Position estimation device, vehicle, own vehicle position estimation method, and computer program

Abstract

To provide a position estimation device in which characteristic points are accurately detected from a picked-up image for parking support at the time of backing operation under a low light intensity environment.SOLUTION: A position estimation device 100, 100a estimating an own vehicle position at the time of backing operation of a vehicle 10, 10a comprises: a backing detection part 111 which detects backing operation; an irradiation instruction part 112 which instructs, when the backing operation is detected, light source devices 400, 800 to radiate near infrared ray toward a rear side of the vehicle; image acquisition part 113 which acquires a picked-up image through imaging the rear side of the vehicle by an imaging device 300 in a state where the near infrared is radiated; and a position estimation part 114 which detects, on the basis of the acquired picked-up image, characteristic points through using a characteristic point extraction device or algorithm based on brightness information, and estimates the own vehicle position through using the detected characteristic points.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a position estimation device, a vehicle, a self-vehicle position estimation method, and a computer program.

Background Art

[0002] As a technology for parking assistance during the reverse operation of a vehicle, various technologies for estimating the self-vehicle position using a captured image obtained by a vehicle-mounted camera have been proposed. As one such technology, a method is known in which feature points in a captured image are detected using a luminance-based feature point extraction algorithm such as FAST (Features from Accelerated Segment Test), the change in the self-vehicle position is estimated from the change over time of such feature points, and further, the self-vehicle position is estimated using such a change in the self-position (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in a low-illumination environment where the illuminance of ambient light is low, such as at night, it is difficult to accurately detect feature points because the difference between ambient light and noise is small. As a result, there is a problem that the estimation accuracy of the self-position decreases. Therefore, a technology capable of accurately detecting feature points from a captured image is desired for parking assistance during reverse operation in a low-illumination environment.

Means for Solving the Problems

[0005] As one aspect of the present disclosure, there is provided a position estimation device (100, 100a) that estimates the position of the host vehicle, which is the position of the vehicle (10, 10a) during reverse operation. This position estimation device includes a reverse detection unit (111) that detects the reverse operation of the vehicle, and when the reverse operation is detected, an irradiation instruction unit (112) that instructs a light source device (400, 800) having a light source capable of irradiating near-infrared light to irradiate the rear of the vehicle with the near-infrared light, an image acquisition unit (113) that acquires an image captured by an imaging device (300) of the vehicle when the rear of the vehicle is imaged in a state where the near-infrared light is irradiated, and a position estimation unit (114) that detects feature points using a feature point extraction device or algorithm based on luminance information based on the acquired captured image and estimates the position of the host vehicle using the detected feature points.

[0006] According to the position estimation device of the above aspect, when the reverse operation is detected, near-infrared light is irradiated to the rear of the vehicle, and feature points are detected based on the captured image obtained by imaging in a state where the near-infrared light is irradiated. Therefore, even in a low-illumination environment where the illuminance of the ambient light is low, the illuminance of the rear of the vehicle can be increased by the near-infrared light. Thus, the difference between the ambient light and the noise can be increased, and feature points can be accurately detected from the captured image for parking assistance during reverse operation in a low-illumination environment.

[0007] The present disclosure can also be realized in various forms. For example, it can be realized in the form of a vehicle, a host vehicle position estimation method, a position estimation device, a computer program for realizing the host vehicle position estimation method, a non-transitory recording medium recording such a computer program, and the like.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Mode for Carrying Out the Invention

