JP7385388B2 - Self-position estimation device - Google Patents

Description

The present invention relates to a self-position estimating device for estimating the self-position of a moving body.

BACKGROUND ART Conventionally, a self-position of a moving body has been acquired and moved, and methods of acquiring the self-position with high precision are being considered. For example, in order to maintain high reliability of the positions of a plurality of moving bodies as a whole, movement control is performed so as to restore the reliability of their own positions (for example, see Patent Document 1). Furthermore, for example, in order to stabilize the accuracy of self-position estimation using navigation satellites, self-position is estimated by selecting a combination of navigation satellites that can always be received (for example, Patent Document 2 reference). Furthermore, position estimation is also performed by integrating the results of environmental measurement using a stereo camera or the like and information acquired using an encoder or the like (for example, see Patent Document 3).

Japanese Patent Application Publication No. 2017-188066 Japanese Patent Application Publication No. 2017-182692 Japanese Patent Application Publication No. 2008-234350

The technique described in Patent Document 1 improves the accuracy of position estimation using a plurality of moving objects, and is not effective in improving the accuracy of self-position estimation of a single moving object.

Further, the technology described in Patent Document 2 uses a plurality of satellites to estimate a position, and the criterion for selecting the plurality of satellites is whether or not a target satellite is available. Therefore, for example, by avoiding a situation where it becomes impossible to receive a satellite signal used for position estimation, it is possible to prevent the positioning accuracy from decreasing. On the other hand, when moving in a space with many obstacles, it may not be possible to select multiple satellites that are not affected by obstacles, and in such situations, the accuracy of position estimation may decrease. There is.

Furthermore, in the technology described in Patent Document 3, position estimation is performed by integrating the results of environmental measurement and information acquired using an encoder, etc., so for example, if the accuracy of one of the , there is also a problem in that the accuracy of the integrated result decreases.

The present invention has been made to solve the above problems, and an object of the present invention is to provide a self-position estimating device that can realize more accurate self-position estimation regarding a moving body.

In order to achieve the above object, a self-position estimating device according to the present invention is a self-position estimating device that estimates the self-position of a mobile object, and includes a sensor section that acquires information used for estimation regarding the position of the mobile object, and a sensor section that acquires information used for estimation regarding the position of the mobile object. A plurality of position estimating sections that estimate the position of the mobile object using the information acquired by the plurality of position estimating sections and obtain reliability regarding the estimation results; , an acquisition unit that acquires the self-position of the mobile object according to the estimation result corresponding to the highest reliability.
With such a configuration, it is possible to obtain the self-position according to the estimation result with the highest reliability, and it is possible to realize more accurate self-position estimation. Furthermore, even if a plurality of moving objects are not present, the accuracy of self-position estimation for a single moving object can be improved. In addition, for example, if a certain position estimation unit estimates the position using GPS (Global Positioning System), even if the user is moving in an environment with many multipaths such as an environment with many obstacles, it is not reliable. By using the result of position estimation other than GPS by another position estimation unit depending on the situation, highly accurate self-position can be obtained. In addition, by acquiring the self-position according to the estimation result with the highest reliability among multiple estimation results, even if the accuracy of one of the estimation results decreases significantly, the other estimation results By acquiring the self-position according to the above, it is possible to avoid a decrease in the accuracy of the self-position.

Further, in the self-position estimating device according to the present invention, the sensor section has a plurality of sensors that acquire different information, and the plurality of position estimation sections each use a plurality of pieces of information acquired by the plurality of sensors. The position of the moving body may be estimated using the following method.
With such a configuration, a plurality of estimation results are obtained using information respectively obtained by a plurality of sensors. Therefore, if the reliability of the estimation result obtained using the information obtained by one of the sensors is low, the self-position may be determined according to the estimation result obtained using the information obtained by the other sensors. be obtained. Each sensor has advantages and disadvantages, but by acquiring self-positions in this way, they can complement each other's weaknesses, and it is possible to achieve more accurate self-position acquisition.

Further, in the self-position estimating device according to the present invention, the plurality of position estimating units may estimate the position of the mobile object using different methods using the same information acquired by the sensor unit.
With such a configuration, the self-position is acquired according to an estimation result obtained by a method with higher reliability. Accuracy may vary depending on the method of estimating the position, but by acquiring the self-position in this manner, it is possible to obtain the self-position with higher accuracy. For example, when estimating a position using a probabilistic method, the estimation result may differ even if a parameter or the like is slightly changed. In such a case, accuracy can be improved by acquiring the self-position using the estimation result of the estimation method with the highest reliability.

Further, in the self-position estimating device according to the present invention, the plurality of position estimating units estimate the position of the moving object, and the obtaining unit selects the estimated position corresponding to the highest reliability among the plurality of reliability. It may be selected as the self-position of the moving object.
With such a configuration, the estimated position with the highest degree of reliability can be set as the self-position of the mobile object, and the process of acquiring the self-position using the estimation result is only the selection, so the process of acquiring the self-position is a simple process.

Further, in the self-position estimating device according to the present invention, the plurality of position estimating sections estimate the movement amount to be added to the previous self-position, and the acquisition section estimates the movement amount to be added to the previous self-position. Among them, the self-position may be acquired by adding the movement amount corresponding to the highest reliability.
With this configuration, even if the estimation result used to obtain the self-position is changed from one estimation result to another depending on the reliability, the self-position after the change can be prevented from changing significantly. It can be avoided. Therefore, estimation results can be seamlessly switched, and smooth self-position estimation can be achieved.

According to the self-position estimating device according to the present invention, it is possible to acquire the self-position according to the estimation result with the highest degree of reliability, and it is possible to realize self-position estimation with higher accuracy.

A block diagram showing the configuration of a moving body according to an embodiment of the present invention Flowchart showing the operation of the self-position estimating device according to the embodiment A block diagram showing an example of another configuration of the mobile body according to the embodiment

DESCRIPTION OF THE PREFERRED EMBODIMENTS A self-position estimating device according to the present invention will be described below using embodiments. Note that in the following embodiments, components and steps denoted by the same reference numerals are the same or equivalent, and a repeated explanation may be omitted. The self-position estimating device according to the present embodiment acquires the self-position according to the estimation result with the highest reliability among the estimation results regarding the position of a moving object estimated by a plurality of position estimators.

