JP2011170843A - Route generation device and method, and mobile device with the route generation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow high-speed movement of a mobile device, by finding out a moving route along which the mobile device can travel at higher speed, in searching of a movement route. <P>SOLUTION: The route generation device 10 for generating a moving route for the mobile device which autonomously moves includes a storage device 3 which stores a plurality of stopover positions set between a movement start position and a target arrival position to provide an indication for generation of the moving route; a position detector 5 which detects a current position of the mobile device; and a route search device 9, which generates during movement of the mobile device a moving route for the mobile device to move to a stopover position to be passed through next among the plurality of stopover positions. The route search device 9 generates a plurality of moving route candidates, selects, from the plurality of moving route candidates, a moving route candidate with the highest high-speed travel possibility by calculating high-speed travel possibility for each of the moving route candidates and mutually comparing the high-speed travel possibilities, and sets the selected moving route candidate as the moving route to be actually used. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a route generation device and method for generating a movement route of a mobile device that moves autonomously, and a mobile device including the route generation device.

A mobile device that moves autonomously is a paved road, an unpaved road (off-road), or a robot or vehicle that moves autonomously indoors.
In order for the mobile device to move and travel autonomously, the mobile device is controlled so as to move along a movement route set on a known map (for example, Patent Document 1 below).

  In order to perform this control, the mobile device includes, for example, a storage device, an inner world information detection device, an obstacle detection device, a route search device, and a control device.

  The storage device stores a map (map information) of an area including the movement range of the moving device and a set route (route information) from the movement start position to the target arrival position. The map information is, for example, a coordinate system using two-dimensional coordinates. The route information is a plurality of via positions set between the movement start position and the target arrival position. These via positions serve as indexes for generating and setting the movement route described later. Note that these via positions may be coordinates in the coordinate system.

  The inner world information detection device detects information about the moving device itself, such as the current position of the moving device, the moving direction of the moving device, and the moving speed.

  The obstacle detection device detects the presence and the position / range of an obstacle such as a stopped vehicle, uneven surface, or wall around the mobile device (for example, in a predetermined area on the traveling side of the mobile device).

  The route search device searches and generates a movement route for the next route position based on the current position of the mobile device, the position / range of the obstacle, and the route information, for example, in a predetermined cycle. That is, the route search device searches and generates a movement route for moving from the current position to the next via position while avoiding the detected obstacle. The generated travel route is set and stored in a storage device for use in travel control of the mobile device.

  Based on the current position of the moving device, the control device controls the moving direction and speed of the moving device so that the moving device travels along the movement path set as described above.

JP-A-7-248820 JP 2009-110154 A

The International Journal Robots Research Vol. 22, no. 7-8, July-August 2003, pp. 583-601 Journal of Field Robots 23 (5), 311-331 (2006), 2006 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com), DOI: 10.1002 / rob. 2011

  If a moving route that can move at a higher speed can be found in the search for the moving route, the moving device can move at a higher speed.

  Accordingly, an object of the present invention is to enable a mobile device to move at high speed by finding a moving route that can move at a higher speed in the search for a moving route.

In order to achieve the above object, according to the present invention, there is provided a route generation device that generates a movement route of a mobile device that moves autonomously.
A storage device that stores a plurality of via positions that are set between the movement start position and the target arrival position and serve as an index for generating the movement path;
A position detecting device for detecting a current position of the moving device;
A route search device that searches and generates a movement route for the mobile device to go to the next passing position among the plurality of via positions based on the current position during movement of the mobile device, and
The route search device
While generating a plurality of moving route candidates, calculating the high-speed driving possibility for each of these moving route candidates,
By comparing these high-speed driving possibilities with each other, a moving route candidate having the highest high-speed driving possibility is selected from the plurality of moving route candidates.
There is provided a route generation device characterized in that the selected movement route candidate is set as a movement route to be actually used.

