route of the vehicle 20 in accordance with information indicating the distance to the object measured by the distance measurement unit 46. For example, the obstacle detection unit 48 extracts one or more object regions indicating one or more objects from the TOF image generated by the distance measurement unit 46, and excludes objects that match predetermined road surface data. The remaining objects may be identified as obstacles. The obstacle detection unit 48 may detect the position of the obstacle in the front space of the vehicle 20 (in other words, the distance from the vehicle 20) and the time until the vehicle 20 collides with the obstacle.

As described above, in the obstacle detection device 24 of the present embodiment, the TOF camera 40 and the obstacle detection unit 48 cooperate to detect the obstacle according to the time until the emitted light is reflected back to the object and returned. To do. The measurement of the distance from the vehicle 20 to the object outside the vehicle 20 and the detection of the obstacle based on the measurement result may be realized by a known method.

The reliability estimation unit 50 estimates the reliability of the obstacle detection result by the obstacle detection unit 48 according to the intensity of background light received by the light receiving unit 44 when the light emitting unit 42 is not emitting light. This reliability can be said to be an index value indicating the level of accuracy of the obstacle detection result by the obstacle detection unit 48.

The reliability estimation unit 50 according to the present embodiment acquires the light emission intensity by the light emission unit 42 and the background light intensity from the TOF camera 40, and compares the two to estimate the reliability. Specifically, the reliability is obtained as a signal-to-noise ratio (S / N ratio), and the ratio between the light emission intensity and the background light intensity is estimated as the reliability. For example, the reliability estimation unit 50 may identify a value obtained by dividing the light emission intensity by the background light intensity as the reliability value.

The obstacle information transmission unit 52 associates the obstacle information related to the obstacle and the reliability information indicating the reliability of the obstacle information and transmits the information to the automatic operation control device 26. The obstacle information includes information indicating the position of the obstacle detected by the obstacle detection unit 48 and the time until the obstacle is reached. The reliability information includes a reliability value estimated by the reliability estimation unit 50.

The automatic driving control device 26 is an automatic driving controller in which an automatic driving control function is implemented, and controls the behavior (also referred to as action or operation) of the vehicle 20 in automatic driving. The automatic operation control device 26 includes an I / O unit 60, a storage unit 62, and a control unit 64. The I / O unit 60 executes communication control according to various communication formats.

The storage unit 62 includes a nonvolatile recording medium such as a flash memory. The storage unit 62 holds a predetermined automatic operation algorithm. The automatic driving algorithm includes an obstacle avoidance algorithm that determines a driving mode (route, speed, etc.) of the vehicle so as to avoid a collision with the obstacle according to the position of the detected obstacle. The obstacle avoidance algorithm includes information indicating a safety distance that is a distance to be maintained between the vehicle and the obstacle for safely avoiding a collision with the obstacle. The storage unit 62 also holds various threshold data that are conditions for determining the mode of automatic driving.

The control unit 64 can be realized by cooperation of hardware resources and software resources, or only by hardware resources. Processors, ROM, RAM, and other LSIs can be used as hardware resources, and programs such as an operating system, application, and firmware can be used as software resources. The control unit 64 includes an obstacle information acquisition unit 66, a behavior determination unit 68, and an operation control unit 70.

The obstacle information acquisition unit 66 acquires the obstacle information transmitted from the obstacle detection device 24 via the I / O unit 60. As described above, the obstacle information includes the obstacle detection result (the position of the obstacle, etc.) by the obstacle detection device 24 and the reliability information indicating the reliability of the detection result.

The behavior determination unit 68 determines the behavior of the vehicle so as to avoid collision with the obstacle according to the information of the obstacle detection result acquired by the obstacle information acquisition unit 66. The behavior of the vehicle for avoiding a collision with an obstacle is hereinafter referred to as “obstacle avoidance behavior”. The behavior determination unit 68 determines the obstacle avoidance behavior according to the obstacle avoidance algorithm stored in the storage unit 62.

For example, the behavior determination unit 68 may determine overtaking of the obstacle as the obstacle avoidance behavior. In this case, the behavior determining unit 68 determines a time and a vehicle speed until the lane change to the right lane is started, and further, a time until the lane change to the left lane is started after the lane change to the right lane is completed. And the vehicle speed may be determined. Moreover, the behavior determination unit 68 may determine to stop before the obstacle as the obstacle avoidance behavior. In this case, the behavior determination unit 68 may determine the waiting time until the actuator of the brake pedal 32 is operated, the operation time of the actuator of the brake pedal 32 and the strength thereof.

