Method and device for detecting abnormalities in a steering robot

Abstract

To provide a method for detecting an abnormality of a steering robot which can achieve the safer steering robot by achieving a duplexed abnormality detection function.SOLUTION: A steering robot is steered by rotating a tool rotating together with a steering wheel by an actuator. A target rotation position of the tool is determined on the basis of a control signal of the actuator. A first rotation difference (step S2) between the target rotation position and a first rotation position of the tool detected by a rotation sensor to be an internal sensor of the steering robot and a second rotation difference (step S5) between the target rotation position and a second rotation position of the tool detected by an optical sensor for monitoring the tool from the outside are monitored. When the first rotation difference is the first threshold or more (YES in step S4) or the second rotation difference is the second threshold or more (YES in step S6), it is determined that an abnormality occurs in the steering robot (step S7).SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present disclosure relates to a method and device for detecting an abnormality in a steering robot. [Background technology]

[0002] Steering robots such as those disclosed in the following Non-Patent Document 1 and Patent Document 1 are known. These steering robots include a jig that is attached to the steering wheel of a vehicle and a main body that is fixed to the vehicle body. The jig is fixed to the steering wheel and rotates together with the steering wheel. The steering wheel can be rotated by moving the jig relative to the main body using an actuator. The steering robot also includes a controller that controls the rotation of the jig, i.e., the rotation of the steering wheel. These steering robots appear to be primarily used to improve the reproducibility of specified steering wheel operations in vehicle dynamics tests conducted on test courses and the like. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] "Steering Robot System" manufactured by HI-TEC srl, handled by Toyo Corporation, [online], [searched April 13, 2020], Internet <URL: https: / / www.toyo.co.jp / mecha / products / detail / srobot.html> [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2012-532766 Summary of the Invention [Problem to be solved by the invention]

[0005] The steering robot described above makes remote control and autonomous driving possible. However, current regulations make it impossible to leave all steering control to such a steering robot when driving on public roads. However, it is possible to use a steering robot to drive a vehicle autonomously, for example, for transportation within private property such as a factory site. However, even when autonomously driving within private property, it is necessary to detect any abnormalities in the steering robot for safety reasons. The steering robot described above is equipped with a self-diagnosis function that uses the motor driver to detect failures in the motor acting as an actuator. However, if there is a bug or malfunction in the self-diagnosis function, there is a possibility that the abnormality will not be detected. For this reason, there is a demand for redundant abnormality detection for steering robots.

[0006] An object of the present disclosure is to provide a method and device for detecting abnormalities in a steering robot, which can realize a safer steering robot by realizing a duplicated abnormality detection function. [Means for solving the problem]

[0007] In the method for detecting an abnormality in a steering robot according to the present disclosure, the steering wheel is rotated by using an actuator of the steering robot to rotate a jig of the steering robot that is attached to the steering wheel and rotates together with the steering wheel, and target rotation positions of the jig and the steering wheel are determined based on control signals from the actuator, a first rotation difference between a first rotation position of the jig and the steering wheel detected by a rotation sensor that is an internal sensor of the steering robot and the target rotation position is monitored, and a second rotation difference between a second rotation position of the jig or the steering wheel detected by an optical sensor that monitors the jig or the steering wheel from the outside and the target rotation position is monitored, and if the first rotation difference is equal to or greater than a first threshold value or if the second rotation difference is equal to or greater than a second threshold value, it is determined that an abnormality has occurred in the steering robot.

[0008] Here, in one abnormality determination, it may be determined whether the first rotation difference is equal to or greater than the first threshold value before determining whether the second rotation difference is equal to or greater than the second threshold value.

[0009] The optical sensor may also be an optical camera that takes a video or image of the jig or the steering wheel.

[0010] Alternatively, the optical sensor may be a LiDAR scanner that detects the shape of the jig or the steering wheel.

[0011] A marking that can be detected by the optical sensor may be provided on the jig or on a portion of the steering wheel whose position changes with rotation.