FIG. 1 is a block diagram showing the configuration of a mobile object 1 according to this embodiment. The mobile object 1 according to the present embodiment acquires a self-position and moves using the self-position, and includes a self-position estimating device 2 for estimating the self-position of the mobile object 1, a moving mechanism 11, and a moving mechanism 11. A control unit 12 is provided. The self-position estimation device 2 also includes a sensor section 21 , a first position estimation section 22 , a second position estimation section 23 , and an acquisition section 24 . In this embodiment, a case will be mainly described in which the self-position estimating device 2 includes two position estimating sections 22 and 23, but this may not be the case. The self-position estimating device 2 may include three or more position estimating units, as described later.

The moving object 1 may be, for example, a moving object that travels on land, a flying object that flies in the air, or a watercraft that moves on water (such as a ship). Often, it may be an underwater vehicle (eg, a submersible, etc.) that moves underwater. In this embodiment, the case where the moving body 1 is a traveling body will be mainly described.

The moving body 1 that is a running body may be, for example, a trolley, a self-driving car, or a moving robot. The robot may be, for example, an entertainment robot, a monitoring robot, a transportation robot, a cleaning robot, or a robot that takes videos or still images. , or other robots.

The flying object may be, for example, a rotary wing aircraft, an airplane, an airship, or any other flying object. From the viewpoint of being movable to any position, it is preferable that the flying object is a rotary wing aircraft. The rotary wing aircraft may be, for example, a helicopter or a multicopter having three or more rotors. The multicopter may be, for example, a quadrotor with four rotors, or may have other numbers of rotors.

Furthermore, the mobile body 1 may be one that moves autonomously, for example. Note that the moving object 1 autonomously may mean that the moving object 1 moves to a destination based on its own judgment, rather than moving in response to an operation instruction received from a user or the like. The destination may be determined manually or automatically, for example. Furthermore, movement to the destination may or may not be performed along a movement route, for example. Furthermore, moving to the destination based on one's own judgment may mean, for example, moving to the destination by the moving object 1 making its own judgment regarding the direction of travel, movement, and stopping. Alternatively, for example, the moving body 1 may move so as not to collide with an obstacle.

Next, the configuration of the self-position estimating device 2 of the mobile object 1 will be explained.
The sensor unit 21 acquires information used for estimating the position of the mobile object 1. Note that the estimation regarding the position of the mobile object 1 may be an estimation regarding only the position of the mobile object 1, or may be an estimation regarding the position and posture (angle) of the mobile object 1. In this embodiment, the latter case will mainly be explained. Here, as shown in FIG. 1, a case in which the sensor unit 21 includes a first sensor 31 and a second sensor 32 will be mainly described, and other cases will be described later. The first and second sensors 31 and 32 are sensors that acquire different information. The sensors that acquire different information may be, for example, sensors that acquire information through different types of sensing (e.g., an image sensor and a ranging sensor), and can acquire information on different objects through the same type of sensing. Sensors (for example, a range sensor that measures a range in a certain direction (such as a front range) of the moving object 1, and a range sensor that measures a range in a different direction (such as a rear range), etc.) There may be. The first and second sensors 31 and 32 are not particularly limited, but each independently includes, for example, a sensor that acquires information around the moving body 1, an angular position sensor, an acceleration sensor, a GPS sensor, etc. It may be. The sensor that acquires information around the moving object 1 may be, for example, a distance measurement sensor, an image sensor, or the like. Note that it is preferable that the first and second sensors 31 and 32 acquire information synchronously. This is because it is preferable that the first and second position estimators 22 and 23 estimate positions at the same time.

Furthermore, when the estimation of the position of the moving body 1 is the estimation of the position and orientation of the moving body 1, the first sensor 31 and the second sensor 32 may have two or more sensors. . For example, the first sensor 31 and/or the second sensor 32 may include an acceleration sensor and a sensor used to detect the attitude of the moving body 1, and may include a GPS sensor and a sensor used to detect the attitude of the moving body 1. It may also have a sensor used for detection. The sensor used to detect the attitude of the moving body 1 may be, for example, a gyro sensor or a geomagnetic sensor (electronic compass). The gyro sensor may be one that detects the attitude of the moving body 1, or may be one that detects the angular velocity or angular acceleration of the moving body 1, for example.

A distance sensor is a sensor that measures the distance to surrounding objects in multiple directions, such as a laser sensor, an ultrasonic sensor, a distance sensor using microwaves, or a stereo image taken by a stereo camera. It may also be a distance sensor or the like. The laser sensor may be a laser range sensor (laser range scanner). Note that these distance measuring sensors are already known and their explanation will be omitted. In this embodiment, the case where the distance measurement sensor is a laser range sensor will mainly be described. Further, the distance measurement sensor may include one laser range sensor, or may include two or more laser range sensors. In the latter case, all directions may be covered by two or more laser range sensors. Furthermore, when the distance measurement sensor is an ultrasonic sensor or a distance sensor using microwaves, distances in multiple directions may be measured by rotating the distance measurement direction of the distance measurement sensor. Distances in a plurality of directions may be measured using a plurality of distance measuring sensors arranged in a plurality of distance sensors. Measuring distances in multiple directions may mean, for example, measuring distances in multiple directions at predetermined angular intervals within a predetermined angular range or all around (360 degrees). The angular intervals may be constant, such as 1 degree, 2 degree, or 5 degree intervals, for example. The information acquired by the ranging sensor may be, for example, distances to surrounding objects at each of a plurality of azimuth angles based on a certain direction of the moving body 1. By using this distance, it becomes possible to know what objects exist around the moving body 1 in the local coordinate system of the moving body 1.

The image sensor may be, for example, a CCD image sensor or a CMOS image sensor. Note that these image sensors are already well known, and their explanation will be omitted. The image sensor may include an optical system, such as a lens, for forming an image on the image sensor. Moreover, the image sensor may be monocular or binocular (stereo camera). An image sensor typically captures a moving image, that is, captures a series of image frames. Since a photographed image is acquired in order to acquire the self-position of the moving body 1, it is preferable that the frame rate is sufficiently large relative to the moving speed. For example, the frame rate may be approximately 30 fps.

The angular position sensor may be, for example, a rotary encoder. Note that the angular position sensor is already well known, and its explanation will be omitted. The angular position sensor may be used, for example, to detect the amount of rotation of a drive shaft such as a wheel of the moving body 1. The information acquired by the angular position sensor may be, for example, the amount of rotation or angle of the rotation axis.