Also, the behavior determination unit 68 determines different obstacle avoidance behaviors according to the reliability information acquired by the obstacle information acquisition unit 66. That is, the behavior determination unit 68 determines the obstacle avoidance behavior according to the obstacle detection result indicated by the obstacle information transmitted from the obstacle detection device 24, but differs depending on the reliability indicated by the obstacle information. Determine obstacle avoidance behavior. In other words, even if the obstacle detection result is the same, the behavior determination unit 68 determines a different obstacle avoidance behavior if the reliability of the obstacle detection result is different.

For example, the storage unit 62 may store one or more predetermined threshold values (hereinafter also referred to as “reliability threshold values”) for determining the level of reliability indicated by the reliability information. The behavior determining unit 68 determines that the reliability of the obstacle detection result is high when the reliability value indicated by the reliability information is equal to or higher than the reliability threshold, and determines the obstacle avoidance behavior corresponding to the high reliability. Also good. On the other hand, the behavior determining unit 68 determines that the reliability of the obstacle detection result is low when the reliability value indicated by the reliability information is less than the reliability threshold value, and determines the obstacle avoidance behavior corresponding to the low reliability. You may decide. Note that the reliability threshold value may be determined by a developer's knowledge, an experiment using the automatic driving system 28, or the like.

Here, an example of determining different obstacle avoidance behavior depending on the reliability is shown. For example, the obstacle avoidance algorithm stored in the storage unit 62 includes a first safe distance when the reliability is relatively high and a safe distance when the obstacle is relatively low, which is greater than the first safe distance. A large second safety distance may be defined. When the reliability indicated by the reliability information is relatively high, the behavior determination unit 68 may determine the obstacle avoidance behavior so that the distance between the host vehicle and the obstacle maintains the first safety distance. On the other hand, when the reliability indicated by the reliability information is relatively low, the behavior determination unit 68 determines the obstacle avoidance behavior so that the distance between the host vehicle and the obstacle maintains the second safety distance. Good.

As another example, when the reliability indicated by the reliability information is relatively low, the behavior determination unit 68 starts the behavior to avoid collision with an obstacle earlier than when the reliability is relatively high. It may be determined as follows. To make the start of the behavior early may be that the start position of the obstacle avoidance behavior is brought forward, or the start time of the obstacle avoidance behavior may be advanced. As yet another example, the behavior determining unit 68 is configured such that when the reliability indicated by the reliability information is relatively low, the distance from the obstacle when avoiding a collision with the obstacle is higher than when the reliability is relatively high. The behavior may be determined so as to increase. The distance to the obstacle here may be a distance that is secured at least between the vehicle and the obstacle.

FIG. 4 shows an example of the obstacle avoidance behavior of the vehicle. FIG. 4A shows an obstacle avoidance behavior when the reliability of the obstacle detection result is relatively high. In FIG. 4A, the behavior determining unit 68 changes the lane to the right when (1) the distance between the vehicle 20 and the obstacle 80 reaches a predetermined behavior start distance 82, and (2) the avoidance distance 84. (3) The obstacle avoidance behavior is determined so as to return to the original lane in front of the obstacle 80.

FIG. 4B shows the obstacle avoidance behavior when the reliability of the obstacle detection result is relatively low. Also in (b) of FIG. 4, the behavior determining unit 68 determines the behavior of the vehicle 20 so as to pass the obstacle 80. However, the behavior determination unit 68 sets the behavior start distance 82 for starting the obstacle avoidance behavior to be larger than that when the reliability is relatively high. That is, the obstacle avoidance behavior is started at an earlier timing than when the reliability is high. If the behavior determination unit 68 determines that it is difficult to change the lane to the right lane as a result of increasing the behavior start distance 82, the behavior determination unit 68 determines the obstacle avoidance behavior so as to stop before the obstacle 80. Furthermore, the behavior determination unit 68 increases the avoidance distance 84 that is secured when passing the side of the obstacle, compared to the case where the reliability is relatively high.