[0012] The abnormality detection device for a steering robot according to the present disclosure includes a steering robot having a jig attached to a steering wheel and rotating together with the steering wheel, an actuator for rotating the jig, and a rotation sensor as an internal sensor for detecting a first rotation position of the jig and the steering wheel, an optical sensor for monitoring the jig or the steering wheel from the outside, and a controller for determining an abnormality in the steering robot, wherein the controller is configured to determine a target rotation position of the jig and the steering wheel based on a control signal from the actuator, monitor a first rotation difference between the first rotation position detected by the rotation sensor and the target rotation position, and monitor a second rotation difference between a second rotation position detected by the optical sensor and the target rotation position, and determine that an abnormality has occurred in the steering robot if the first rotation difference is equal to or greater than a first threshold value or if the second rotation difference is equal to or greater than a second threshold value. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a schematic plan view illustrating a configuration of an anomaly detection device according to an embodiment. [Figure 2]FIG. 1 is a front view of the jig of the device attached to the steering wheel. [Figure 3] This is a modified example of the jig. [Figure 4] FIG. 2 is a block diagram of the device. [Figure 5] 1 is a flowchart of an anomaly detection method according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, an embodiment of the abnormality detection device for the steering robot 1 will be described with reference to the drawings.

[0015] The vehicle to which the steering robot 1 is attached is, for example, a truck used in a business establishment such as a factory, and FIG. 1 is a plan view of the truck's cabin. As shown in FIG. 1, the anomaly detection device of this embodiment includes the steering robot 1, a controller 2, and an optical camera 3 as an optical sensor. The steering robot 1 is a known steering robot, such as the robot disclosed in the aforementioned [Non-Patent Document 1]. Therefore, only an overview of the steering robot 1 will be described, and a detailed description of its configuration, such as its drive mechanism, will be omitted.

[0016] The vehicle can be remotely driven or automatically driven without a passenger on board using the steering robot 1. Although not shown, the vehicle is also equipped with a pedal robot that operates the brake pedal and accelerator pedal, and a gear robot that operates the gear selector. The steering robot 1 is controlled in coordination with these robots. In addition, the vehicle can have a passenger sit in the driver's seat 4 with the steering robot 1 installed, or the passenger can drive without activating the steering robot 1. The passenger can also drive with the steering robot 1 activated, and the passenger's driving can be recorded in this case.

[0017] The steering robot 1 is equipped with a jig 10 that is attached to the steering wheel 5 and rotates together with the steering wheel 5. The jig 10 of this embodiment has a ring shape that is attached to the front of the rim of the steering wheel 5. As described above, when a passenger drives the vehicle, the jig 10 has a ring shape so that the passenger, i.e., the driver, can operate the jig 10 like a steering wheel. However, if passenger driving is not taken into consideration, the jig 10 that rotates together with the steering wheel 5 does not have to be ring-shaped.

[0018] The steering robot also includes a bracket 11 that holds the jig 10. The bracket 11 has a built-in motor 12 (see FIG. 4) that serves as an actuator for rotating the jig 10. In this embodiment, the bracket 11 is fixed to the inner surface of the front windshield by a suction cup. To rotate the jig 10 with the motor 12, it is necessary to receive the reaction force, and this reaction force is received by the bracket 11 that is fixed to the vehicle body. Note that the bracket 11 need only be fixed to the vehicle body, and is not limited to being fixed by a suction cup. For example, the bracket 11 may be composed of a pole provided between the upper surface of the floor and the inner surface of the roof or windshield, and an arm that extends from the pole toward the jig 10.

[0019] In this embodiment, the controller 2 of the anomaly detection device is installed on the floor between the driver's seat 4 and the passenger seat. The controller 2 in this embodiment also functions as a controller for steering control of the steering robot 1. The controller 2 is an electronic device including a CPU, ROM, RAM, storage such as an SSD or an HHD, an I / O device, etc. The above-mentioned steering robot 1 is connected to this controller 2. Since the controller 2 also functions as a controller for the steering robot 1, the detection results of various sensors mounted on the steering robot 1 are grasped by the controller 2. In addition, various actuators mounted on the steering robot 1 are also controlled by the controller 2.