[0009] A. First Embodiment: A1. Device Configuration: The position estimation device 100 shown in FIG. 1 is mounted on and used in the vehicle 10. First, the configuration of the vehicle 10 will be described. The vehicle 10 is a moving body that travels on wheels and may be any type of moving body such as a vehicle equipped with a gasoline engine, a battery electric vehicle (BEV), a hybrid vehicle, or a fuel cell vehicle. The vehicle 10 has a manual driving mode and an automatic driving mode as driving modes. Manual driving is a driving mode in which the driver of the vehicle 10 operates a steering wheel, an accelerator pedal, a brake pedal, etc. to drive the vehicle 10. Automatic driving is a driving mode in which at least one of acceleration / deceleration, steering, and braking of the vehicle 10 is autonomously controlled without depending on the driver's instruction. In this embodiment, automatic driving is executed when performing an operation of reversing and parking (hereinafter referred to as "reverse parking operation"). In this embodiment, the switching between manual driving and automatic driving is performed when a predetermined operation such as pressing a physical operation button (not shown) or pressing an operation button on a menu screen displayed on an operation display (not shown) is executed by the driver.

[0010] In addition to the position estimation device 100 described in detail later, the vehicle 10 includes an operation control unit 200, an imaging device 300, a light source device 400, a driving device 510, a steering device 520, and a braking device 530.

[0011] The operation control unit 200 controls the operation of the vehicle 10. Specifically, the operation control unit 200 controls the operation of the actuators that change the acceleration, deceleration, and steering angle of the vehicle 10, thereby controlling the operation of the vehicle 10, that is, the "running", "turning", and "stopping" of the vehicle 10. In the present embodiment, the above-described actuators include the actuator of the drive device 510, the actuator of the steering device 520, and the actuator of the braking device 530. The operation control unit 200 can transmit and receive data to and from the position estimation device 100, the drive device 510, the steering device 520, and the braking device 530 via a network such as CAN (Control Area Network). In the present embodiment, the operation control unit 200 is configured as an ECU (Electronic Control Unit) including a CPU and a memory.

[0012] The drive device 510 is a device for driving the vehicle 10. For example, a motor or an inverter device for controlling the rotation of the motor is applicable. The steering device 520 is a device for changing the traveling direction of the vehicle 10. For example, a power steering motor or a hydraulic device is applicable. The braking device 530 is a device for applying brakes to the vehicle 10. For example, a hydraulic device or a valve is applicable.

[0013] The imaging device 300 captures the rear of the vehicle 10 to obtain a captured image. The imaging device 300 has a light-receiving element capable of receiving the reflected light of near-infrared light irradiated from the light source device 400, and captures a subject projected by such reflected light. "Near-infrared light" means, for example, electromagnetic waves with wavelengths of 780 nm (nanometers) to 2500 nm. In the present embodiment, the imaging device 300 is attached to the upper part of the outer surface facing the rear among the body of the vehicle 10. Therefore, compared with the configuration in which the imaging device 300 is attached to the interior of the vehicle 10, there are fewer obstacles and the influence of the indoor light environment is low, so it is possible to obtain a more suitable captured image for obtaining feature points. The angle of view of the imaging device 300 and the orientation of the imaging device 300 are such that the road surface portion where the vehicle 10 is scheduled to travel when reversing, the portions shifted from such road surface portion by at least the width of one vehicle to the left and right respectively, and the regions from these portions to positions at least the height of the vehicle 10 vertically upward can be included in the imaging range. The imaging device 300 is electrically connected to the position estimation device 100 and is configured to be able to transmit the image data of the captured image to the position estimation device 100.

[0014] The light source device 400 has a light source (not shown) capable of irradiating near-infrared light and can irradiate the rear of the vehicle 10. In the present embodiment, like the imaging device 300, the light source device 400 is attached to the upper outer side of the outer surface facing the rear of the body of the vehicle 10. Therefore, compared with the configuration in which the light source device 400 is attached inside the passenger compartment of the vehicle 10, there are fewer obstacles and the near-infrared light can be appropriately irradiated to the rear of the vehicle 10. The irradiation range of the light source device 400 is a part of the imaging range of the imaging device 300 that is closer to the vehicle 10. Specifically, it is a range that can irradiate at least an area from the current position of the vehicle 10 (hereinafter also referred to as the "own vehicle position") along the vehicle length direction to a predetermined distance backward in the road surface portion where the vehicle is planned to travel when reversing. Since the light source device 400 irradiates near-infrared light to the rear of the vehicle 10 and the imaging device 300 images the subject reflected by the light, objects (ground features) existing behind the vehicle 10 can be imaged even in an environment with low illuminance such as at night. The light source device 400 is electrically connected to the position estimation device 100, and the on / off of the irradiation of near-infrared light is controlled by the position estimation device 100.