As described above, the behavior determining unit 68 is configured to increase the distance between the vehicle 20 and the obstacle 80 when the reliability of the obstacle detection result by the obstacle detection device 24 is low than when the reliability is high. Determine avoidance behavior. Thereby, even if the reliability of the obstacle detection result by the obstacle detection device 24 is low, the certainty of collision avoidance between the vehicle 20 and the obstacle 80 can be improved. In addition, when the reliability of the obstacle detection result is low, the behavior determination unit 68 may make either one of the behavior start distance 82 and the avoidance distance 84 longer than when the reliability is high.

2, the operation control unit 70 is the behavior determined by the behavior determination unit 68, and in the present embodiment, causes the vehicle 20 to execute an obstacle avoidance behavior in particular. Specifically, the driving control unit 70 calculates a control value for the control target (steering 30 or the like) of the driving operation unit 22 for causing the vehicle 20 to execute the obstacle avoidance behavior. The operation control unit 70 transmits a control command indicating the calculated control value to each control target ECU or controller via the I / O unit 60.

The operation of the automatic driving system 28 having the above configuration will be described.

FIG. 5 is a flowchart showing the operation of the automatic driving system 28 according to the first embodiment. The TOF camera 40 of the obstacle detection device 24 measures the distance from the vehicle to an object existing in the traveling direction of the vehicle (eg, forward) (S10). The obstacle detection unit 48 of the obstacle detection device 24 detects an obstacle to the traveling of the vehicle based on the distance information to the object measured by the TOF camera 40 (S12). When an obstacle is detected (Y in S14), the reliability estimation unit 50 of the obstacle detection device 24 determines the obstacle detection result according to the ratio between the emission intensity of the TOF camera 40 and the current background light intensity. Is estimated (S16). The obstacle information transmission unit 52 of the obstacle detection device 24 transmits the obstacle detection result and its reliability information to the automatic operation control device 26 (S18). The obstacle detection device 24 periodically repeats the processes of S10 to S18.

The behavior determination unit 68 of the automatic operation control device 26 determines the obstacle avoidance behavior according to the obstacle detection result by the obstacle detection device 24 (S20). When the reliability of the obstacle detection result is equal to or greater than the predetermined threshold (Y in S22), the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior determined in S20 (S24). ). When the reliability of the obstacle detection result is less than the predetermined threshold (N in S22), the behavior determination unit 68 of the automatic driving control device 26 changes the obstacle avoidance behavior determined in S20 to a safer mode (S26). ). In this case, the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior resulting from the change in S26 (S24). If no obstacle is detected by the obstacle detection device 24 (N in S14), the subsequent processing is skipped and the flow of FIG.

In the change of the obstacle avoidance behavior in S26, the behavior determination unit 68 increases the distance between the vehicle 20 and the obstacle when avoiding the obstacle, for example, when the reliability is equal to or higher than a predetermined threshold. Obstacle avoidance behavior may be changed. In addition, when the obstacle avoidance behavior that avoids the obstacle and continues to proceed is determined in S20, the behavior determining unit 68 sets the behavior so as to stop before the obstacle as a safer obstacle avoidance behavior. It may be changed. In the example of FIG. 5, the obstacle avoidance behavior is changed in S26, but the behavior determination unit 68 uses a parameter associated with the reliability value among a plurality of predetermined parameters in S20. Different obstacle avoidance behavior may be determined according to the reliability value.

According to the automatic operation control device 26 of the first embodiment, the obstacle avoidance behavior for avoiding the collision with the obstacle detected by the obstacle detection device 24 is adapted to the reliability estimated by the obstacle detection device 24. The vehicle 20 can be executed in such a manner. Further, according to the obstacle detection device 24, the reliability of the obstacle detection result is estimated by comparing the light emission intensity detectable by the TOF camera 40 and the background light intensity. This eliminates the need for a new sensor or the like for obtaining reliability, and can suppress the number of parts and cost.

(Second embodiment)
The functional blocks of the automatic driving system 28 of the second embodiment are the same as those of the first embodiment (FIG. 2). However, the obstacle detection device 24 of the second embodiment has an obstacle according to the reliability of the obstacle detection result. The detection process and the reliability estimation process are repeated a plurality of times. Hereinafter, the configuration of the obstacle detection device 24 will be described.