The acceleration sensor measures the acceleration of the moving body 1. This acceleration sensor may be, for example, a two-axis acceleration sensor or a three-axis acceleration sensor. Note that the acceleration sensor is already well known, and its explanation will be omitted. If the moving direction is limited within a predetermined plane (for example, when the moving body 1 is a traveling body that moves on a floor surface), a two-axis acceleration sensor may be used. The information acquired by the acceleration sensor may be, for example, the acceleration of the moving body 1.

A GPS sensor is a sensor that receives radio waves from GPS satellites. Note that the GPS sensor is already well known, and its explanation will be omitted. The information acquired by the GPS sensor may be, for example, radio waves transmitted from a GPS satellite.

The information acquired by the gyro sensor may be, for example, the attitude (angle), angular velocity, or angular acceleration of the moving body 1. Further, the information acquired by the geomagnetic sensor may be, for example, the attitude of the mobile object 1. The attitude of the moving body 1 may be, for example, the azimuth of the moving body 1, or the azimuth and elevation/depression angle of the moving body 1.

The first and second position estimating units 22 and 23 estimate the position of the mobile object 1 using the information acquired by the first and second sensors 31 and 32 of the sensor unit 21, respectively. Obtain the confidence level regarding the estimation result. It is assumed that the first and second position estimating units 22 and 23 perform estimation and reliability regarding the positions of the first and second sensors 31 and 32, respectively, according to their types. For example, when the first sensor 31 is a ranging sensor, the first position estimating unit 22 estimates the position using information acquired by the ranging sensor and obtains reliability. When the sensor 31 is an image sensor, the first position estimating unit 22 estimates the position and obtains reliability using information acquired by the image sensor. Therefore, the first and second position estimating units 22 and 23 each independently perform, for example, information acquired by a sensor that acquires information around the mobile object 1, angle Information acquired by a position sensor, information acquired by an acceleration sensor, information acquired by a GPS sensor, etc. may be used to estimate the position and acquire reliability. The information acquired by the sensor that acquires information around the moving object 1 may be, for example, information acquired by a distance measurement sensor, information acquired by an image sensor, or the like. Hereinafter, the estimation result and reliability acquired by the first position estimation unit 22 using the sensing result of the first sensor 31 will be referred to as the first estimation result and the first reliability, and the second position The estimation result and reliability obtained by the estimation unit 23 using the sensing result of the second sensor 32 may be referred to as a second estimation result and a second reliability.

Here, estimation regarding a position using information acquired by a sensor and acquisition of reliability will be explained. Note that the estimation regarding the position of the mobile object 1 may be, for example, an estimation of the position of the mobile object 1 itself (for example, the position in the world coordinate system), and the self-position of the mobile object 1 according to the previous estimation result. It may also be an estimation of the amount of movement (difference in position) to be added to. In the latter case, when estimating the amount of movement and estimating the position of the mobile object 1 using the estimated amount of movement, for example, by estimating the position using an angular position sensor, an acceleration sensor, etc. The amount of movement itself may be used, and when the amount of movement is not estimated, such as when estimating the position using a GPS sensor, the amount of movement may be obtained by calculating the difference between the estimated self-positions. good.

First, estimation regarding a position using information acquired by a ranging sensor will be explained. The first position estimating unit 22 and/or the second position estimating unit 23 uses the distance measured by the ranging sensor and the environmental map indicating the position of obstacles in the moving area of the moving object 1 to 1 may be estimated. Strictly speaking, the measured distance is the measured distance in each direction. It is assumed that the environmental map is stored in a recording medium (not shown) that can be accessed by the first position estimating section 22 and/or the second position estimating section 23. The obstacle may be an object such as a wall or equipment whose distance is to be measured by the distance sensor. The estimated position is usually a position in the world coordinate system. Since such self-position estimation using distance measurement results is already known in SLAM (Simultaneous Localization and Mapping) using distance measurement results, detailed explanation thereof will be omitted. Note that the first position estimation unit 22 and/or the second position estimation unit 23 may or may not create an environmental map. In the former case, the first position estimating unit 22 and/or the second position estimating unit 23 may be similar to SLAM using distance measurement results; in the latter case, the first position estimating unit 22 and/or the second position estimating unit 23 The position estimating unit 22 and/or the second position estimating unit 23 may perform self-position estimation using an existing environmental map.

The reliability of the estimation result regarding the position using the distance measurement results indicates the degree to which the estimation result can be trusted, that is, the certainty. It may be obtained depending on the accuracy of the measurement result, it may be both, or it may be obtained by other methods. Therefore, the reliability is, for example, a match between the measurement results by the first sensor 31 and/or the second sensor 32 and the distance to the obstacle indicated by the environmental map when the mobile object 1 is present at the estimated position. The higher the degree of , the larger the value may be. In this case, the reliability may be calculated as follows.
Reliability = (ΣF(Li))/N1

Here, the function F(x) is a decreasing function that becomes 1 when x=0 and becomes smaller as x becomes a larger positive value. However, F(x)≧0. Furthermore, Li is the absolute value of the difference between the distance measurement result in the i-th distance measurement direction and the distance to the edge of the obstacle indicated by the environmental map. Further, N1 is the total number of i, that is, the total number of distance measurements, and the sum Σ is taken for all i. Since it is standardized by N1, the reliability is a real number greater than or equal to 0 and less than or equal to 1. Note that F(x) may be any function, and may be, for example, F(x)=exp(-α×x). α is a positive real number.

Moreover, since the distance measurement result by a laser etc. has a larger error due to the influence of reflection etc. as the distance becomes longer, the reliability may be, for example, a value that becomes smaller as the distance measurement result becomes larger. In this case, the reliability may be calculated as follows.
Reliability = (ΣG(Di))/N1

Here, the function G(x) is a decreasing function that becomes 1 when x=0 and becomes smaller as x becomes a larger positive value. However, G(x)≧0. Further, Di is the distance measurement result in the i-th distance measurement direction. Further, N1 and the total sum Σ are as described above. In this case as well, since it is standardized by N1, the reliability is a real number greater than or equal to 0 and less than or equal to 1. Further, G(x) may also be the same as the above function F(x). In addition, for example, the reliability increases as the degree of matching between the distance measurement result and the position of an obstacle shown on the environmental map increases, and the reliability decreases as the distance measurement result increases. good.

