Joint angle estimation device, joint angle estimation method and program product

Abstract

The invention relates to a hitch angle estimation device, a hitch angle estimation method, and a program product. A hitch angle estimation device according to the present disclosure performs conversion from an observation coordinate system to a world coordinate system on a point sequence representing a tow bar and a lower end portion of a trailer in an image captured by a camera mounted on a vehicle that tows the trailer via the tow bar. In the present invention, linear fitting of a point row representing a tow bar and a lower end of a trailer after the execution of the transformation is performed is performed, and a hitch angle of the trailer in a world coordinate system is calculated on the basis of a straight line representing the tow bar in the world coordinate system and a straight line representing the lower end of the trailer in the world coordinate system obtained by the linear fitting. And executing Kalman filtering processing on the hitching angle of the trailer in the world coordinate system.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a hitch angle estimation device, a hitch angle estimation method, and a program product. BACKGROUND

[0002] In Patent Literature 1 (Japanese Patent Application Publication No. 2023-178952), hitch angle estimation based on an image is described. In the technology described in Patent Literature 1, Kalman filtering is applied to each hitch angle measurement value based on a three-dimensional angle measurement value. Further, in the technology described in Patent Literature 1, each wheel and line path (track) detected in an image is related to a corresponding two-dimensional hitch angle using a Kalman filtering process.

[0003] Further, in Patent Literature 1, a technology for appropriately estimating a hitch angle of a trailer in a world coordinate system based on an image captured by a camera mounted on a vehicle towing a trailer via a drawbar in a case where the image is not clear is not described. Therefore, in the technology described in Patent Literature 1, the hitch angle of the trailer in the world coordinate system cannot be appropriately estimated based on the image captured by the camera mounted on the vehicle towing the trailer via the drawbar in a case where the image is not clear, and the like. SUMMARY

[0004] In view of the above, an object of the present disclosure is to provide a hitch angle estimation device, a hitch angle estimation method, and a program product that can appropriately estimate a hitch angle of a trailer in a world coordinate system based on an image captured by a camera mounted on a vehicle towing a trailer via a drawbar in a case where the image is not clear, and the like.

[0005] (1) One aspect of the present disclosure is a hitching angle estimation device having a processor that performs the following actions: performs a transformation from an observation coordinate system to a world coordinate system on a point row representing a drawbar on an image captured by a camera mounted on a vehicle towing a trailer via the drawbar and a point row representing a lower end portion of the trailer on the image; performs a linear fitting of the point row representing the drawbar after the transformation is performed and a linear fitting of the point row representing the lower end portion of the trailer after the transformation is performed; calculates a hitching angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performs Kalman filter processing on the hitching angle of the trailer in the world coordinate system, the processor using the hitching angle of the trailer in the world coordinate system as an observation function and using a motion model representing the vehicle, the trailer, and the drawbar obtained by performing the transformation from the observation coordinate system to the world coordinate system on the image as a state transition function, thereby outputting the hitching angle of the trailer in the world coordinate system after the Kalman filter processing is performed.

[0006] (2) In the hitching angle estimation device of aspect (1), the Kalman filter for the Kalman filter processing can be an Unscented Kalman Filter (UKF).

[0007] (3) In the hitching angle estimation device of aspect (1) or (2), the motion model representing the vehicle, the trailer, and the drawbar can include the hitching angle of the trailer in the world coordinate system, a steering angle of the vehicle in the world coordinate system, a speed of the vehicle in the world coordinate system, a wheelbase of the vehicle in the world coordinate system, a hitching length of the vehicle in the world coordinate system, and a trailer beam length in the world coordinate system.

[0008] (4) One aspect of the present disclosure is a hitching angle estimation method including: performing transformation from an observation coordinate system to a world coordinate system on a point row representing a drawbar on an image captured by a camera mounted on a vehicle towing a trailer via the drawbar and a point row representing a lower end portion of the trailer on the image; performing linear fitting of the point row representing the drawbar after the transformation is performed and linear fitting of the point row representing the lower end portion of the trailer after the transformation is performed; calculating a hitching angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performing Kalman filter processing on the hitching angle of the trailer in the world coordinate system, in which the hitching angle of the trailer in the world coordinate system is used as an observation function and a motion model representing the vehicle, the trailer, and the drawbar obtained by performing transformation from the observation coordinate system to the world coordinate system on the image is used as a state transition function, thereby outputting the hitching angle of the trailer in the world coordinate system after the Kalman filter processing is performed.