The reliability estimation unit 50 re-executes the distance measurement processing on the TOF camera 40 when the reliability of the first obstacle detection result by the obstacle detection unit 48 is less than a predetermined reliability threshold. Instruct. In response to this instruction, the light emitting unit 42 increases the intensity of the light emission pulse, that is, newly emits measurement light having a higher intensity than before. The light receiving unit 44 receives reflected light obtained by reflecting the newly emitted measurement light on an object outside the vehicle 20. The distance measuring unit 46 newly measures the distance to the object outside the vehicle 20 according to the time from the emission of the measurement light whose intensity has been increased until the reflected light is received.

The obstacle detection unit 48 detects the presence of an obstacle according to the distance to the object outside the vehicle 20 measured by the distance measurement unit 46 as in the first embodiment. The reliability estimation unit 50 estimates the reliability of the new obstacle detection result by the obstacle detection unit 48 by comparing the intensity of the background light and the intensity of the newly emitted measurement light. Specifically, as in the first embodiment, the ratio (S / N ratio) between the intensity of the background light and the intensity of the newly emitted measurement light is calculated as the reliability.

FIG. 6 is a flowchart showing the operation of the automatic driving system 28 according to the second embodiment. The processes of S30 to S36 and S44 to S52 of FIG. 6 correspond to S10 to S26 of FIG. 5 described in the first embodiment.

The TOF camera 40 of the obstacle detection device 24 measures the distance from the vehicle to the object existing in the traveling direction of the vehicle (S30). The obstacle detection unit 48 of the obstacle detection device 24 detects an obstacle to the traveling of the vehicle based on the distance information to the object measured by the TOF camera 40 (S32). When an obstacle is detected (Y in S34), the reliability estimation unit 50 of the obstacle detection device 24 determines the obstacle detection result according to the ratio between the emission intensity of the TOF camera 40 and the current background light intensity. Is estimated (S36).

If S30 to S36 are the first execution (N in S38) and the reliability of the obstacle detection result is equal to or higher than a predetermined threshold (Y in S40), the obstacle information transmission unit 52 of the obstacle detection device 24 will be described. Transmits the obstacle detection result and its reliability information to the automatic operation control device 26 (S44). On the other hand, if the reliability of the obstacle detection result is less than the predetermined threshold (N in S40), the TOF camera 40 of the obstacle detection device 24 newly emits measurement light with increased intensity (S42). The processes of S30 to S36 are repeated.

When S30 to S36 are the second execution (Y of S38), the obstacle information transmission unit 52 of the obstacle detection device 24 performs the second obstacle regardless of the reliability of the second obstacle detection result. The detection result and its reliability information are transmitted to the automatic operation control device 26 (S44). As a modification, the obstacle detection result with high reliability and the reliability information among the first obstacle detection result and the second obstacle detection result may be transmitted to the automatic operation control device 26. The obstacle detection device 24 periodically repeats the processes of S30 to S44.

The behavior determination unit 68 of the automatic operation control device 26 determines the obstacle avoidance behavior according to the obstacle detection result by the obstacle detection device 24 (S46). When the reliability of the obstacle detection result is equal to or higher than the predetermined threshold (Y in S48), the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior determined in S46 (S50). ). When the reliability of the obstacle detection result is less than the predetermined threshold (N in S48), the behavior determination unit 68 of the automatic driving control device 26 changes the obstacle avoidance behavior determined in S46 to a safer mode (S52). ). In this case, the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior resulting from the change in S52 (S50). If an obstacle is not detected by the obstacle detection device 24 (N in S34), the subsequent processing is skipped and the flow of FIG.

According to the automatic driving system 28 of the second embodiment, it is easy to obtain an obstacle detection result with high reliability, and it is easy to cause the vehicle to execute an obstacle avoidance behavior based on the obstacle detection result with high reliability. As a result, it is possible to more reliably prevent the avoidance operation from being executed at an unnecessary timing, or the obstacle avoidance in time and the collision with the obstacle.

(Third embodiment)
Although the configuration of the driving operation unit 22 and the automatic driving control device 26 of the third embodiment is the same as that of the first embodiment (FIG. 2), the obstacle detection device 24 of the third embodiment is the obstacle detection of the first embodiment. The reliability of the obstacle detection result is estimated by a method different from that of the device 24. Hereinafter, the configuration of the obstacle detection device 24 will be described.