[0020] The controller 2 is also connected to an optical camera 3 that captures video or images of the jig 10 or the steering wheel 5. In this embodiment, the optical camera 3 is fixed by a suction cup to glass that is fitted into the rear panel diagonally behind the driver's seat 4. In this embodiment, the optical camera 3 is placed diagonally behind the driver's seat 4, taking into consideration the case where an occupant sits in the driver's seat 4. By placing the optical camera 3 in this position, the jig 10 or the steering wheel 5 can be stably photographed even if an occupant sits in the driver's seat 4.

[0021] The optical camera 3 is an optical sensor that monitors the jig 10 or the steering wheel 5 from the outside. In this embodiment, the optical camera 3 is used as the optical sensor, but other types of optical sensors may be used. For example, a LiDAR (Light Detection And Ranging) scanner that detects the shape of the jig 10 or the steering wheel 5 may be used as the optical sensor. The LiDAR scanner detects the shape of an object by scanning using a ranging sensor that uses laser light. Ranging sensors used in LiDAR scanners include a ToF (Time of Flight) system and an FMCW (Frequency Modulated Continuous Wave) system, but the system is not limited thereto. Some LiDAR scanners can measure the brightness of the object surface as well as the object shape.

[0022] In this embodiment, the jig 10 or the steering wheel 5 is monitored from the outside by an optical sensor such as an optical camera 3 or a LiDAR scanner; specifically, their rotational positions are monitored. Here, the optical camera 3 may not be able to capture a sufficient image of the jig 10 or the steering wheel 5. For example, if the steering wheel 5 is hidden by the jig 10, it may not be possible to capture a sufficient image of the entire steering wheel 5. However, since the jig 10 and the steering wheel 5 rotate together, the rotational positions of the jig 10 and the steering wheel 5 can be determined by monitoring either one of them. In this embodiment, the jig 10 is attached to the steering wheel 5, and therefore the rotational position of the jig 10 is detected by the optical camera 3.

[0023] The following description will be given taking as an example the monitoring of the jig 10 by the optical camera 3. Although the optical camera 3 may be a still image camera, in this embodiment the optical camera 3 acquires images. Furthermore, the optical camera 3 in this embodiment acquires color images. However, the optical camera 3 may also acquire infrared images.

[0024] First, initial calibration is performed when the optical camera 3 is installed. This is an initial step required for the controller 2 to process the image captured by the optical camera 3 and detect the rotational position of the jig 10 in the image. Calibration is necessary because the position of the jig 10 in the image frame and the brightness of the image change depending on the installation state of the optical camera 3. Furthermore, in order to detect the rotational position from the appearance, i.e., shape, etc., of the jig 10 in the image, calibration of the rotational position is also performed based on an image of the vehicle traveling straight ahead, etc. In this embodiment, as shown in FIG. 2, in order to make it easier to detect the rotational position of the jig 10, markings 13 that can be detected by the optical camera 3 are applied to parts of the jig 10 whose position changes with rotation.

[0025] In this embodiment, the marking 13 is a yellow stripe and is provided at the 12 o'clock position on the steering wheel 5 when the steering wheel 5 is traveling straight ahead. In this embodiment, only one marking 13 is provided, but multiple markings 13 may be provided. When multiple markings 13 are provided, it is preferable that each marking has a different color. The rotational position of the jig 10 is detected based on the image from the optical camera 3. However, if there is only one marking 13, it is possible that the marking 13 may not be clearly visible depending on the angle of view, or that the marking 13 may not be clearly visible due to the influence of external light. In consideration of such cases, providing multiple markings 13 allows any one of the markings 13 to be clearly visible, improving detection accuracy.

[0026] 3 shows a modified steering wheel 5 and jig 10. In this modified example, an optical camera 3 serving as an optical sensor detects the rotational position of the steering wheel 5, not the jig 10. For this reason, three markings 13 are provided on the steering wheel 5, not the jig 10. As in this modified example, the steering wheel 5, not the jig 10, may be monitored from the outside by the optical camera 3. Alternatively, both the steering wheel 5 and the jig 10, which rotate together, may be monitored from the outside by the optical camera 3.