[0009] (5) One aspect of the present disclosure is a program product in which a computer program for causing a processor to perform the following actions is recorded: performing transformation from an observation coordinate system to a world coordinate system on a point row representing a drawbar on an image captured by a camera mounted on a vehicle towing a trailer via the drawbar and a point row representing a lower end portion of the trailer on the image; performing linear fitting of the point row representing the drawbar after the transformation is performed and linear fitting of the point row representing the lower end portion of the trailer after the transformation is performed; calculating a hitching angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performing Kalman filter processing on the hitching angle of the trailer in the world coordinate system, in which the hitching angle of the trailer in the world coordinate system is used as an observation function and a motion model representing the vehicle, the trailer, and the drawbar obtained by performing transformation from the observation coordinate system to the world coordinate system on the image is used as a state transition function, thereby outputting the hitching angle of the trailer in the world coordinate system after the Kalman filter processing is performed.

[0010] According to the present disclosure, even in the case where an image captured by a camera mounted on a vehicle towing a trailer via a drawbar is not clear, etc., a hitching angle of the trailer in a world coordinate system can be appropriately estimated based on the image. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1is a diagram showing one example of a vehicle to which the hitching angle estimation device of the first embodiment is applied.

[0012] Figure 2A is a diagram in which the vehicle 1, the trailer TR, and the drawbar DB are viewed from the side.

[0013] Figure 2B is a diagram in which the vehicle 1, the trailer TR, and the drawbar DB are viewed from below (from the underside). Figure 2A

[0014] Figure 3A is a diagram showing one example of an image IM containing the trailer TR and the drawbar DB, which is captured by the camera 11.

[0015] Figure 3B is a diagram showing one example of a point row PTR1 representing the lower end portion TR1 of the trailer TR and a point row PDB representing the drawbar DB, which are extracted by the extraction section 3C from the image IM shown in Figure 3A

[0016] Figure 3C is a diagram showing one example of a straight line LTR1 in the world coordinate system representing the lower end portion TR1 of the trailer TR and a straight line LDB in the world coordinate system representing the drawbar DB, which are obtained by performing a transformation from the observation coordinate system to the world coordinate system on the point row PTR1 and the point row PDB shown in Figure 3B

[0017] Figure 4 is a diagram showing one example of the hitching angle of the trailer in the world coordinate system, which is calculated by the calculation section.

[0018] Figure 5 is a diagram showing one example of a motion model of a vehicle, a trailer, and a drawbar.

[0019] Figure 6 is a flowchart for explaining one example of the processing performed in the hitching angle estimation device of the first embodiment. DETAILED DESCRIPTION

[0020] Hereinafter, with reference to the drawings, embodiments of the hitching angle estimation device, the hitching angle estimation method, and the program product of the present disclosure will be described.

[0021] <First Embodiment>

[0022] Figure 1 is a diagram showing one example of the vehicle 1 to which the hitching angle estimation device 16 of the first embodiment is applied. Figure 2A and Figure 2B is a diagram showing Figure 1 ​​​Fig. 1 is a diagram showing an example of a relationship of a vehicle 1, a trailer TR, and a drawbar DB. In detail, Figure 2A Fig. 2 is a diagram showing the vehicle 1, the trailer TR, and the drawbar DB as viewed from the side, Figure 2B Fig. 3 is a diagram showing the vehicle 1, the trailer TR, and the drawbar DB as viewed from below (from the lower side) of the vehicle 1. Figure 2A Fig. 4 is a diagram showing the vehicle 1, the trailer TR, and the drawbar DB as viewed from above.