[0015] The position estimation device 100 estimates the own vehicle position. In the present embodiment, the position estimation device 100 is configured as an ECU including a CPU 110, a ROM 120, and a RAM 130. The CPU 110 functions as a reverse detection unit 111, an irradiation instruction unit 112, an image acquisition unit 113, and a position estimation unit 114 by expanding and executing a control program stored in advance in the ROM 120 in the RAM 130.

[0016] The reverse detection unit 111 detects the reverse operation of the vehicle 10. Such detection may be performed, for example, by obtaining a detection result indicating that the shift range is "R" (reverse) from a sensor that detects the shift range mounted on the vehicle 10. Alternatively, for example, it may be detected using the detection results of sensors that detect the rotation direction of the wheels or the rotation direction of the axles.

[0017] The irradiation instruction unit 112 instructs the light source device 400 to turn on and off the irradiation of near-infrared light. The instruction of "turn on the irradiation of near-infrared light" corresponds to the instruction to irradiate the rear of the vehicle 10 with near-infrared light. The image acquisition unit 113 acquires the captured image obtained by the imaging device 300. That is, the image acquisition unit 113 acquires the captured image obtained by the imaging device 300 imaging the rear of the vehicle 10.

[0018] Based on the acquired captured image, the position estimation unit 114 detects feature points using a feature point extraction device or algorithm based on luminance information, and estimates the position of the host vehicle using the detected feature points. The "feature point extraction device or algorithm based on luminance information" refers to an algorithm or a device that executes the algorithm for detecting feature points from the difference information by comparing the luminance of each pixel with a predetermined threshold luminance based on the captured image obtained by the imaging device 300 imaging the traveling direction during the backward movement of the vehicle 10 and the periphery of the vehicle 10. In the present embodiment, as such a device or algorithm, FAST (Features from Accelerated Segment Test) or a device that executes FAST is used. In FAST, corner points are detected as feature points. Then, the position estimation unit 114 identifies the change in the position of the same corner point at different times, and uses the change in such position to identify the change in the position of the host vehicle. When the position of a predetermined parking space is preset as the position of the host vehicle at the start of travel, for example, the current position of the host vehicle can be identified by using the change in the position of the host vehicle with reference to such position.

[0019] The method for detecting feature points (corner points) by FAST will be described with reference to FIG. 2. In FIG. 2, an example of a captured image, image F1, is shown at the upper left. Also, at the lower right, a certain pixel block B1 included in the image F1 is shown enlarged.

[0020] In FAST, first, the central pixel in a 9-pixel × 9-pixel pixel block B1 is set as the target pixel px, and a total of 16 surrounding pixels px0 to px15 that are positioned to surround the target pixel px are set. Next, it is specified which of the following Brighter pixels and Darker pixels each of the surrounding pixels px0 to px15 corresponds to. At this time, the target pixels are set so as to go around the periphery of the target pixel px counterclockwise from pixel px0 to pixel px15, such that pixel px0 is the first target pixel to be specified, pixel px1 is the second target pixel to be specified, pixel px2 is the third target pixel to be specified, and so on. In the following, the luminance Bn (n is 0 to 15) means the luminance value of each of the surrounding pixels px0 to px15. Also, L means the luminance value of the target pixel px. Further, Bth means the luminance threshold value. (i) Brighter pixel: A pixel that satisfies L + Bth < Bn (ii) Darker pixel: A pixel that satisfies L - Bth > Bn