[0027] As described above, the rotational position of the jig 10 is detected using the optical camera 3, but the rotational position is also detected by an internal sensor of the steering robot 1. The internal sensor used here is a rotation sensor that detects the rotational position of the jig 10, which is part of the steering robot 1, and more specifically, is the encoder 14 of the motor 12. In this embodiment, the encoder 14 is built into the motor 12, and the rotational position of the motor 12 can be detected by the encoder 14. Note that other types of rotation sensors, such as a potentiometer, may be used instead of the encoder 14. Note that, for the sake of distinction, the rotational position detected using the encoder 14 will be referred to as a first rotational position, and the rotational position detected using the optical camera 3 will be referred to as a second rotational position, hereinafter.

[0028] The first rotational position detected by the encoder 14 is used for abnormality detection control by the abnormality detection device, but is also used for normal steering control of the steering robot 1. As described above, the controller 2 also performs steering control of the steering robot 1 before abnormality detection control. In steering control, the controller 2 sends a control signal to the motor 12 to rotate the jig 10 so as to rotate the steering wheel 5 in a desired rotational state. Hereinafter, the rotational position of the jig 10 determined based on this control signal will be referred to as the target rotational position. In steering control, the controller 2 performs feedback control using the determined target rotational position.

[0029] The anomaly detection device of this embodiment performs two types of anomaly detection: one using the rotational difference between the target rotational position and the first rotational position, and the other using the rotational difference between the target rotational position and the second rotational position. To distinguish between these two types of anomaly detection functions, the rotational difference between the target rotational position and the first rotational position is referred to as the "first rotational difference," and the rotational difference between the target rotational position and the second rotational position is referred to as the "second rotational difference." FIG. 4 shows a block diagram of the anomaly detection device. The top row in FIG. 4 shows the function of detecting anomalies based on the first rotational difference using the encoder 14 as a rotation sensor of the steering robot 1. Meanwhile, the bottom row in FIG. 4 shows the function of detecting anomalies based on the second rotational difference using the optical camera 3 as an optical sensor that externally monitors the steering robot 1.

[0030] More specifically, the rotational position is detected as the rotational angle of the jig 10 and the steering wheel 5. Normally, the steering wheel 5 makes three to four rotations from a clockwise fully locked position to a counterclockwise fully locked position. Therefore, even when the steering wheel 5 is in the state shown in FIG. 2, it is not necessarily in a straight-ahead state (0°), but may be in a state where it has rotated one turn clockwise (+360°) or one turn counterclockwise (-360°). The encoder 14 is of an absolute type, and can distinguish and detect these rotation angles. Note that even if the encoder 14 is of an incremental type, it is possible to distinguish and detect these rotation angles by processing the output signal of the encoder 14.

[0031] On the other hand, the optical camera 3, which serves as an optical sensor that monitors the jig 10 from the outside, cannot distinguish these rotation angles from an image at a certain point in time, i.e., still image data. However, in this embodiment, the optical camera 3 detects images, i.e., continuous data, and the controller 2 processes this continuous data, so these rotation angles can be distinguished. Furthermore, even if the optical camera 3 acquires still images instead of images, abnormality determination is performed at predetermined time intervals, so if multiple still image data sets at predetermined time intervals are treated as continuous data, these rotation angles can be distinguished.

[0032] Alternatively, even when the data are not processed as continuous data, it is rare for the first rotational position and the second rotational position to deviate by even one rotation, and an abnormality is determined before the deviation reaches that extent. Therefore, when detecting the second rotational position using the optical camera 3, the detection result of the first rotational position based on the encoder 14 may be used. For example, when detecting the second rotational position from still image data acquired by the optical camera 3, the detection result of the first rotational position can also be used to determine whether the rotation is clockwise or counterclockwise and how many rotations have occurred since the straight-ahead state.

[0033] 4, the output from the encoder 14 is used by a steering angle calculation unit 20 of the controller 2 to calculate a first rotational position of the jig 10 and the steering wheel 5. Based on the calculated first rotational position, a steering angle difference calculation unit 21 calculates a first rotational difference between the target rotational position and the first rotational position. Based on the calculated first rotational difference, an abnormality determination unit 22 determines whether or not an abnormality has occurred.

[0034] Specifically, it is determined whether the first rotation difference is equal to or greater than a predetermined threshold. For the sake of distinction, this predetermined threshold for the first rotation difference is referred to as the first threshold. If the first rotation difference is equal to or greater than the first threshold, it means that there is a large deviation between the target position and the first rotation position detected by the encoder 14. In this case, it can be determined that the steering robot 1 is not functioning normally, or that the steering control result is not the desired result due to the influence of the road surface, etc. In other words, it is determined that some kind of abnormality has occurred.