[0023] In the example shown in Figure 1 , Figure 2A and Figure 2B , the vehicle 1 is towing the trailer TR via the drawbar DB. The vehicle 1 is provided with a camera 11, an HMI (Human Machine Interface) 12, a vehicle speed sensor 13, a steering angle sensor 14, a vehicle control device 15, a steering actuator 15A, a brake actuator 15B, a drive actuator 15C, and a hitch angle estimation device 16. The camera 11 is, for example, disposed at a rear end 1R of the vehicle 1. The camera 11 captures an image (for example, a fisheye lens image or the like) IM (see Figure 2A ) containing the trailer TR and the drawbar DB, and transmits the image IM to the hitch angle estimation device 16. Figure 3A

[0024] As shown in Figure 2A and Figure 2B , the drawbar DB is fixed with respect to the trailer TR, and is rotatably connected to the vehicle 1 with the hitch ball HB as a center.

[0025] Figure 3A - Figure 3C Fig. 5 is a diagram showing an example of an image IM containing the trailer TR and the drawbar DB, which is captured by the camera 11. In detail, Figure 3A Fig. 6 is a diagram showing an example of an image IM containing the trailer TR and the drawbar DB, which is captured by the camera 11. Figure 3B Fig. 7 is a diagram showing an example of a point row PTR1 representing a lower end portion TR1 of the trailer TR and a point row PDB representing the drawbar DB, which are extracted from the image IM shown in Figure 3A Fig. 8 is a diagram showing an example of a straight line LTR1 in the world coordinate system representing the lower end portion TR1 of the trailer TR and a straight line LDB in the world coordinate system representing the drawbar DB, which are obtained by performing a transformation from an observation coordinate system to a world coordinate system on the point row PTR1 and the point row PDB shown in Figure 3C Fig. 9, and performing linear fitting by a linear fitting section 3E described later. Figure 3B

[0026] In Figure 1 , Figure 3A - Figure 3C ​​In the illustrated example, the HMI 12 has functions such as accepting various operations by the driver of the vehicle 1 and transmits signals indicating the driver's operations to the vehicle control device 15. The vehicle speed sensor 13 detects the speed (vehicle speed) V [m / s] of the vehicle 1 and transmits this detection result to the vehicle control device 15 and the engagement angle estimation device 16. The steering angle sensor 14 detects the steering angle Φ [deg] and transmits this detection result to the vehicle control device 15 and the engagement angle estimation device 16. The vehicle control device 15 controls the steering actuator 15A, the brake actuator 15B, and the drive actuator 15C based on signals and other signals transmitted from the HMI 12.

[0027] The hook angle estimation device 16 is composed of a microcomputer equipped with a communication interface (I / F) 161, a memory 162, and a processor 163. The communication interface 161 has an interface circuit for connecting the hook angle estimation device 16 to the camera 11, the HMI 12, the vehicle speed sensor 13, the steering angle sensor 14, and the vehicle control device 15. The memory 162 stores programs and various data used in the processing executed by the processor 163. Specifically, the memory 162 stores, for example, the wheelbase WB of the vehicle 1 (see Figure 2A ), the rear overhang OH of the vehicle 1 (refer to Figure 2A The wheelbase WB of the vehicle 1, the rear overhang OH of the vehicle 1, etc. are written into the memory 162, for example, when the vehicle 1 is manufactured. In addition, the memory 162 stores the trailer beam length TBL (the length from the hitch ball HB to the wheel of the trailer TR) (see Figure 2A ) is stored. The trailer beam length TBL is calculated based on the behavior of the trailer TR during the calibration run of the trailer TR, for example, and is written to the memory 162 after the calibration run of the trailer TR. The processor 163 functions as an acquisition unit 3A, an estimation unit 3B, an extraction unit 3C, a conversion unit 3D, a linear fitting unit 3E, a calculation unit 3F, and a processing unit 3G.

[0028] The acquisition unit 3A acquires the image IM including the trailer TR and the drawbar DB captured by the camera 11 . The acquisition unit 3A also acquires the vehicle speed V detected by the vehicle speed sensor 13 and the steering angle Φ detected by the steering angle sensor 14 .