[0021] As described above, for each of the surrounding pixels px0 to px15, it is determined whether it corresponds to a Brighter pixel or a Darker pixel. As a result, when Brighter pixels are continuously identified a predetermined number of threshold times, and when Darker pixels are continuously identified a predetermined number of threshold times, the target pixel px is detected as a feature point (corner point). The above-mentioned "predetermined number of threshold times" is 9 times. Note that it may be set to other numbers such as 10 times, not limited to 9 times. When detecting corner points with such FAST, in a low illuminance environment, the signal-to-noise ratio (the ratio of the reflected light of near-infrared light to the noise light caused by other ambient light) of the luminance value of each pixel in the captured image is low, and there is a possibility that an accurate luminance value cannot be obtained at the position corresponding to each pixel. In this case, it is impossible to accurately determine whether it is a Brighter pixel or a Darker pixel. As a result, it is impossible to accurately identify the feature point (corner point), and furthermore, there is a possibility that the position of the host vehicle cannot be accurately identified. Therefore, in the position estimation device 100 of the present embodiment, by executing the host vehicle position estimation process described later, in order to provide parking support during the reverse operation in a low illuminance environment, the feature point (corner point) is accurately detected from the captured image.

[0022] A2. Host vehicle position estimation process: The own vehicle position estimation process shown in FIG. 3 is a process for estimating the position of the own vehicle and is executed as part of the parking support process. The "parking support process" is a process for automatically parking the vehicle 10 in a parking space by automatic driving. When an operator or the like presses an operation button (not shown) provided in the vehicle 10, the own vehicle position estimation process is executed as part of the parking support process. At the same time, the driving operation support process is executed by the operation control unit 200. The "driving operation support process" includes a process of determining a route to the planned parking position based on the estimated own vehicle position, a process of setting target positions at predetermined intervals (for example, every 0.5 m) for traveling the determined route, a process of determining an acceleration amount, a steering amount, and a braking amount for moving from the current position to the first target position, and a process of creating a command for realizing the determined acceleration amount, steering amount, and braking amount and outputting it to the drive device 510, the steering device 520, and the braking device 530. As an execution scene of the parking support process, for example, a scene is assumed in which the driver of the vehicle 10 stops the vehicle near a parking lot and presses the operation button in that state to execute the parking support process.

[0023] The reverse detection unit 111 determines whether or not the reverse operation of the vehicle 10 has been detected (step S105). If it is determined that the reverse operation has not been detected (step S105: NO), step S105 is executed again. On the other hand, if it is determined that the reverse operation has been detected (step S105: YES), the irradiation instruction unit 112 instructs the irradiation of near-infrared light. That is, the irradiation instruction unit 112 instructs to irradiate the near-infrared light to the rear of the vehicle 10 (step S110). The irradiation time at this time is set to a time that can ensure the time during which the imaging device 300 can capture an image while the infrared light is irradiated. The image acquisition unit 113 acquires a captured image obtained by the imaging device 300 capturing the rear of the vehicle 10 in a state where the near-infrared light is irradiated (step S115). The position estimation unit 114 estimates the own vehicle position using the captured image acquired in step S115 (step S120). After the completion of step S120, the process returns to step S105.

[0024] In FIG. 4, the detailed procedure of step S120 in FIG. 3 is shown. The position estimation unit 114 executes a predetermined type of image processing on the captured image obtained in step S115 (step S205). Examples of the predetermined type of image processing include demosaicing, noise removal, and distortion correction. Next, the position estimation unit 114 executes grayscale conversion processing on the captured image after the image processing (step S210). In step S210, for example, 8-bit grayscale conversion is executed. Note that the grayscale conversion may be performed with an arbitrary number of bits, not limited to 8 bits. The position estimation unit 114 creates a multi-stage reduced image for the grayscale-converted captured image (step S215). In step S215, for example, an 8-stage reduced image is created. The position estimation unit 114 executes FAST processing to detect feature points (corner points) (step S220). In the above description using FIG. 2, the target pixel px in the pixel block B1 was used as the singular point. However, while sequentially changing this target pixel px in the captured image (for example, image F1), the above-described identification of Brighter pixels and Darker pixels and the process of identifying whether these pixels continue for a predetermined number of times are executed. Then, the above process is executed not only for the pixel block B1 but also for all pixel blocks in the image F1. The position estimation unit 114 estimates the position of the host vehicle using the feature points detected in step S220 (step S225).