Note that even when the estimated position cannot be uniquely determined, the reliability may be set to a low value. For example, if the mobile object 1 exists in a monotonous environment such as a long corridor, the estimated position may not be uniquely determined using the distance measurement results. In such a case, the reliability may be set to a low value.

Next, estimation regarding the position using information acquired by the image sensor will be explained. The first position estimating unit 22 and/or the second position estimating unit 23 estimates the position of the mobile object 1 using the information acquired by the image sensor, thereby responding to the estimation result and the estimation result. It is also possible to obtain the reliability. In the estimation regarding the position, the estimation regarding the position may be performed using the feature points specified in the photographed image acquired by the image sensor and an environmental map showing the position of the feature points in the movement area of the mobile object 1. . It is assumed that the environmental map is stored in a recording medium (not shown) that can be accessed by the first position estimating section 22 and/or the second position estimating section 23. The estimated position is usually a position in the world coordinate system. Since such self-position estimation using captured images is already known as visual-SLAM, detailed explanation thereof will be omitted. Note that the first position estimation unit 22 and/or the second position estimation unit 23 may or may not create an environmental map. In the former case, the first position estimation unit 22 and/or the second position estimation unit 23 may be similar to visual-SLAM, and in the latter case, the first position estimation unit 22 and/or the second position estimation unit 23 may be similar to visual-SLAM. 22 and/or the second position estimation unit 23 may perform self-position estimation using an existing environmental map.

Note that the feature points specified in the captured image may be, for example, a predetermined number of FAST key points used in self-position estimation similar to ORB-SLAM, or may be SIFT key points, SURF key points, etc. good. In addition, the reliability indicates the degree to which the estimation result is reliable, that is, the certainty.For example, the reliability is obtained according to the degree of matching between the feature points specified in the captured image and the feature points included in the environmental map. It may be acquired according to the precision of the feature points specified in the photographed image, it may be both, or it may be acquired by other methods. Therefore, the reliability is calculated based on, for example, the feature descriptor (feature quantity) of the feature points specified in the captured image acquired by the image sensor and the feature points when the mobile object 1 is present at the estimated position. The value may be increased as the degree of similarity between the feature point and the feature descriptor included in the corresponding environmental map increases. In that case, the reliability may be calculated as follows.
Reliability = (ΣRi)/N2

Here, Ri is the degree of similarity between the feature descriptor of the i-th feature point identified in the photographed image and the feature descriptor of the feature point included in the environmental map corresponding to that feature point, and is from 0 to 1. It is preferable that the value be normalized to a value up to . As for the degree of similarity, it is assumed that the larger the value, the greater the degree of similarity. Further, if the environmental map does not include a feature point corresponding to the i-th feature point specified in the photographed image, the degree of similarity may be 0. Further, N2 is the total number of i, that is, the total number of feature points specified in the photographed image, and the sum Σ is taken for all i. Since it is standardized by N2, the reliability is a real number greater than or equal to 0 and less than or equal to 1. Note that when the feature descriptor is a binary vector (for example, in the case of ORB-SLAM), the degree of similarity may be calculated using Hamming distance. If the feature descriptor is not a binary vector (for example, in the case of SIFT keypoints or SURF keypoints), the degree of similarity may be calculated using cosine distance or Euclidean distance.

In addition, since feature points in a photographed image cannot be properly identified if it is overexposed and has blown out highlights, or if it is underexposed and has blown out shadows, the reliability is For example, the larger the evaluation value used in specifying the feature point, the larger the value may be. Note that it is assumed that feature point candidates with high evaluation values are specified as feature points. For example, in ORB-SLAM, a predetermined number of FAST key points are selected from the top of the Harris corner measure, so the evaluation value may be the Harris corner measure. That is, the reliability may be a representative value of a Harris corner measure regarding a plurality of feature points specified in a photographed image. The representative value may be, for example, an average value, a median value, a maximum value, or the like. For example, when specifying feature points (key points) by SIFT, in order to delete low-contrast key point candidates, the size of the extreme value of the key point candidate points is calculated, and the candidate points whose size is less than a threshold are Delete. Therefore, a representative value of the magnitude of the extreme value regarding the plurality of identified feature points may be used as the reliability. The same applies to other feature points. In addition, the reliability increases as the degree of similarity between the feature descriptors of feature points identified in the photographed image and the feature descriptors of feature points included in the environmental map corresponding to those feature points increases. In addition, the larger the evaluation value used in specifying the feature point, the larger the value may be.

For information on key points, feature points, feature descriptors, etc., please refer to the following documents.
Reference 1: Edward Rosten, Tom Drummond, "Machine learning for high-speed corner detection", in 9th European Conference on Computer Vision, vol. 1, 2006, pp. 430-443.
Reference 2: C. Harris, M. Stephens, "A Combined Corner and Edge Detector", In Proceedings of the 4th Alvey Vision Conference (1988), pp. 147-151.
Reference 3: Ethan Rublee, Vincent Rabaud, Kurt Konolige, Gary R. Bradski, "ORB: An efficient alternative to SIFT or SURF", ICCV 2011, pp.2564-2571.
Reference 4: D. Lowe, "Distinctive image features from scale-invariant keypoints", Int. Journal of Computer Vision (2004), 60(2), pp.91-110.
Reference 5: H. Bay, A. Ess, T. Tuytelaars, L. Van Gool, "SURF: Speeded Up Robust Features", Computer Vision and Image Understanding (CVIU) (2008), 110(3), pp.346- 359.
Reference 6: R. Mur-Artal, JM Montiel, JD Tardos, "ORB-SLAM: A Versatile and Accurate Monocular SLAM System", IEEE Transactions on Robotics 31 (2015), pp.1147-1163.

Next, estimation regarding position using information acquired by an angular position sensor such as an encoder will be explained. The amount of rotation of a wheel or the like can be determined based on the amount of rotation and angle acquired by the angular position sensor. Therefore, the first position estimating unit 22 and/or the second position estimating unit 23 may use the information obtained by the angular position sensor to obtain the estimation result regarding the position. Here, an example of a two-wheeled moving object 1 will be specifically described.