[0029] like Figure 3AAs shown, the inference unit 3B infers the lower end portion TR1 of the trailer TR and the drawbar DB on the image IM based on the image IM acquired by the acquisition unit 3A. In detail, the inference unit 3B, for example, uses a model obtained by performing learning using training data, and thereby infers the lower end portion TR1 of the trailer TR and the drawbar DB on the image IM based on the image IM acquired by the acquisition unit 3A, wherein the training data is a dataset of a learning image captured by a learning camera mounted on a learning vehicle that tows a learning trailer via a learning drawbar, and a label indicating a lower end portion of the learning trailer and a drawbar on the learning image.

[0030] As shown in Figure 3A and Figure 3B , the extraction unit 3C extracts a point row PTR1 on the image IM indicating the lower end portion TR1 of the trailer TR and a point row PDB on the image IM indicating the drawbar DB based on the lower end portion TR1 of the trailer TR and the drawbar DB on the image IM obtained by the inference by the inference unit 3B.

[0031] The conversion unit 3D performs conversion from an observation coordinate system to a world coordinate system on the image IM shown in Figure 3B . That is, the conversion unit 3D performs conversion from an observation coordinate system to a world coordinate system on the point row PTR1 on the image IM shown in Figure 3B indicating the lower end portion TR1 of the trailer TR and the point row PDB indicating the drawbar DB.

[0032] The linear fitting unit 3E performs linear fitting on the point row PTR1 indicating the lower end portion TR1 of the trailer TR after the conversion by the conversion unit 3D is performed, and thereby generates a straight line LTR1 in the world coordinate system indicating the lower end portion TR1 of the trailer TR, as shown in Figure 3C . In addition, the linear fitting unit 3E performs linear fitting on the point row PDB indicating the drawbar DB after the conversion by the conversion unit 3D is performed, and thereby generates a straight line LDB in the world coordinate system indicating the drawbar DB, as shown in Figure 3C .

[0033] The calculation unit 3F calculates the hitch angle ψ [deg] of the trailer TR in the world coordinate system based on the straight line LTR1 in the world coordinate system indicating the lower end portion TR1 of the trailer TR, the straight line LDB in the world coordinate system indicating the drawbar DB, and the like obtained by the linear fitting shown in Figure 3C .

[0034] Figure 4 is a graph indicating one example of the hitch angle ψ [deg] of the trailer TR in the world coordinate system calculated by the calculation unit 3F.Figure 4 the longitudinal axis indicates the hitch angle ψ of the trailer TR in the world coordinate system, Figure 4 the horizontal axis indicates time.

[0035] In Figure 4 , the period in which the hitch angle ψ of the trailer TR in the world coordinate system is zero corresponds to the period in which the steering angle Φ of the vehicle 1 is zero (the period in which the vehicle 1 is straight running). The period in which the hitch angle ψ of the trailer TR in the world coordinate system is greater than zero corresponds to the period in which the steering angle Φ of the vehicle 1 is greater than zero (the period in which the vehicle 1 is turning).

[0036] In Figure 1 - Figure 4 , the processing section 3G performs Kalman filter processing on the hitch angle ψ of the trailer TR in the world coordinate system calculated by the calculation section 3F. The processing section 3G performs the Kalman filter processing using Unscented Kalman Filter (UKF).

[0037] In detail, the processing section 3G uses the hitch angle ψ of the trailer TR in the world coordinate system calculated by the calculation section 3F (refer to Figure 4 ) as an observation function in the Kalman filter processing. Further, the processing section 3G uses a motion model (refer to Figure 5 ) representing the vehicle 1, the trailer TR, and the drawbar DB, which is obtained by performing the transformation from the observation coordinate system to the world coordinate system on the image IM, as a state transition function in the Kalman filter processing.

[0038] Figure 5 is a diagram representing one example of the motion model of the vehicle 1, the trailer TR, and the drawbar DB.