[0025] As a result of repeatedly executing the above-described host vehicle position estimation process and driving operation support process, when the parking of the vehicle 10 in the parking space is completed, the host vehicle position estimation process and the driving operation support process (that is, the parking support process) are completed.

[0026] According to the position estimation device 100 of the first embodiment described above, when a reverse operation is detected, near-infrared light is irradiated to the rear of the vehicle 10, and feature points are extracted based on the captured image obtained by capturing an image in a state where the near-infrared light is irradiated. Therefore, even in a situation where the illuminance of the ambient light is low, the illuminance at the rear of the vehicle 10 can be increased by the near-infrared light. Accordingly, the difference between the ambient light and the noise can be increased, and feature points can be accurately detected from the captured image for parking assistance during the reverse operation in a low-illuminance environment.

[0027] Further, the feature point extraction device or algorithm compares the luminance in pixel units with a predetermined threshold luminance based on the captured image obtained by the imaging device 300 capturing the traveling direction during the reverse operation of the vehicle 10 and the periphery of the vehicle 10, and detects feature points from the differential information. Therefore, feature points can be accurately detected.

[0028] Further, since the imaging device 300 is attached to the outer surface of the body of the vehicle 10, a more suitable captured image for obtaining feature points can be obtained compared to a configuration in which the imaging device 300 is attached inside the passenger compartment of the vehicle 10. Also, since the light source device 400 is also attached to the outer surface of the body of the vehicle 10, near-infrared light can be appropriately irradiated to the rear of the vehicle 10.

[0029] B. Second Embodiment: The vehicle 10a of the second embodiment shown in FIG. 5 is different from the vehicle 10 of the first embodiment in that it does not include the light source device 400, includes an additional communication device 600, and includes a position estimation device 100a instead of the position estimation device 100. Since other configurations in the vehicle 10a of the second embodiment are the same as those in the vehicle 10 of the first embodiment, the same reference numerals are given and detailed descriptions thereof are omitted.

[0030] The position estimation device 100a of the second embodiment differs from the position estimation device 100 of the first embodiment in that the CPU 110 also functions as the communication control unit 115. Since the other configurations of the position estimation device 100a in the second embodiment are the same as those of the position estimation device 100 in the first embodiment, the same reference numerals are given and the detailed description thereof is omitted.

[0031] The communication device 600 performs wireless communication. Examples of the wireless communication include wireless communication using a wireless communication service provided by a wireless communication carrier such as 4G (fourth generation) communication or 5G (fifth generation) communication, and wireless LAN. The communication device 600 is electrically connected to the position estimation device 100a, outputs communication data as a wireless signal in response to an instruction from the position estimation device 100a, and receives a wireless signal and outputs the communication data obtained from such a signal to the position estimation device 100a.

[0032] The communication control unit 115 communicates with a light source device (light source device 800) described later via the communication device 600. The communication control unit 115 is a functional unit realized by the CPU 110 expanding and executing a control program stored in advance in the ROM 120 in the RAM 130, similar to the other functional units 111 to 114.

[0033] The light source device 800 is a ground object separate from the vehicle 10a and is installed on the ground around a parking space, that is, an area where the reverse operation of the vehicle 10a is performed. Similar to the light source device 400 of the first embodiment, the light source device 800 is configured to be able to irradiate near-infrared light toward the parking space. The light source device 800 includes a control device and a communication device (not shown). The control device included in the light source device 800 controls the on / off of the irradiation of the near-infrared light. The communication device included in the light source device 800 communicates with the vehicle 10a via the network 700. The network 700 includes a wireless base station 750 near the parking space, and wireless signals are exchanged between the communication device 600 of the vehicle 10a and the wireless base station 750.