FIG. 7 shows the configuration of the obstacle detection device 24 according to the third embodiment. The obstacle detection device 24 further includes a background light intensity storage unit 54. The background light intensity storage unit 54 stores information indicating the intensity of background light (hereinafter referred to as “BG0”) measured in advance in an environment (in this embodiment, indoors) in which a noise light source such as the sun is excluded or reduced. To do. The reliability estimation unit 50 compares the background light intensity (BG0) stored in the background light intensity storage unit 54 with the current background light intensity (hereinafter referred to as “BGx”), and compares it with an obstacle detection unit. The reliability of the obstacle detection result by 48 is estimated.

The current background light intensity (BGx) is the intensity of the background light received by the light receiving unit 44 when the light emitting unit 42 does not emit light in an environment where the vehicle 20 is traveling. In other words, it is the intensity of the background light at the time of measuring the distance from the object by the TOF camera 40, and in other words, the intensity of the background light at the time of the obstacle detection processing by the obstacle detection unit 48.

The reliability estimation unit 50 may estimate the reliability according to the difference or ratio between BG0 and BGx. The reliability estimation unit 50 according to the present embodiment determines that the reliability is low when the difference between BG0 and BGx is equal to or greater than a predetermined threshold A, and generates reliability information indicating the low reliability. On the other hand, if the difference between BG0 and BGx is less than the threshold value A, the reliability estimation unit 50 determines that the reliability is high and generates reliability information indicating high reliability. The threshold A is set to a value that is assumed that the reliability of the obstacle detection result is low when the difference between BG0 and BGx exceeds this value. The threshold A may be determined based on the developer's knowledge, an experiment by the automatic driving system 28, or the like.

FIG. 8 is a flowchart showing the operation of the automatic driving system 28 according to the third embodiment. S60 to S64, S68, and S72 to S82 in FIG. 8 correspond to S30 to S34, S38, and S42 to S52 in FIG. 6 described in the second embodiment.

The TOF camera 40 of the obstacle detection device 24 measures the distance from the vehicle to the object existing in the traveling direction of the vehicle (S60). The obstacle detection unit 48 of the obstacle detection device 24 detects an obstacle to the traveling of the vehicle based on the distance information to the object measured by the TOF camera 40 (S62). When an obstacle is detected (Y in S64), the reliability estimation unit 50 of the obstacle detection device 24 acquires the current background light intensity BGx from the TOF camera 40 (S66).

If S60 to S66 are the first execution (N in S68) and the difference between BG0 stored in advance and BGx is less than a predetermined threshold A (N in S70), the obstacle detection device 24 has a fault. The object information transmission part 52 transmits an obstacle detection result and its reliability information to the automatic driving control device 26 (S74). On the other hand, if the difference between BG0 and BGx is greater than or equal to the threshold value A (Y in S70), the TOF camera 40 of the obstacle detection device 24 newly emits measurement light with increased intensity (S72), and S60 to The process of S66 is repeated.

When S60 to S66 are the second execution (Y of S68), the obstacle information transmission unit 52 of the obstacle detection device 24 determines the second obstacle detection result and its result regardless of the difference between BG0 and BGx. The reliability information is transmitted to the automatic operation control device 26 (S74). As a modification, even if the obstacle detection result with the small difference between BG0 and BGx out of the first obstacle detection result and the second obstacle detection result and its reliability information are transmitted to the automatic operation control device 26, Good. The obstacle detection device 24 periodically repeats the processing of S60 to S74.

The behavior determination unit 68 of the automatic operation control device 26 determines the obstacle avoidance behavior according to the obstacle detection result by the obstacle detection device 24 (S76). If the reliability of the obstacle detection result is equal to or higher than the predetermined threshold (Y in S78), the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior determined in S76 (S80). ). When the reliability of the obstacle detection result is less than the predetermined threshold (N in S78), the behavior determination unit 68 of the automatic driving control device 26 changes the obstacle avoidance behavior determined in S76 to a safer mode (S82). ). In this case, the driving control unit 70 of the automatic driving control device 26 causes the vehicle 20 to execute the obstacle avoidance behavior resulting from the change in S82 (S80). If no obstacle is detected by the obstacle detection device 24 (N in S64), the subsequent processing is skipped and the flow of this figure is terminated.

According to the automatic driving system 28 of the third embodiment, the reliability of the obstacle detection result is estimated by comparing the current background light intensity detectable by the TOF camera 40 and the indoor background light intensity measured in advance. . This eliminates the need for a new sensor or the like for obtaining reliability, and can suppress the number of parts and cost.