When the rotation amounts of the left and right wheels acquired by the angular position sensor are respectively Δθ _L and Δθ _R , and the wheel diameter is R, the movement amounts Δd _L and Δd _R of the left and right wheels are as follows, respectively.
Δd _L =RΔθ _L
Δd _R =RΔθ _R

Further, if the distance between the left and right wheels is L, the moving amount Δd and the turning amount Δθ of the moving body 1 can be calculated as shown in the following equation.
Δd=(Δd _R +Δd _L )/2
Δθ=(Δd _R −Δd _L )/L

When the position of the moving body 1 is (X, Y, Θ), the relative movement amount (ΔX, ΔY, ΔΘ) of the moving body 1 is expressed by the following equation. Note that (X, Y) may be coordinate values in a two-dimensional XY coordinate system, and Θ may be an angle with respect to the X axis in the two-dimensional coordinate system. Further, the position (X, Y, Θ) of the moving body 1 may be a position in the world coordinate system.
(ΔX, ΔY, ΔΘ) = (Δd×cosΘ, Δd×sinΘ, Δθ)

Therefore, by adding the relative movement amount (ΔX, ΔY, ΔΘ) to the previous (time: t-1) self-position (X _t-1 , Y _{t-1 , Θ t-1 ), the current (time: t-1) self-position (X t-1 , Y t-1} , Θ _t-1 ) t)'s self-position (X _t , Y _t , Θ _t ) can be obtained.
(X _t , Y _t , Θ _t )=(X _t-1 , Y _t-1 , Θ _t-1 )+(ΔX, ΔY, ΔΘ)

Note that the relative movement amount in the above equation is the movement amount between t-1 and t. The previous self-position may be estimated using information acquired by an angular position sensor, for example, or may be a self-position acquired by the acquisition unit 24, as described later. In addition, although the case where the moving body 1 is of a two-wheeled type has been specifically explained here, even if the moving body 1 is of another type, the angular position sensor such as an encoder can be used in the same manner. It goes without saying that the position of the mobile object 1 can be estimated using the acquired information.

The reliability of the estimation result regarding the position using the information acquired by the angular position sensor is such that the estimation result can be trusted. In position estimation using an angular position sensor, reliability is reduced due to the influence of the road surface, for example. Specifically, on wet driving surfaces, gravel, sand, grass, and other driving surfaces, the wheels tend to slip, and the reliability of position estimation using the angular position sensor decreases. Therefore, the first position estimating unit 22 and/or the second position estimating unit 23 determines the state of the running road surface corresponding to the current position of the mobile object 1 using a map showing the area of the running road surface where the wheels are slippery. (a state indicating whether the wheels are slippery) may be specified, and the reliability may be obtained according to the identification result. For example, when the moving body 1 is on a slippery road surface, low reliability may be obtained, and when the moving object 1 is on a non-slip road surface, high reliability may be obtained. In this case, for example, reliability may be acquired depending on the degree of slipperiness of the road surface. Specifically, the reliability may be obtained such that the slippery the road surface on which the moving object 1 is present becomes, the lower the reliability is obtained. Moreover, the reliability in this case is usually the reliability regarding the above-mentioned relative movement amounts (ΔX, ΔY, ΔΘ). Further, whether or not the mobile object 1 is present on a slippery road surface may be determined using, for example, an estimated position estimated using an angular position sensor, or using a self-position acquired by the acquisition unit 24. It may be determined that In the latter case, the determination may be made using, for example, the previous self-position.

Note that when the current self-position is estimated using only the information acquired by the angular position sensor, low reliability may be obtained when there is a slippery road surface on the route to the current self-position. high reliability may be obtained when the route is free of slippery driving surfaces. Further, in this case, for example, reliability may be acquired depending on the degree of slippery road surface included in the route. Specifically, the more slippery the road surface included in the route, the lower the reliability may be acquired.

Furthermore, when position estimation is performed using only information acquired by an angular position sensor such as an encoder, it is known that the longer the moving distance, the lower the reliability becomes. This is because errors accumulate. Therefore, for example, when the moving object 1 is present at a location whose position has been determined in advance, and when the current position estimated using the angular position sensor is to be adjusted to the predetermined position, The reliability may be acquired depending on the moving distance after alignment. In that case, the longer the moving distance, the lower the reliability will be acquired.

Next, estimation regarding the position using information acquired by the acceleration sensor will be explained. The acceleration sensor acquires the acceleration of the moving body 1. Then, the first position estimating section 22 and/or the second position estimating section 23 can obtain the amount of movement (distance of movement) by performing second-order integration on the obtained acceleration. By sequentially adding the travel distances thus obtained, the current position can be estimated. In this case, the longer the travel distance, the lower the reliability. Furthermore, in this case, since only the moving distance can be obtained, the initial position must be input. Therefore, for example, the reliability may be acquired according to the moving distance of the mobile body 1 from the position where the initial position is set. Furthermore, when positioning is performed in addition to the initial position, the reliability may be acquired according to the distance traveled by the mobile body 1 after the positioning. In this case, the longer the moving distance from the initial position or the moving distance after alignment becomes, the lower the reliability will be obtained.

In addition, in estimating the position using distance measurement results, photographed images, rotation amount of wheels, acceleration, etc., state space models such as particle filters and Kalman filters may be used, and the Nelder-Mead method (Nelder-Mead method) may be used. ; downhill simplex method) may be used, or other methods may be used. When a particle filter is used, for example, a representative value of the likelihood of each particle (for example, an average value, a median value, a maximum value, etc.) may be acquired as the reliability according to the estimation result.

Next, position estimation using information acquired by the GPS sensor will be explained. The GPS sensor receives radio waves from multiple GPS satellites. Then, the first position estimating section 22 and/or the second position estimating section 23 may perform position estimation using the received radio waves from the plurality of GPS satellites. Note that GPS position estimation is already well known, and detailed explanation thereof will be omitted. The reliability according to the GPS position estimation may be obtained using, for example, the degree of multipath detected in received radio waves. In this case, for example, the more multipaths there are, the lower the reliability may be obtained. For example, the first position estimating unit 22 and/or the second position estimating unit 23 may determine the current position of the mobile object 1 using a map indicating whether radio waves from GPS satellites can be appropriately received. The reception state of the corresponding radio wave (the state indicating whether or not the radio wave can be properly received) may be specified, and the reliability may be obtained according to the identification result. For example, if the mobile object 1 is located in an area where radio waves from GPS satellites can be properly received, high reliability will be obtained, and if the mobile object 1 is located in an area where radio waves are poorly received (for example, an area with many multipaths or indoors), high reliability will be obtained. low reliability may be obtained. In this case, the reliability may be acquired depending on the radio wave reception situation, for example. Specifically, the better the radio waves from the GPS satellite corresponding to the area where the mobile object 1 is present can be received, the higher the reliability may be obtained.