According to a preferred embodiment of the present invention, the high-speed driving possibility is based on a route length of a moving route candidate, a minimum curvature radius of the moving route candidate, and a difference in height of a road surface area along the moving route candidate. Including at least one of the road surface goodness and the proximity of the currently used moving route and the moving route candidate,
The route search device calculates the high-speed driving possibility according to the value of each element.

  Preferably, the high-speed driving possibility is higher as the value of each element is larger.

  According to the present invention, there is provided a moving device comprising: the above-described route generating device; and a control device that controls the moving device to move along the moving route generated by the route generating device. The

According to another embodiment of the present invention, the route search device is configured to detect a local position in the vicinity of the current position based on a current position of the mobile device and an obstacle detected by an obstacle detection device provided in the mobile device. Generate a map,
In the local map, the route search device extends a large number of linear search lines from the current position to an undetected area or an obstacle where no obstacle is detected, thereby causing obstacles on both sides. A travelable part of the mobile device sandwiched between objects is specified, the plurality of travel route candidates reaching the travelable part are generated, and a high speed travel possibility is calculated for each of the travel route candidates.

In this case, preferably, the route search device is a local map,
(A) When there are obstacles on both the left and right sides in the direction toward the passing position that passes next, extract the travelable area of the mobile device that exists between the left and right obstacles;
(B) generating a plurality of inscribed circles located in the travelable area and in contact with the boundaries of the travelable area;
(C) A road line is generated based on a line passing through the center of each inscribed circle,
(D) A plurality of linear search lines extending radially from the current position, and one end search line among a plurality of search lines that are continuously adjacent to each other and reach the undetected region without colliding with an obstacle. The search line adjacent to the left obstacle identifies the collision point as the first collision point, and the search line adjacent to the other search line of the plurality of search lines collides with the right obstacle. The collision point to be identified as the second collision point,
(E) The first collision point and the second collision point are compared, and the one closer to the current position among the two collision points is set as the target collision point,
(F) A line segment extending from the target collision point, orthogonal to the road line, and extending to the obstacle on the opposite side of the target collision point is generated as the travelable location,
(G) The plurality of movement route candidates that reach the line segment are generated, and the high-speed travel possibility is calculated for each of the movement route candidates.

Instead, the route search device in a local map,
(A) A plurality of linear search lines extending radially from the current position, and one end search line among a plurality of search lines that are adjacent to each other and reach the undetected region without colliding with an obstacle; An adjacent search line identifies a collision point that collides with an obstacle on the left as a first collision point, and a search line adjacent to the other search line of the plurality of search lines collides with an obstacle on the right side. Identify the collision point as the second collision point,
(B) The first collision point and the second collision point are compared, and the one closer to the current position among the two collision points is set as the target collision point,
(C) In the obstacle on the side opposite to the target collision point, a point where the distance from the target collision point is the shortest is set, and a line segment extending from the point to the target collision point is generated as the travelable location. ,
(D) The plurality of travel route candidates that reach the line segment may be generated, and the high-speed travel possibility may be calculated for each of the travel route candidates.

According to yet another embodiment of the present invention, the moving device is a vehicle that travels on a road surface by traveling wheels that are rotationally driven,
The vehicle is equipped with a vibration sensor that measures the vibration of the vehicle when traveling,
Based on the vibration measured by the vibration sensor, a road surface determination device for determining whether the road surface on which the vehicle is currently traveling is an unpaved road surface or a paved road surface is provided,
The high-speed driving possibility includes not only the road surface goodness as an element thereof but also one or both of the route length and the minimum curvature radius,
The route search device calculates the value of each element by converting it to a standard value, calculates the high-speed driving possibility reflecting the standard value of each element,
The route search device, when it is determined that the road surface is an unpaved road surface for the same moving route candidate, than the case where the road surface is determined to be a paved road surface and the minimum curvature. Calculate a smaller radius standard value.