[0039] In Figure 1 - Figure 5 , the motion model representing the vehicle 1, the trailer TR, and the drawbar DB includes the steering angle Φ of the vehicle 1 in the world coordinate system (refer to Figure 5 ). The steering angle Φ detected by the steering angle sensor 14 is used as the steering angle Φ of the vehicle 1 in the world coordinate system.

[0040] Further, the motion model representing the vehicle 1, the trailer TR, and the drawbar DB includes the speed V of the vehicle 1 in the world coordinate system. The speed V of the vehicle 1 detected by the vehicle speed sensor 13 is used as the speed V of the vehicle 1 in the world coordinate system.

[0041] Also, the motion model representing the vehicle 1, the trailer TR, and the drawbar DB includes the wheel base WB of the vehicle 1 in the world coordinate system (refer to Figure 5 ). The wheel base WB of the vehicle 1 written to the memory 162 (refer to Figure 2A ) is used as the wheel base WB of the vehicle 1 in the world coordinate system.

[0042] The hitch length HL of the vehicle 1 in the world coordinate system is included in the motion model of the vehicle 1, the trailer TR, and the drawbar DB (refer to Figure 5 ). As shown in Figure 5 , the hitch length HL of the vehicle 1 in the world coordinate system is the sum of the overhang OH of the vehicle 1 in the world coordinate system and the length ΔL in the world coordinate system from the rear end 1R of the vehicle 1 to the hitch ball HB. The overhang OH of the vehicle 1 in the world coordinate system is used as the overhang OH of the vehicle 1 in the world coordinate system, which is written to the memory 162 (refer to Figure 2A ). The length ΔL in the world coordinate system from the rear end 1R of the vehicle 1 to the hitch ball HB is used as the length ΔL, which is calculated based on the image IM (refer to Figure 3A ) containing the trailer TR and the drawbar DB taken by the camera 11. Figure 2A

[0043] Further, the trailer beam length TBL in the world coordinate system is included in the motion model of the vehicle 1, the trailer TR, and the drawbar DB (refer to Figure 5 ). The trailer beam length TBL in the world coordinate system is used as the trailer beam length TBL, which is calculated based on the behavior of the trailer TR in the calibration run of the trailer TR as described above and written to the memory 162 (refer to Figure 2A ).

[0044] Furthermore, the hitch angle ψ of the trailer TR in the world coordinate system is included in the motion model of the vehicle 1, the trailer TR, and the drawbar DB (refer to Figure 5 ). The hitch angle ψ of the trailer TR in the world coordinate system included in the motion model of the vehicle 1, the trailer TR, and the drawbar DB is used as the hitch angle ψ of the trailer TR in the world coordinate system, which satisfies the following equation.

[0045] ψ = - (V / WB) (tan Φ + (WB / TBL) x sin ψ + (HL / TBL) x tan Φ x cos ψ)

[0046] As with the hitch angle ψ of the trailer TR in the world coordinate system, that is, the hitch angle ψ of the trailer TR in the world coordinate system calculated by the calculation section 3F, shown in Figure 4 , the hitch angle ψ of the trailer TR in the world coordinate system in the above equation varies depending on the change in the steering angle Φ of the vehicle 1 in the world coordinate system.

[0047] In the example shown in Figure 1 - Figure 5 , the processing section 3G outputs the hitch angle ψ of the trailer TR in the world coordinate system after the above Kalman filter processing is performed.

[0048] That is, in the example shown in Figure 1 - Figure 5 , the processing section 3G outputs the hitch angle ψ of the trailer TR in the world coordinate system after the above Kalman filter processing is performed.In the example shown, the processing section 3G estimates the hitch angle ψ of the trailer TR in the world coordinate system by appropriately combining the observation function and the state transition function (in detail, reflecting the hitch angle ψ of the trailer TR in the world coordinate system calculated last time).

[0049] Therefore, in Figure 1 - Figure 5 In the example shown, even in a case where the image IM captured by the camera 11 mounted on the vehicle 1 that is towing the trailer TR via the drawbar DB is not clear, etc., the hitch angle ψ of the trailer TR in the world coordinate system can be appropriately estimated based on the image IM.