Note that in estimating the position using information acquired by a distance sensor, an image sensor, and an angular position sensor, it is usually possible to estimate the position and orientation of the mobile object 1. On the other hand, in estimating the position using information acquired by the GPS sensor and the acceleration sensor, only the estimating the position of the mobile object 1 is usually performed. Therefore, when estimation regarding the attitude of the mobile object 1 is also required, estimation regarding the attitude may be performed using information acquired by a sensor used for detecting the attitude, such as a gyro sensor or a geomagnetic sensor. For example, when information is acquired by a gyro sensor or geomagnetic sensor that detects the attitude of the moving body 1, the acquired information itself may be used as the attitude of the moving body 1. For example, when the angular velocity and angular acceleration of the moving body 1 are acquired by a gyro sensor that is an angular velocity sensor or an angular acceleration sensor, the angle (amount of change in attitude) and angular acceleration obtained by integrating the angular velocity The attitude of the moving body 1 may be obtained by integrating the angle (amount of change in attitude) obtained by second-order integration of .

When acquiring the attitude of the mobile body 1 using a gyro sensor or a geomagnetic sensor, reliability may or may not be acquired for the acquired attitude. In the former case, it is preferable that the final reliability (position and orientation reliability) is obtained based on the position reliability and the orientation reliability, and in the latter case, The position reliability may be used as the position and orientation reliability. When the reliability of the position and orientation is generated based on the reliability of the position and the reliability of the orientation of the moving object 1, the reliability of the position and orientation is determined such that, for example, the higher the reliability of the position, the higher the reliability of the orientation. The larger the posture reliability, the larger the posture may be.

Here, a method for obtaining posture reliability will be briefly described. When acquiring an attitude (angle) using a gyro sensor, it is usually necessary to set an initial angle. Therefore, for example, the reliability may be acquired according to the cumulative result of the amount of change in angle after setting the initial angle. In this case, for example, the reliability may be lowered as the cumulative amount of change in angle from the initial angle setting increases. Moreover, a geomagnetic sensor usually detects a weak magnetic field to detect an orientation. Therefore, when the intensity of the magnetic field detected by the geomagnetic sensor is large, it is considered that a magnet exists near the geomagnetic sensor, and the error in orientation may increase due to its influence. Therefore, for example, the reliability may be acquired depending on the magnitude of the strength of the magnetic field detected by the geomagnetic sensor. In this case, for example, the higher the intensity of the magnetic field detected by the geomagnetic sensor, the lower the reliability may be acquired.

Further, the reliability acquired by the first and second position estimating units 22 and 23, that is, the first and second reliability, may be standardized so that they can be appropriately compared. suitable. Specifically, if the first and second reliability values are approximately the same, it is preferable that the first and second estimated positions are each standardized so that they have approximately the same reliability. It is. The same applies when three or more reliability levels are obtained by three or more position estimators.

The acquisition unit 24 acquires the self-position of the mobile body 1 according to the estimation result corresponding to the higher reliability of the first and second reliability. For example, when the first and second estimation results are estimated positions, the acquisition unit 24 determines the estimated position of the mobile object 1 corresponding to the higher reliability of the first and second reliability. You may also select it as your own location. Further, for example, when the first and second estimation results are estimated movement amounts, the acquisition unit 24 uses the higher reliability of the first and second reliability for the previous self-position. The self-position may be acquired by adding the movement amount corresponding to the degree. In this way, by adding the movement amount with higher reliability to the previous self-position, changes in the self-position can be made smoother. For example, when selecting the estimated position with higher reliability as described above, the self-position may change significantly when switching from one estimated position to the other. On the other hand, if the estimated movement amount is added to the previous self-position, the self-position will not change significantly even if one movement amount is switched to the other movement amount. Note that when acquiring the self-position using the amount of movement, it is preferable that the acquisition unit 24 retains the previously acquired self-position. In addition, as mentioned above, when selecting the estimated position with higher reliability, the self-position may change significantly, so if the amount of change exceeds a predetermined threshold, the The unit 24 may acquire the self-position so as to smoothly connect the two. Further, when the first and second reliability levels are the same, the acquisition unit 24 may acquire the self-position of the mobile object 1 using one of the randomly selected estimation results, or the previous The self-position of the mobile body 1 may be acquired using the estimation result on the same side as the selection.

Next, the configuration of the mobile body 1 other than the self-position estimating device 2 will be explained.
The moving mechanism 11 moves the moving body 1. The moving mechanism 11 may or may not be able to move the moving body 1 in all directions, for example. Being able to move in all directions means being able to move in any direction. The movement mechanism 11 may include, for example, a running part (for example, wheels, endless track, etc.) and a driving means (for example, a motor, an engine, etc.) for driving the running part, and allows the flying object to fly. It may have a flight section, or it may have a navigation section for navigating a surface vehicle or an underwater vehicle. The flight section may include, for example, a rotor and a drive means for driving the rotor. Further, the navigation section may include, for example, a screw and a drive means for driving the screw. Furthermore, the moving mechanism 11 may be equipped with a mechanism that can obtain the amount of rotation of a wheel, for example, an angular position sensor such as an encoder. In addition, when the moving mechanism 11 is capable of moving the moving body 1 in all directions, the traveling portion thereof may be omnidirectional moving wheels (for example, an omni wheel, a mecanum wheel, etc.). As this moving mechanism 11, a known mechanism can be used, so a detailed explanation thereof will be omitted.

The movement control unit 12 controls the movement of the mobile body 1 by controlling the movement mechanism 11 using the self-position acquired by the acquisition unit 24. The movement control may include control of the direction of movement of the moving body 1, start/stop of movement, and the like. For example, if a movement route has been set, the movement control unit 12 may control the movement mechanism 11 so that the moving body 1 moves along the movement route. More specifically, the movement control unit 12 may control the movement mechanism 11 so that the acquired self-position is along the movement route. Furthermore, the movement control unit 12 may control movement using a map. Since the control of the movement mechanism 11 by the movement control unit 12 is well known, detailed explanation thereof will be omitted.

Next, the operation of the self-position estimating device 2 will be explained using the flowchart of FIG. 2.
(Step S101) The acquisition unit 24 determines whether to acquire the self-position. If the self-position is to be acquired, the process proceeds to step S102; otherwise, the process of step S101 is repeated until it is determined that the self-position is to be acquired. Note that the acquisition unit 24 may, for example, periodically determine to acquire the self-position.