The present invention has been described based on the first to third embodiments. Those skilled in the art will understand that these embodiments are exemplifications, and that various modifications can be made to each component or combination of each treatment process, and such modifications are within the scope of the present invention. is there.

Modifications related to the second embodiment and the third embodiment will be described. In the second example and the third example, the obstacle detection process and the reliability estimation process were repeated at most twice. As a modification, when the reliability of the obstacle detection result does not reach a predetermined reliability threshold, the obstacle detection process and the reliability estimation process may be repeated three or more times. And every time it repeats, you may increase the emitted light intensity by the TOF camera 40 (light emission part 42) in steps. This makes it easier to obtain highly reliable obstacle detection results even outdoors where there is a lot of noise.

Modifications related to the second embodiment and the third embodiment will be described. In the second embodiment and the third embodiment, when the reliability of the obstacle detection result based on the output data of the TOF camera 40 does not reach the predetermined reliability threshold, the light emission intensity by the TOF camera 40 is increased. As a modification, the automatic driving system 28 may include another sensor having a method different from that of the TOF camera 40. Another sensor may be, for example, any one or combination of a CCD image sensor, a CMOS image sensor, a sonar, and a radar. When the reliability of the obstacle detection result based on the output data of the TOF camera 40 does not reach the predetermined reliability threshold, the obstacle detection unit 48 detects the presence of the obstacle based on the output result of another sensor. Also good. The obstacle information transmitting unit 52 may transmit an obstacle detection result based on the output result of another sensor to the automatic driving control device 26 so that the automatic driving control device 26 determines the obstacle avoidance behavior.

Modifications related to the first to third embodiments will be described. The behavior determination unit 68 of the automatic operation control device 26 may evaluate the reliability estimated by the obstacle detection device 24 in three or more stages such as high, medium, and low. The behavior determination unit 68 may determine the travel route, speed, etc. of the vehicle so that the lower the reliability estimated by the obstacle detection device 24, the faster the start of the behavior for avoiding collision with the obstacle. . In addition, the behavior determination unit 68 determines the travel route, speed, and the like of the vehicle so that the lower the reliability estimated by the obstacle detection device 24, the greater the distance from the obstacle when avoiding a collision with the obstacle. May be.

Any combination of the above-described examples and modifications is also useful as an embodiment of the present invention. The new embodiment resulting from the combination has the effects of the combined example and modification. Further, it should be understood by those skilled in the art that the functions to be fulfilled by the constituent elements described in the claims are realized by the individual constituent elements shown in the embodiments and the modified examples or by their cooperation.

It should be noted that the techniques described in the embodiments and the modifications may be specified by the following items.

[Item 1]
An acquisition unit that acquires the detection result of the obstacle, a determination unit that determines the behavior of the vehicle so as to avoid a collision with the obstacle according to the detection result acquired by the acquisition unit, and a behavior determined by the determination unit An automatic operation control device comprising: an operation control unit that causes the vehicle to execute the operation. The acquisition unit further acquires information indicating the reliability of the obstacle detection result, and the determination unit determines different behaviors according to the information indicating the reliability.

According to this automatic driving control device, it is possible to cause the vehicle to execute a suitable obstacle avoidance behavior according to the reliability of the input obstacle detection result.

[Item 2]
When the reliability of the obstacle detection result is relatively low, the determination unit determines to start the behavior earlier to avoid collision with the obstacle than when the reliability is relatively high. Also good.

¡According to this aspect, it is possible to improve the certainty of avoiding collision with an obstacle.

[Item 3]
When the reliability of the obstacle detection result is relatively low, the determination unit behaves so as to increase the distance to the obstacle when avoiding a collision with the obstacle compared to the case where the reliability is relatively high. May be determined.

[Item 4]
An automatic driving system including an obstacle detection device and an automatic driving control device. The obstacle detection device includes a light emitting unit that emits light toward the outside of the vehicle, a detection unit that detects an obstacle according to the time until the light emitted from the light emitting unit is reflected back to the object, and the light emission. An estimation unit for estimating the reliability of the detection result by the detection unit according to the intensity of the background light received when the unit is not emitting light, information on the obstacle detected by the detection unit, and the reliability estimated by the estimation unit And a transmitter for transmitting the information to the automatic operation control device. The automatic driving control device determines the behavior of the vehicle according to the obstacle information transmitted from the obstacle detection device so as to avoid a collision with the obstacle, and the behavior determined by the decision unit. An operation control unit to be executed by the vehicle. The determination unit determines different behaviors according to the reliability information transmitted from the obstacle detection device.