[0034] Also in the vehicle 10a of the second embodiment having the above configuration, the host vehicle position estimation process of the first embodiment shown in FIG. 3 is executed. However, in step S110, the irradiation instruction unit 112 instructs to irradiate near-infrared light onto the parking space, that is, the rear of the vehicle, by using the communication with the light source device 800 by the communication control unit 115.

[0035] The position estimation device 100a of the second embodiment described above has the same effects as the position estimation device 100 of the first embodiment. In addition, the position estimation device 100a further includes a communication control unit 115 for communicating with the light source device 800 via the communication device 600 included in the vehicle 10a. Further, when the reverse operation is detected, the communication control unit 115 instructs the light source device 800 to irradiate near-infrared light to the rear of the vehicle 10a by using the communication with the light source device 800 by the communication control unit 115. Therefore, even in a configuration where the vehicle 10a does not mount the light source device, near-infrared light can be irradiated to the rear of the vehicle 10a during the reverse operation.

[0036] C. Other Embodiments: (C1) In each embodiment, the "feature point extraction device or algorithm based on luminance information" was FAST or a device that executes FAST, but the present disclosure is not limited to this. Any algorithm that compares the luminance of each pixel with a predetermined threshold luminance based on the captured image obtained by the imaging devices 400 and 800 imaging the traveling direction during the reverse operation of the vehicle 10 and the periphery of the vehicle 10, such as ORB (Oriented FAST and Rotated Test), and detects feature points from the difference information, or a device that executes that algorithm may be used.

[0037] (C2) In the first embodiment, the imaging device 300 and the light source device 400 were attached to the outer surface of the body of the vehicle 10. Also, in the second embodiment, the imaging device 300 was attached to the outer surface of the body of the vehicle 10. However, the present disclosure is not limited to this. For example, at least one of the imaging device 300 and the light source device 400 may be provided inside the vehicle compartments of the vehicles 10 and 10a. In such a configuration, for the devices provided inside the vehicle compartment, the measures for waterproofing can be simplified, and the measures for coping with the high temperature caused by direct sunlight can also be simplified. Furthermore, by providing it inside the vehicle compartment, it will be in an environment where temperature changes and vibrations are suppressed compared to the configuration attached to the outer surface of the body, and the countermeasures against temperature changes and vibrations can be simplified. For this reason, the manufacturing cost of the vehicles 10 and 10a can be suppressed.

[0038] (C3) In each embodiment, the vehicles 10 and 10a were provided with the imaging device 300 that images a subject projected by the reflected light of near-infrared light. However, in addition to such an imaging device 300, an imaging device (camera) that images a subject projected by the reflected light of visible light may be provided, and feature points may also be detected using the imaging image (hereinafter referred to as "visible light image") obtained by such an imaging device. Specifically, feature points are detected from the imaging image (hereinafter referred to as "near-infrared light image") obtained by the imaging device 300, and separately, feature points are detected from the visible light image. Then, when the feature points detected based on the near-infrared light image match the feature points detected from the visible light image, such feature points may be used for estimating the position of the host vehicle. According to such a configuration, it is possible to suppress the erroneously detected feature points from being used for self-position estimation, and self-position estimation can be performed with higher accuracy. In the above configuration, during the day, feature points may be detected from two types of imaging images, and at night, feature points may be extracted only from the near-infrared light image in the same manner as in each embodiment. By doing so, it is possible to suppress the occurrence of the same problems as in the prior art (feature points cannot be detected accurately) at night.

[0039] (C4) In each embodiment, the "parking assistance process" included vehicle position estimation processing and driving operation assistance processing. However, instead of the driving operation assistance processing, any type of assistance processing that utilizes the estimated own vehicle position may be included. For example, during parking operation, the driving modes of vehicles 10 and 10a are manual driving modes. Based on the own vehicle position estimated by the own vehicle position estimation processing, it is determined whether vehicles 10 and 10a will collide with surrounding ground objects (e.g., curbs, telegraph poles, etc.). If it is determined that there will be a collision, processing such as warning the driver with a warning sound may be included.