(Step S102) The first and second sensors 31 and 32 each acquire information used for estimation regarding the position.

(Step S103) The first position estimating unit 22 uses the information acquired by the first sensor 31 to acquire a first estimation result regarding the position of the mobile object 1, and according to the first estimation result. Obtain the confidence level.

(Step S104) The second position estimating unit 23 uses the information acquired by the second sensor 32 to acquire a second estimation result regarding the position of the mobile object 1, and according to the second estimation result. Obtain the confidence level. Note that it is preferable that the first and second estimation results are acquired synchronously.

(Step S105) The acquisition unit 24 acquires the self-position using the estimation result corresponding to the higher reliability of the first and second reliability. Then, the process returns to step S101.

Although the flowchart in FIG. 2 only shows the process of acquiring the self-position, in the moving body 1, the movement control unit 12 uses the acquired self-position to control the movement. The movement mechanism 11 may perform a movement process according to the above. Moreover, the order of processing in the flowchart of FIG. 2 is an example, and the order of each step may be changed as long as the same result can be obtained. Further, in the flowchart of FIG. 2, the process is terminated by turning off the power or by an interrupt to terminate the process.

Next, the operation of the mobile body 1 according to this embodiment will be explained using a simple specific example. In this specific example, it is assumed that the moving body 1 is used as a transport robot in a factory. Further, it is assumed that the first sensor 31 is an encoder that obtains the rotation amount of the wheel, and the second sensor 32 is a laser range sensor. Further, it is assumed that the first position estimating unit 22 obtains the reliability using a map showing areas where wheels are likely to slip. Further, the first and second position estimating units 22 and 23 estimate the amount of movement to be added to the previous self-position, and the acquisition unit 24 adds the one with higher reliability to the previous self-position. It is assumed that the current self-position is obtained by adding the amount of movement.

First, it is assumed that the moving object 1 is moving in an area where there is a lot of pedestrian traffic, but where the wheels are not easily slippery. In such a situation, the amount of rotation of the wheel is measured by the first sensor 31, the amount of movement that is the first estimation result is obtained according to the measurement result, and the first reliability regarding the amount of movement is determined. Suppose that it has been acquired. Further, the distance to the obstacle is measured by the second sensor 32, and a movement amount that is a second estimation result is obtained according to the measurement result, and a second reliability regarding the movement amount is obtained. (Steps S101 to S104). Note that in this case, the first reliability is a high value. This is because the moving body 1 does not exist in an area where the wheels are likely to slip. On the other hand, the second reliability is a low value. This is because there are many obstacles (humans) that are not included in the environmental map. Therefore, the acquisition unit 24 selects the amount of movement that is the first estimation result that corresponds to the first reliability that is the higher value, and adds the selected amount of movement to the previous self-position. The self-position is acquired and passed to the movement control unit 12 (step S105). Then, movement is controlled according to the self-position.

Assume that the moving body 1 then moves to an area where there is no pedestrian traffic and where the wheels are likely to slip. In such a situation, information is acquired by the first and second sensors 31 and 32, and the first and second position estimators 22 and 23 combine the first and second estimation results with the first and second Assume that reliability of 2 is obtained for each (steps S101 to S104). Note that in this case, the first reliability will be a low value. This is because the moving body 1 exists in an area where the wheels are likely to slip. On the other hand, the second reliability is a high value. This is because there are no obstacles (humans) that are not included in the environmental map. Therefore, the acquisition unit 24 selects the amount of movement that is the second estimation result that corresponds to the second reliability that is a higher value, and adds the selected amount of movement to the previous self-position. The self-position is acquired and passed to the movement control unit 12 (step S105). Then, movement is controlled according to the self-position.

As described above, according to the mobile object 1 according to the present embodiment, the self-position of the mobile object 1 is acquired using the estimation result corresponding to the higher reliability of the first and second reliability. Therefore, more accurate self-position acquisition can be achieved. Furthermore, when the first and second position estimation units 22 and 23 estimate the amount of movement, the current self-position can be obtained by adding the estimated movement amount to the previous self-position. Even if the amount of movement used for acquiring the self-position is changed according to the reliability, it is possible to avoid a large change in the self-position, and it is possible to achieve smooth acquisition of the self-position.

Note that, in this embodiment, the case where the sensor section 21 includes the first and second sensors 31 and 32 has been mainly described, but this may not be the case. As shown in FIG. 3, the sensor section 21 may include only one sensor 33. In this case, the first and second position estimation units 22 and 23 use the same information acquired by the sensor 33 of the sensor unit 21 to estimate the position of the mobile object 1 using different methods. Good too. For example, when position estimation is performed using a state space model such as a particle filter or a Kalman filter, or the Nelder-Mead method, the first and second position estimation units 22 and 23 use different models or methods to estimate the position. An estimate may be made. Specifically, when the sensor 33 is a ranging sensor, the first position estimating unit 22 uses a particle filter to estimate the position, and the second position estimating unit 23 uses a Kalman filter to estimate the position. You may make an estimate regarding. In addition, when the first and second position estimators 22 and 23 perform position estimation using the same model or the same method, the first and second position estimators 22 and 23 use different settings (parameters). Estimation regarding position may be performed using a model or method. For example, when the first and second position estimators 22 and 23 use particle filters, the numbers of particles may be different from each other. Further, for example, when the first and second position estimating units 22 and 23 use the Nelder-Mead method, the number of vertices of the simplex may be different from each other, and the initial positions of the vertices of the simplex may be different from each other. good. Note that even when the sensor unit 21 has two sensors that acquire the same information, the first and second position estimation units 22 and 23 similarly determine the position of the mobile object 1 using different methods. You may make an estimate regarding.

Furthermore, it is considered difficult for the first and second position estimators 22 and 23 to perform position estimation using different methods using information acquired by the GPS sensor. It is also considered difficult to obtain different information depending on the sensor using two GPS sensors. Therefore, when it has a GPS sensor, the sensor section 21 also has other types of sensors other than the GPS sensor, and one of the first and second position estimating sections 22 and 23 is , it is preferable that the position is estimated using information acquired by a GPS sensor, and the other position estimator estimates the position using information acquired by a sensor other than the GPS sensor.