[0050] In detail, in Figure 1 - Figure 5 In the example shown, in a case where the image IM captured by the camera 11 mounted on the vehicle 1 that is towing the trailer TR via the drawbar DB is not clear, instead of estimating the hitch angle ψ of the trailer TR in the world coordinate system based only on the unclear image IM, the hitch angle ψ of the trailer TR in the world coordinate system at the time of estimating the hitch angle ψ of the trailer TR in the world coordinate system last time is considered, and the hitch angle ψ of the trailer TR in the world coordinate system this time is estimated. Therefore, it is possible to suppress the possibility that the hitch angle ψ of the trailer TR in the world coordinate system is inappropriately estimated based on the unclear image IM.

[0051] Figure 6 is a flowchart for describing one example of the process performed in the hitch angle estimation device 16 of the first embodiment.

[0052] In Figure 6 In the example shown, in step S10, the acquisition section 3A acquires the image IM including the trailer TR and the drawbar DB captured by the camera 11. Further, the acquisition section 3A acquires the vehicle speed V detected by the vehicle speed sensor 13 and the steering angle Φ detected by the steering angle sensor 14. Also, the acquisition section 3A acquires the wheel base WB of the vehicle 1, the rear overhang OH of the vehicle 1, the trailer beam length TBL, and the like written to the memory 162.

[0053] In step S11, the estimation section 3B estimates the lower end portion TR1 of the trailer TR and the drawbar DB on the image IM based on the image IM acquired in step S10.

[0054] In step S12, the extraction section 3C extracts the point row PTR1 on the image IM indicating the lower end portion TR1 of the trailer TR and the point row PDB on the image IM indicating the drawbar DB based on the lower end portion TR1 of the trailer TR and the drawbar DB on the image IM obtained by the estimation in step S11.

[0055] In step S13, the conversion section 3D performs conversion from the observation coordinate system to the world coordinate system on the point row PTR1 representing the lower end portion TR1 of the trailer TR and the point row PDB representing the drawbar DB on the image IM.

[0056] In step S14, the linear fitting section 3E performs linear fitting of the point row PTR1 representing the lower end portion TR1 of the trailer TR after step S13 is performed, and generates a straight line LTR1 representing the lower end portion TR1 of the trailer TR in the world coordinate system. Further, the linear fitting section 3E performs linear fitting of the point row PDB representing the drawbar DB after step S13 is performed, and generates a straight line LDB representing the drawbar DB in the world coordinate system.

[0057] In step S15, the calculation section 3F calculates the hitching angle ψ [deg] of the trailer TR in the world coordinate system on the basis of the straight line LTR1 representing the lower end portion TR1 of the trailer TR in the world coordinate system, the straight line LDB representing the drawbar DB in the world coordinate system, and the like, which are obtained by linear fitting as shown in the drawing. Figure 3C

[0058] In step S16, the processing section 3G performs Kalman filter processing on the hitching angle ψ of the trailer TR in the world coordinate system calculated in step S15.

[0059] <Second Embodiment>

[0060] The vehicle 1 to which the hitching angle estimation device 16 of the second embodiment is applied is configured similarly to the vehicle 1 to which the hitching angle estimation device 16 of the first embodiment is applied, except for the points described later.

[0061] As described above, in the vehicle 1 to which the hitching angle estimation device 16 of the first embodiment is applied, the conversion section 3D performs conversion from the observation coordinate system to the world coordinate system on the image IM shown in the drawing by using a publicly known technique. Figure 3B

[0062] On the other hand, in the vehicle 1 to which the hitching angle estimation device 16 of the second embodiment is applied, the conversion section 3D performs conversion from the observation coordinate system to the world coordinate system on the image IM shown in the drawing by using a technique other than the publicly known technique (for example, a technique specific to the manufacturer of the vehicle 1). Figure 3B