[0040] (C5) The position estimation devices 100 and 100a and their methods described in the present disclosure may be realized by a dedicated computer provided by configuring a processor and a memory programmed to execute one or more functions embodied by a computer program. Alternatively, the position estimation devices 100 and 100a and their methods described in the present disclosure may be realized by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Or, the position estimation devices 100 and 100a and their methods described in the present disclosure may be realized by one or more dedicated computers configured by a combination of a processor programmed to execute one or more functions and a memory and a processor configured by one or more hardware logic circuits. Also, the computer program may be stored in a computer-readable non-transitory tangible recording medium as instructions to be executed by a computer.

[0041] The present disclosure is not limited to the above-described embodiments and can be realized in various configurations without departing from the gist thereof. For example, the technical features in each embodiment corresponding to the technical features in the form described in the summary of the invention can be appropriately replaced or combined in order to solve some or all of the above-described problems or to achieve some or all of the above-described effects. Also, if the technical feature is not described as essential in this specification, it can be appropriately deleted.

[0042] The present disclosure may be implemented in the following forms. [Form 1] A position estimation device (100, 100a) that estimates the own vehicle position, which is the position of the vehicle (10, 10a) during a reverse movement of the vehicle, a reverse detection unit (111) that detects the reverse movement of the vehicle; an irradiation instruction unit (112) that, when the reverse movement is detected, instructs a light source device (400, 800) having a light source capable of irradiating near-infrared light to irradiate the rear of the vehicle with the near-infrared light; an image acquisition unit (113) that acquires an image captured image obtained by the imaging device (300) of the vehicle capturing the rear of the vehicle in a state where the near-infrared light is irradiated; a position estimation unit (114) that detects feature points using a feature point extraction device or algorithm based on luminance information based on the acquired captured image, and estimates the own vehicle position using the detected feature points; A position estimation device comprising: [Form 2] In the position estimation device according to Form 1, the feature point extraction device or algorithm compares the luminance in pixel units with a predetermined threshold luminance based on the captured image obtained by the captured image capturing the traveling direction during the reverse movement of the vehicle and the periphery of the vehicle, and detects the feature points from the difference information. A position estimation device. [Form 3] In the position estimation device (100a) according to Form 1, the light source device (800) is provided on a feature other than the vehicle and on a feature around the area where the reverse movement is performed, the position estimation device further includes a communication control unit (115) for communicating with the light source device via a communication device (600) of the vehicle, The irradiation instruction unit is a position estimation device that, when the backward movement is detected, uses communication with the light source device by the communication control unit to instruct the light source device to irradiate the near-infrared light to the rear of the vehicle. [Embodiment 4] A vehicle (10, 10a), The position estimation device (100, 100a) according to any one of Embodiments 1 to 3, An operation control unit (200) that controls a parking operation including the backward movement of the vehicle by using the own vehicle position estimated by the position estimation device, A vehicle comprising the same. [Embodiment 5] An own vehicle position estimation method for estimating an own vehicle position that is the position of the vehicle during a backward movement of the vehicle, A step (S105) of detecting a backward movement of the vehicle; A step (S110) of, when the backward movement is detected, instructing a light source device having a light source capable of irradiating near-infrared light to irradiate the near-infrared light to the rear of the vehicle; A step (S115) of acquiring a captured image obtained by a imaging device of the vehicle capturing the rear of the vehicle in a state where the light is irradiated; A step (S120) of detecting feature points by using a feature point extraction device or algorithm based on luminance information based on the acquired captured image, and estimating the own vehicle position by using the detected feature points; An own vehicle position estimation method comprising the same. [Embodiment 6] A computer program for estimating an own vehicle position that is the position of the vehicle during a backward movement of the vehicle, A function of detecting a backward movement of the vehicle, A function of, when the backward movement is detected, instructing a light source device having a light source capable of irradiating near-infrared light to irradiate the near-infrared light to the rear of the vehicle; A function of acquiring a captured image obtained by a imaging device of the vehicle capturing the rear of the vehicle in a state where the light is irradiated; A function of detecting feature points using a feature point extraction algorithm based on luminance information based on the acquired captured image, and estimating the position of the host vehicle using the detected feature points; A computer program for causing a computer to realize the above.