According to this automatic driving system, it is possible to cause the vehicle to execute a suitable obstacle avoidance behavior according to the reliability estimated by the obstacle detection result together with the obstacle detection.

[Item 5]
The estimation unit may estimate the reliability by comparing the intensity of background light with the intensity of light emitted by the light emitting unit.

According to this aspect, the reliability of the obstacle detection result can be estimated without providing a new sensor or the like.

[Item 6]
When the reliability estimated by the estimation unit is less than a predetermined threshold, the light emitting unit newly emits light with increased intensity, and the detection unit is in a time until the newly emitted light returns. Accordingly, the presence of an obstacle is newly detected, and the estimation unit compares the intensity of the background light with the intensity of the newly emitted light to estimate the reliability of the new detection result by the detection unit. Also good.

According to this aspect, it becomes easy to obtain the detection result of the obstacle with high reliability.

[Item 7]
The obstacle detection device may further include a storage unit that stores the intensity of background light indoors. The estimation unit may estimate the reliability by comparing the intensity of the background light stored in the storage unit with the current intensity of the background light.

The obstacle detection device of the present disclosure is useful because the reliability of the obstacle detection result is improved and the vehicle can have a suitable obstacle avoidance behavior.

DESCRIPTION OF SYMBOLS 20 Vehicle 24 Obstacle detection apparatus 26 Automatic driving control apparatus 28 Automatic driving system 40 TOF camera 42 Light emission part 44 Light receiving part 46 Distance measurement part 48 Obstacle detection part 50 Reliability estimation part 54 Background light intensity memory | storage part 66 Obstacle information acquisition Unit 68 Behavior determining unit 70 Operation control unit

Claims

An acquisition unit for acquiring obstacle detection results;
In accordance with the detection result acquired by the acquisition unit, a determination unit that determines the behavior of the vehicle so as to avoid a collision with an obstacle,
An operation control unit for causing the vehicle to execute the behavior determined by the determination unit;
With
The acquisition unit further acquires information indicating the reliability of the obstacle detection result,
The automatic operation control device, wherein the determining unit determines different behaviors according to information indicating the reliability.
The determination unit determines that the start of the behavior for avoiding a collision with the obstacle is earlier when the reliability of the obstacle detection result is relatively low than when the reliability is relatively high. The automatic operation control device according to claim 1.
When the reliability of the obstacle detection result is relatively low, the determination unit increases the distance from the obstacle when avoiding a collision with the obstacle compared to the case where the reliability is relatively high. The automatic operation control device according to claim 1 or 2, wherein the behavior is determined.
Equipped with obstacle detection device and automatic operation control device,
The obstacle detection device includes:
A light emitting unit that emits light toward the outside of the vehicle;
A detection unit that detects an obstacle according to a time until the light emitted from the light-emitting unit is reflected by the object and returns;
An estimation unit that estimates the reliability of the detection result by the detection unit according to the intensity of background light received when the light emitting unit is not emitting light,
Including information on the obstacle detected by the detection unit, and a transmission unit that transmits information on the reliability estimated by the estimation unit to the automatic operation control device,
The automatic operation control device is
A determination unit that determines the behavior of the vehicle so as to avoid a collision with the obstacle according to the information of the obstacle transmitted from the obstacle detection device;
An operation control unit that causes the vehicle to execute the behavior determined by the determination unit,
The said determination part determines a different behavior according to the information of the reliability transmitted from the said obstacle detection apparatus, The automatic driving system characterized by the above-mentioned.
The automatic operation system according to claim 4, wherein the estimation unit estimates the reliability by comparing the intensity of the background light and the intensity of light emitted by the light emitting unit.
When the reliability estimated by the estimation unit is less than a predetermined threshold,
The light emitting unit newly emits light with increased intensity,
The detection unit newly detects the presence of an obstacle according to the time until the newly emitted light returns,
6. The automatic driving system according to claim 5, wherein the estimation unit compares the intensity of the background light with the intensity of the newly emitted light to estimate the reliability of a new detection result by the detection unit. .
The obstacle detection device further includes a storage unit that stores the intensity of background light indoors,
The automatic operation system according to claim 4, wherein the estimation unit estimates the reliability by comparing a background light intensity stored in the storage unit with a current background light intensity.