[0035] The same applies to the abnormality determination based on the other optical camera 3. The image from the optical camera 3 is image-processed by the steering angle calculation unit 23 of the controller 2, and the second rotational positions of the jig 10 and the steering wheel 5 are calculated. In this image processing, the detection accuracy can be improved if the above-mentioned markings 13 are provided. Based on the calculated second rotational position, the steering angle difference calculation unit 24 calculates a second rotational difference between the target rotational position and the second rotational position. Based on the calculated second rotational difference, the abnormality determination unit 25 determines whether or not an abnormality has occurred.

[0036] Specifically, it is determined whether the second rotation difference is equal to or greater than a predetermined second threshold. If the second rotation difference is equal to or greater than the first threshold, it is determined that an abnormality has occurred. The first threshold and the second threshold may be the same value or may be different values. Since the processing method in the controller 2 based on the detection result of the encoder 14 is different from the processing method in the controller 2 based on the detection result of the optical camera 3, it may be possible to improve the accuracy of abnormality determination by setting the first threshold and the second threshold to different values.

[0037] If the abnormality determination unit 22 or 25 determines that an abnormality has occurred, a signal indicating the abnormality is sent to the abnormality processing unit 26, and a control signal for abnormality processing is sent from the abnormality processing unit 26 to the motor 12, which is the object of control. The details of the abnormality processing will be described later.

[0038] FIG. 5 shows a flowchart of the abnormality detection method by the abnormality detection device described above. A single abnormality determination shown in the flowchart of FIG. 5 is repeatedly performed at predetermined time intervals, for example, at 10 to 20 Hz. In this embodiment, first, in step S1, a steering angle is calculated as a first rotational position based on the detection result of encoder 14. Next, in step S2, a steering angle difference corresponding to a first rotational difference between the target rotational position and the first rotational position is calculated. Then, in step S3, it is determined whether the angle difference corresponding to the calculated first rotational difference is equal to or greater than a first threshold value. If the result in step S3 is affirmative, abnormality processing is performed in step S7. After step S7, the process returns to step S1.

[0039] On the other hand, if step S3 is negative, then in step S4, a steering angle is calculated as a second rotation position based on the detection result of the optical camera 3. Next, in step S5, a steering angle difference corresponding to a second rotation difference between the target rotation position and the second rotation position is calculated. Then, in step S6, it is determined whether the angle difference corresponding to the calculated second rotation difference is equal to or greater than a second threshold value. If step S6 is positive, then abnormality processing is performed in step S7. After step S7, the processing returns to step S1. The flowchart of FIG. 5 is ended simultaneously when the system of the steering robot 1 is shut down.

[0040] In the abnormality processing in step S7 when an abnormality is detected, for example, an operating command such as a stop command may be sent to the motor 12 to stop the steering control by the steering robot 1 (see FIG. 4). If the remote driving or automatic driving of the vehicle is also stopped at the same time, the vehicle's running is stopped in coordination with the other robots described above. A warning may be displayed in the cabin or a warning sound may be emitted. As described above, since there may be occupants in the vehicle, these warnings are intended to warn the occupants.

[0041] That is, in this embodiment, abnormality detection is performed in a doubled manner by detecting abnormalities using the encoder 14 as a rotation sensor of the steering robot 1 and the optical camera 3 as an optical sensor that monitors the steering robot 1 from the outside. As a result, a safer steering robot 1 can be realized.

[0042] In particular, the steering robot 1 of this embodiment uses the steering robot disclosed in the aforementioned [Non-Patent Document 1]. Therefore, if an attempt is made to duplicate anomaly detection within the steering robot 1, the steering robot 1 itself must be modified, and building a duplicated anomaly detection system is time-consuming. However, as in this embodiment, by adding an optical sensor such as an optical camera 3 to the steering robot 1 to duplicate the anomaly detection system, there is no need to modify the steering robot 1 itself, and a duplicated anomaly detection system can be built relatively easily. Furthermore, while the steering robot 1 of this embodiment can be operated by a human, modifying the steering robot 1 itself would increase its size, making it difficult to operate by a human. However, in this embodiment, there is no need to modify the steering robot 1 itself, and human operation is possible.