Moreover, according to the present invention, there is provided a route generation method for generating a movement route of a mobile device that moves autonomously,
A storage device that stores a plurality of via positions that are set between a movement start position and a target arrival position and serves as an index for generating the movement route, and a position detection device that detects a current position of the movement device,
While moving a mobile device, based on the current position, when searching and generating a movement route for the mobile device to go to the next via position among the plurality of via positions,
While generating a plurality of moving route candidates, calculating the high-speed driving possibility for each of these moving route candidates,
By comparing these high-speed driving possibilities with each other, a moving route candidate having the highest high-speed driving possibility is selected from the plurality of moving route candidates.
There is provided a route generation method characterized in that the selected movement route candidate is set as a movement route to be actually used.

  According to the present invention described above, the route search device generates a plurality of moving route candidates, calculates a high-speed driving possibility for each of these moving route candidates, and compares these high-speed driving possibilities with each other, The travel route candidate having the highest possibility of traveling at high speed is selected from the plurality of travel route candidates, and the selected travel route candidate is set as the travel route to be actually used. A route is set. Thereby, the moving device can be moved at high speed.

It is a lineblock diagram of the course generating device by a 1st embodiment of the present invention. It is a graph which shows normalization of each element of high-speed driving possibility. Each evaluation area ER of a movement route candidate is shown. It is a flowchart which shows the method of calculating the evaluation value of an evaluation area | region. It is the elements on larger scale which show the evaluation area | region in FIG. It is a figure for demonstrating the road surface state of an evaluation area | region. It is a graph used in order to calculate a road surface favorable degree. It is a figure for demonstrating calculation of the distance of the present movement path | route and a movement path | route candidate. 3 is a flowchart illustrating a route generation method according to the first embodiment of the present invention. It is a local map which a route search device generates. It is explanatory drawing of the path | route production | generation method by 3rd Embodiment of this invention. It is a flowchart which shows the route generation method by 3rd Embodiment of this invention. It is another explanatory view of the route generation method according to the third embodiment of the present invention. It is explanatory drawing of the path | route production | generation method by 4th Embodiment of this invention. It is explanatory drawing of the path | route production | generation method by 5th Embodiment of this invention. It is a block diagram of the path | route production | generation apparatus by 6th Embodiment of this invention. The relationship between a running speed and the average value of the amplitude of vibration is shown. It is a graph which shows normalization of each element of the high-speed driving | running | working possibility by 6th Embodiment.

  Embodiments of the present invention will be described with reference to the drawings.

[First Embodiment]
The route generation device 10 according to the present embodiment includes a storage device 3, a position detection device 5, a route search device 9, and an obstacle detection device 7, as shown in FIG.

  The storage device 3 stores map information related to a map including the movement range of the moving device, and route position information indicating a position on the map (that is, a route position). The map information may be information relating to a two-dimensional map. This map information may be, for example, an xy coordinate system, and a position on the map is specified by given x and y coordinates. It's okay. The via position information is position information (for example, x coordinate and y coordinate) of a plurality of via positions set between the movement start position of the mobile device and the target arrival position. The plurality of via positions are used by the route search device 9 as an index for generating a movement route. Note that the plurality of via positions may include a movement start position and a target arrival position.

The position detection device 5 detects the current position of the moving device. The position detection device 5 is provided in the moving device. The position detection device 5 may be, for example, a GPS device that receives information about a position from a GPS system (GPS satellite) using a GPS system and detects the current position of the mobile device based on the information.
Instead, the position detecting device 5 detects the moving speed and moving direction of the moving device as needed, and calculates and detects the current position of the moving device based on the detected moving speed, moving direction, and initial position information of the moving device. You may do.