[0063] ​​​While the embodiments of the hitching angle estimation device, the hitching angle estimation method, and the program product of the present disclosure have been described above with reference to the accompanying drawings, the hitching angle estimation device, the hitching angle estimation method, and the program product of the present disclosure are not limited to the above-described embodiments, and can be appropriately changed without departing from the gist of the present disclosure. The configurations of the respective examples of the above-described embodiments can also be appropriately combined. In the respective examples of the above-described embodiments, the processing performed in the hitching angle estimation device 16 has been described as software processing performed by executing a program, but the processing performed in the hitching angle estimation device 16 can also be processing performed by hardware. Alternatively, the processing performed in the hitching angle estimation device 16 can also be processing that combines both software and hardware. Furthermore, the program stored in the memory 162 of the hitching angle estimation device 16 (a program that realizes the functions of the processor 163 of the hitching angle estimation device 16) can be provided, circulated, and the like, for example, recorded in a computer-readable storage medium (program product) such as a semiconductor memory, a magnetic recording medium, an optical recording medium, or the like.

Claims

1. A hitching angle estimation device having a processor, wherein the processor performs the following actions: performing a transformation from an observation coordinate system to a world coordinate system on a point series representing a drawbar on an image captured by a camera mounted on a vehicle towing a trailer via the drawbar and a point series representing a lower end portion of the trailer on the image; performing a linear fitting of the point series representing the drawbar after the transformation is performed and a linear fitting of the point series representing the lower end portion of the trailer after the transformation is performed; calculating a hitching angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performing a Kalman filter process on the hitching angle of the trailer in the world coordinate system, the processor outputting the hitching angle of the trailer in the world coordinate system after the Kalman filter process is performed, using the hitching angle of the trailer in the world coordinate system as an observation function and using a motion model representing the vehicle, the trailer and the drawbar obtained by performing the transformation from the observation coordinate system to the world coordinate system on the image as a state transition function.

2. The hitching angle estimation device according to claim 1, wherein the Kalman filter used for the Kalman filter process is an unscented Kalman filter.

3. The hitching angle estimation device according to claim 1, wherein the motion model representing the vehicle, the trailer and the drawbar includes the hitching angle of the trailer in the world coordinate system, a steering angle of the vehicle in the world coordinate system, a speed of the vehicle in the world coordinate system, a wheelbase of the vehicle in the world coordinate system, a hitching length of the vehicle in the world coordinate system and a trailer beam length in the world coordinate system.

4. A hitching angle estimation method comprising: performing a transformation from an observation coordinate system to a world coordinate system on a point series representing a drawbar on an image captured by a camera mounted on a vehicle towing a trailer via the drawbar and a point series representing a lower end portion of the trailer on the image; performing a linear fitting of the point series representing the drawbar after the transformation is performed and a linear fitting of the point series representing the lower end portion of the trailer after the transformation is performed; calculating a hitching angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performing a Kalman filter process on the hitching angle of the trailer in the world coordinate system. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ In the hitch angle estimation method, a hitch angle of the trailer in the world coordinate system is used as an observation function, a motion model of the vehicle, the trailer, and the drawbar obtained by performing a transformation from the observation coordinate system to the world coordinate system on the image is used as a state transition function, and thus a hitch angle of the trailer in the world coordinate system after the Kalman filter processing is output.

5. A program product having a computer program recorded thereon, the computer program for causing a processor to perform the following actions: performing a transformation from an observation coordinate system to a world coordinate system on a point row representing the drawbar on an image captured by a camera mounted on a vehicle towing a trailer via a drawbar and a point row representing a lower end portion of the trailer on the image; performing a linear fitting of the point row representing the drawbar after the transformation is performed and a linear fitting of the point row representing the lower end portion of the trailer after the transformation is performed; calculating a hitch angle of the trailer in the world coordinate system based on a straight line representing the drawbar in the world coordinate system and a straight line representing the lower end portion of the trailer in the world coordinate system obtained by the linear fitting; and performing a Kalman filter processing on the hitch angle of the trailer in the world coordinate system, wherein a hitch angle of the trailer in the world coordinate system is used as an observation function, a motion model of the vehicle, the trailer, and the drawbar obtained by performing a transformation from the observation coordinate system to the world coordinate system on the image is used as a state transition function, and thus a hitch angle of the trailer in the world coordinate system after the Kalman filter processing is output.

Images