[0043] According to the abnormality detection method of this embodiment, the jig 10, which is attached to the steering wheel 5 and rotates together with the steering wheel 5, is rotated using a motor 12 serving as an actuator, thereby rotating the steering wheel 5. Target rotational positions of the jig 10 and the steering wheel 5 are determined based on control signals from the motor 12. A first rotational difference between a first rotational position detected by an encoder 14 serving as a rotation sensor, which is an internal sensor, and the target rotational position is monitored. A second rotational difference between a second rotational position detected by an optical camera 3 serving as an optical sensor that externally monitors the jig 10 or the steering wheel 5 and the target rotational position is also monitored. Then, if the first rotational difference is equal to or greater than a first threshold value, or if the second rotational difference is equal to or greater than a second threshold value, it is determined that an abnormality has occurred.

[0044] Therefore, dual abnormality detection is performed by the rotation sensor as an internal sensor and the optical sensor, thereby realizing a safer steering robot 1. One abnormality detection is based on the internal sensor of the steering robot 1 itself, and the other abnormality detection is by the optical sensor that monitors the steering robot 1 from the outside. Therefore, by detecting abnormalities from both the inside and outside of the steering robot 1, more reliable abnormality detection can be achieved.

[0045] According to this embodiment, as shown in the flowchart of FIG. 5 , during one abnormality determination, the encoder 14 is used to determine whether the first rotation difference is equal to or greater than the first threshold value before the optical camera 3 is used to determine whether the second rotation difference is equal to or greater than the second threshold value. That is, abnormality detection by the encoder 14 as a rotation sensor is prioritized over abnormality detection by the optical camera 3 as an optical sensor. This allows for earlier abnormality detection. Furthermore, because the optical sensor that monitors the steering robot 1 from the outside may be affected by external light, etc., prioritizing abnormality detection by the highly reliable rotation sensor allows for more reliable abnormality detection.

[0046] Furthermore, in the above embodiment, the optical sensor is the optical camera 3 that captures video or images of the jig 10 or the steering wheel 5, so that the cost can be reduced. Furthermore, it is also easy to miniaturize the device for the detection method. Furthermore, image processing technology has improved dramatically in recent years, so there are fewer restrictions on the jig 10 or the steering wheel 5 to be monitored, such as restrictions on shape and reflectivity.

[0047] The optical sensor may be a LiDAR scanner that detects the shape of the jig 10 or the steering wheel 5. In this case, it is less susceptible to the influence of external light, and the accuracy of abnormality detection can be improved. In recent years, the sensitivity performance of the optical camera 3 has also improved, but when using the optical camera 3, lighting may be required at night, but with a LiDAR scanner, such equipment is not necessary.

[0048] Furthermore, according to the above embodiment, the marking 13 that can be detected by an optical sensor such as the optical camera 3 or a LiDAR scanner is applied to a portion of the jig 10 or the steering wheel 5 whose position changes with rotation. This improves the detection accuracy of the second rotation position by the optical sensor. As a result, the accuracy of abnormality detection can be improved.

[0049] In the above embodiment, accuracy is improved by using the color of the marking 13 with the optical camera 3 that captures color images. However, the optical camera 3 may also capture monochrome images or videos. In that case, it is sufficient to use a camera with a surface that improves the brightness of the marking in the captured image or video. As mentioned above, some LiDAR scanners can detect brightness. In this case, it is also sufficient to use a camera with a surface that improves the brightness of the marking in the captured shape data. Furthermore, the marking 13 may be detected not by its color or brightness but by the shape of the symbol mark.

[0050] According to the abnormality detection device of this embodiment, the controller 2 is configured to determine target rotational positions of the jig 10 and the steering wheel 5 based on a control signal from the motor 12 serving as an actuator. The controller 2 is also configured to monitor a first rotational difference between a first rotational position detected by the encoder 14 serving as a rotation sensor and the target rotational position. The controller 2 is further configured to monitor a second rotational difference between a second rotational position detected by the optical camera 3 serving as an optical sensor and the target rotational position. The controller 2 is configured to determine that an abnormality has occurred when the first rotational difference is equal to or greater than the first threshold value or when the second rotational difference is equal to or greater than the second threshold value.