Further, in the present embodiment, the first position estimating section 22 and/or the second position estimating section 23 may estimate the self-position using a plurality of pieces of information from which the self-position can be estimated. For example, the first position estimation unit 22 and/or the second position estimation unit 23 uses the distance measurement result obtained by the distance measurement sensor and the rotation amount and angle of the drive shaft obtained by the angular position sensor. The self-position may also be estimated. In this case, for example, a state space model such as a particle filter may be used. Further, in this case, the sensor section 21 may include, for example, two or more types of sensors, such as a distance measurement sensor and an angular position sensor. Moreover, in this case, reliability according to the likelihood of a particle filter or the like may be acquired.

Furthermore, in this embodiment, as described above, the self-position estimating device 2 may include three or more position estimating units. In this case, the acquisition unit 14 acquires the self-position of the mobile object 1 according to the estimation result corresponding to the highest reliability among the multiple reliabilities respectively acquired by the multiple position estimation units. good. Further, in this case, the sensor section 21 may have the same number of sensors as the position estimating section. Then, the plurality of position estimators may estimate the position of the mobile object 1 using the plurality of pieces of information acquired by the plurality of sensors. Further, as described above, when the sensor unit 21 has one sensor 33, the plurality of position estimating units use the same information acquired by the sensor unit 21 to estimate the location of the mobile object by different methods. An estimate may be made regarding the location of. When three or more position estimating units estimate the position of the mobile body 1, the acquisition unit 14 selects the estimated position corresponding to the highest reliability among the plurality of reliability levels as the self-position of the mobile body 1. You may also select it as In addition, when three or more position estimating units estimate the amount of movement to be added to the previous self-position, the acquisition unit 14 adds the previous self-position to the highest reliability among the plurality of reliabilities. The self-position may be acquired by adding the amount of movement corresponding to . Furthermore, in this case, the sensor section 21 may have a smaller number of sensors than the position estimating section. For example, when the self-position estimating device 2 includes N position estimating units, the sensor unit 21 may include K sensors that acquire different information. Note that N is an integer of 3 or more, and K is an integer that satisfies 1<K<N. Then, the K position estimation units estimate the position of the mobile object 1 using the plurality of pieces of information acquired by the K sensors, and the remaining (NK) position estimation units The position of the mobile object 1 may be estimated using the same information as the two position estimators. In this case, the two or more position estimators that estimate the position of the mobile body 1 using information from a certain sensor may estimate the position of the mobile body 1 using different methods.

In addition, if the self-position acquired by the self-position estimating device 2 is also used in another device, etc., the self-position estimating device 2 has an output section that outputs the self-position acquired by the acquiring section 24. may have. The output may be transmitted to a predetermined device via a communication line, or may be stored on a recording medium, for example. Note that the output unit may or may not include a device that performs output (eg, a communication device, etc.). Furthermore, the output section may be realized by hardware or by software such as a driver that drives these devices.

Furthermore, in the above embodiments, each process or each function may be realized by being centrally processed by a single device or a single system, or may be realized by being distributedly processed by multiple devices or multiple systems. This may be realized by

In addition, in the above embodiment, the information exchange performed between each component is performed by one component, for example, when the two components that exchange the information are physically different. This may be done by outputting information and receiving the information by another component, or by one component if the two components passing that information are physically the same. This may be performed by moving from a phase of processing corresponding to the component to a phase of processing corresponding to the other component.

In the above embodiments, information related to processing executed by each component, for example, information accepted, acquired, selected, generated, transmitted, or received by each component. Information such as threshold values, formulas, addresses, etc. used by each component in processing may be held temporarily or for a long period of time in a recording medium (not shown), even if not specified in the above description. Further, the information may be stored in the recording medium (not shown) by each component or by a storage unit (not shown). Further, each component or a reading unit (not shown) may read information from the recording medium (not shown).

In addition, in the above-described embodiment, if the information used in each component, for example, information such as threshold values, addresses, various setting values, etc. used by each component in processing, may be changed by the user, the above-mentioned Even if it is not specified in the description, the user may or may not be able to change the information as appropriate. If the information can be changed by the user, the change is realized by, for example, a reception unit (not shown) that receives change instructions from the user, and a change unit (not shown) that changes the information in accordance with the change instruction. It's okay. The acceptance of the change instruction by the reception unit (not shown) may be, for example, acceptance from an input device, information transmitted via a communication line, or information read from a predetermined recording medium. .

Further, in the above embodiment, when two or more components included in the self-position estimating device 2 have a communication device, an input device, etc., the two or more components physically have a single device. or may have separate devices.

Furthermore, in the embodiments described above, each component may be configured by dedicated hardware, or components that can be realized by software may be realized by executing a program. For example, each component can be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory. At the time of execution, the program execution section may execute the program while accessing the storage section or recording medium. Further, the program may be executed by being downloaded from a server or the like, or may be executed by reading a program recorded on a predetermined recording medium (for example, an optical disk, a magnetic disk, a semiconductor memory, etc.). good. Further, this program may be used as a program constituting a program product. Further, the number of computers that execute the program may be one or more. That is, centralized processing or distributed processing may be performed.

Furthermore, it goes without saying that the present invention is not limited to the above-described embodiments, and that various modifications can be made, and these are also included within the scope of the present invention.

As described above, the self-position estimating device according to the present invention has the effect of being able to acquire a highly accurate self-position, and is useful, for example, as a self-position estimating device used in a mobile object that moves autonomously.

1 Mobile object, 2 Self-position estimating device, 21 Sensor section, 22 First position estimating section, 23 Second position estimating section, 24 Acquisition section, 31 First sensor, 32 Second sensor

Claims (3)

A self-position estimating device that estimates the self-position of a moving object,
a sensor unit that acquires information used for estimation regarding the position of the mobile object;
a plurality of position estimating units that use the same information acquired by the sensor unit to estimate the position of the mobile object using different methods , and obtain reliability regarding the estimation results;
A self-position estimating device comprising: an acquisition unit that acquires the self-position of the mobile body according to the estimation result corresponding to the highest reliability among the plurality of reliabilities respectively acquired by the plurality of position estimation units. . The plurality of position estimating units estimate the position of the mobile object,
The self-position estimating device according to claim 1 , wherein the acquisition unit selects the estimated position corresponding to the highest reliability among the plurality of reliability as the self-position of the mobile body. The plurality of position estimation units estimate a movement amount to be added to the previous self-position,
The self-position estimating device according to claim 1 , wherein the acquisition unit acquires the self-position by adding a movement amount corresponding to the highest reliability among the plurality of reliability to the previous self-position.

Images