Explanation of Signs

[0043] 10, 10a... vehicle, 100, 100a... position estimation device, 111... reverse detection unit, 400, 800... light source device, 112... irradiation instruction unit, 300... imaging device, 113... image acquisition unit, 114... position estimation unit

Claims

1. A position estimation device (100, 100a) for estimating the position of the host vehicle, which is the position of the vehicle (10, 10a) during a reverse movement of the vehicle, comprising: a reverse detection unit (111) for detecting a reverse movement of the vehicle; an irradiation instruction unit (112) for instructing the light source device (400, 800) having a light source capable of irradiating near-infrared light to irradiate the rear of the vehicle with the near-infrared light when the reverse movement is detected; an image acquisition unit (113) for acquiring an image captured image obtained by the imaging device (300) of the vehicle capturing the rear of the vehicle in a state where the near-infrared light is irradiated; a position estimation unit (114) for detecting feature points using a feature point extraction device or algorithm based on luminance information based on the acquired captured image, and estimating the position of the host vehicle using the detected feature points; A position estimation device comprising:

2. In the position estimation device according to Claim 1, the feature point extraction device or algorithm compares the luminance in pixel units with a predetermined threshold luminance based on the captured image obtained by capturing the traveling direction during the reverse movement of the vehicle and the periphery of the vehicle in the captured image, and detects the feature points from the difference information. A position estimation device.

3. In the position estimation device (100a) according to Claim 1, the light source device (800) is provided on a ground object different from the vehicle and around the ground object in the area where the reverse movement is performed, the position estimation device further comprises a communication control unit (115) for communicating with the light source device via a communication device (600) of the vehicle, the irradiation instruction unit instructs the light source device to irradiate the rear of the vehicle with the near-infrared light using communication with the light source device by the communication control unit when the reverse movement is detected. A position estimation device.

4. A vehicle (10, 10a), comprising: the position estimation device (100, 100a) according to any one of Claims 1 to 3; an operation control unit (200) for controlling a parking operation including the reverse movement of the vehicle using the position of the host vehicle estimated by the position estimation device; A vehicle comprising:

5. A method for estimating the position of the host vehicle, which is the position of the vehicle during a reverse movement of the vehicle, comprising: a step (S105) of detecting a reverse movement of the vehicle; When the reverse movement is detected, a step of instructing a light source device having a light source capable of irradiating near-infrared light to irradiate the rear of the vehicle with the near-infrared light (S110); A step of acquiring a captured image obtained by the imaging device of the vehicle capturing the rear of the vehicle in a state where the light is irradiated (S115); Based on the acquired captured image, a step of detecting feature points using a feature point extraction device or algorithm based on luminance information, and estimating the position of the host vehicle using the detected feature points (S120); A host vehicle position estimation method comprising the above.

6. A computer program for estimating the position of the host vehicle, which is the position of the vehicle during the reverse movement of the vehicle, A function of detecting the reverse movement of the vehicle; When the reverse movement is detected, a function of instructing a light source device having a light source capable of irradiating near-infrared light to irradiate the rear of the vehicle with the near-infrared light; A function of acquiring a captured image obtained by the imaging device of the vehicle capturing the rear of the vehicle in a state where the light is irradiated; Based on the acquired captured image, a function of detecting feature points using a feature point extraction algorithm based on luminance information, and estimating the position of the host vehicle using the detected feature points; A computer program for causing a computer to realize the above.

Images