[0051] Therefore, dual abnormality detection is performed by the rotation sensor as an internal sensor and the optical sensor, thereby realizing a safer steering robot 1. One abnormality detection is based on the internal sensor of the steering robot 1 itself, and the other abnormality detection is by the optical sensor that monitors the steering robot 1 from the outside. Therefore, by detecting abnormalities from both the inside and outside of the steering robot 1, more reliable abnormality detection can be achieved.

[0052] In the above embodiment, the jig 10 is attached to the front side of the steering wheel 5, but it may also be attached to the back side of the steering wheel 5. Also, the camera 3 is installed diagonally behind the driver's seat 4, but it may also be installed in another position, such as above the steering wheel 5. Particularly in trucks and the like, the rim surface of the steering wheel 5 is often close to horizontal, and in such cases it is easier to monitor the rotational position of the steering wheel 5 and the jig 10 from above. Furthermore, the vehicle on which the abnormality detection device is installed is not limited to trucks as in the above embodiment.

[0053] Furthermore, this disclosure can improve the safety of remote-controlled and autonomous vehicle driving, thereby contributing, for example, to Goal 9 of the Sustainable Development Goals (SDGs) led by the United Nations, which is to "Develop resilient infrastructure, promote inclusive and sustainable industrialization, and foster technological innovation." [Explanation of symbols]

[0054] 1 Steering robot 2 Controller 3 Optical camera (optical sensor) 5 Steering wheel 10 Jig 12 Motor (actuator) 13 Marking 14 Encoder (rotation sensor)

Claims

1. A method for detecting an abnormality in a steering robot, comprising: a jig of the steering robot attached to the steering wheel and rotating together with the steering wheel is rotated by an actuator of the steering robot, thereby rotating the steering wheel; determining target rotation positions of the jig and the steering wheel based on the control signal of the actuator; monitor a first rotation difference between a first rotation position of the jig and the steering wheel, which is detected by a rotation sensor that is an internal sensor of the steering robot, and the target rotation position, and also monitor a second rotation difference between a second rotation position of the jig or the steering wheel, which is detected by an optical sensor that monitors the jig or the steering wheel from the outside, and the target rotation position; An abnormality detection method for a steering robot, which determines that an abnormality has occurred if the first rotation difference is equal to or greater than a first threshold value, or if the second rotation difference is equal to or greater than a second threshold value.

2. 2. The method for detecting an abnormality in a steering robot according to claim 1, wherein, during a single abnormality determination, it is determined whether the first rotation difference is equal to or greater than the first threshold value before it is determined whether the second rotation difference is equal to or greater than the second threshold value.

3. 3. The method for detecting an abnormality in a steering robot according to claim 1, wherein the optical sensor is an optical camera that takes a video or image of the jig or the steering wheel.

4. The method for detecting an abnormality in a steering robot according to claim 1 or 2, wherein the optical sensor is a LiDAR scanner that detects the shape of the jig or the steering wheel.

5. 5. A method for detecting an abnormality in a steering robot according to claim 1, wherein a marking detectable by the optical sensor is applied to a portion of the jig or the steering wheel whose position changes with rotation.

6. An abnormality detection device for a steering robot, the steering robot having a jig attached to a steering wheel and rotating together with the steering wheel, an actuator that rotates the jig, and a rotation sensor as an internal sensor that detects first rotation positions of the jig and the steering wheel; an optical sensor that monitors the jig or the steering wheel from the outside; a controller for determining an abnormality in the steering robot, an abnormality detection device for a steering robot, wherein the controller is configured to determine target rotational positions of the jig and the steering wheel based on a control signal from the actuator, monitor a first rotational difference between the first rotational position detected by the rotation sensor and the target rotational position, and monitor a second rotational difference between the second rotational position detected by the optical sensor and the target rotational position, and determine that an abnormality has occurred if the first rotational difference is equal to or greater than a first threshold value or if the second rotational difference is equal to or greater than a second threshold value.

Images