The obstacle detection device 7 is provided in the moving device, and detects an obstacle that hinders the movement of the moving device. The moving device detects an obstacle such as a stopped vehicle, uneven surface, wall, and the position / range thereof around the moving device (for example, in a predetermined region on the side where the moving device travels). The obstacle detection device 7 may detect an obstacle by emitting laser light and detecting reflected light of the laser light, for example. In this case, the obstacle detection device 7 detects a laser light source that emits a plurality of laser beams in the traveling direction of the moving device, the reflected light of the emitted laser beams, and the presence / absence of an obstacle and its position / range. And a reflected light detection device for detection. The laser light source emits a plurality of laser beams at an angle inclined with respect to a horizontal direction perpendicular to the traveling direction of the moving device and inclined downward in the vertical direction. The reflected light detection device receives the reflected light of the emitted laser beam, and detects the presence / absence of an obstacle and its position / range based on the received reflected light.
The obstacle detection device 7 is not limited to the above-described device, and may be another known device. For example, in addition to the obstacle detection device using laser light, a device using a CCD camera or the like may be used as the obstacle detection device 7.

  Based on the current position detected by the position detection device 5 during movement of the mobile device, the route search device 9 determines a travel route for the mobile device to go to the next via position among the plurality of via locations. Search and generate. The route search device 9 may be configured by a computer.

  The route search device 9 generates a plurality of travel route candidates, calculates a high-speed travel possibility for each of the travel route candidates, and compares the high-speed travel possibilities with each other. A travel route candidate having the highest possibility of traveling at high speed is selected, and the selected travel route candidate is set as a travel route to be actually used.

The route search device 9 may generate a plurality of moving route candidates by the following equation (Eq1).

κ (S) = a ₁ + a ₂ S + a ₃ S ² +... + a _{n + 1} Sn ⁿ (Eq1)

Κ (S) represented by this equation (Eq1) is the curvature of the moving route candidate (for example, Non-Patent Document 1).
In this formula (Eq1),
n is an integer of 1 or 2 or more,
S indicates a path length along the moving path candidate from the current position of the mobile device to an arbitrary position on the moving path candidate, and the upper limit value of S is the end position of the moving path candidate from the current position Is the travel path length along the travel path candidate to
a _{1 to} a _{n + 1} are coefficients.
In this formula (Eq1), the upper limit value of _{S, a 1, a 2,} a 1 ···, and by _{a n + 1} of a plurality of at least one mutually different kappa (S), respectively, a plurality of moving path candidate It may be specified and generated.

  The above-mentioned κ (S) representing each moving route candidate is obtained by following the azimuth angle of the moving route candidate (that is, the tangential direction of the moving route candidate) and the position on the moving route candidate using S in the equation (Eq1) as a parameter. [Equation 1] (Eq2) and (Eq3).

  The high-speed driving possibility will be described. The high-speed travel possibility includes, as its elements, at least the route length of the travel route candidate, the minimum radius of curvature of the travel route candidate, the road surface goodness, and the proximity of the travel route currently used and the travel route candidate. Contains either (all in this example).

The route search device calculates the high-speed driving possibility according to the value of each element. In this embodiment, the high-speed driving possibility increases as the values of these elements increase. Preferably, the high-speed driving possibility is calculated as a sum of standard values of each element. The standard value is, for example, a value in the range of 0 to 1. If the value that each element can take is not in the range of 0 to 1, the standard value is converted to be in this range. Then, the sum of the standard value of the route length of the moving route candidate, the standard value of the minimum curvature radius of the moving route candidate, and the standard value of the proximity of the currently used moving route and the moving route candidate is calculated. Let it be a high-speed driving possibility.
In addition, the longer the route length of the moving route candidate is, the lower the necessity of temporarily stopping the movement of the mobile device because the next moving route candidate is not found. As a result, the time required to reach the target arrival position is shortened. Also, the higher the proximity is, the smaller the degree of deviation from the currently used moving route from the moving route candidate is, so that the moving route candidate is more likely to have a smaller number of curves. If there are few curves, the part which can move at high speed will increase.

(Path length)
The route length of the travel route candidate is a length along the travel route candidate from the start point to the end point (end position) of the travel route candidate. That is, the upper limit value of S in κ (S).

  The route length of the moving route candidate is converted into the above-described standard value according to the relationship of FIG. In FIG. 2A, the path length and its standard value are in a proportional relationship. In the example of FIG. 2A, when the length of the generated travel route candidate exceeds the set value, the standard value is set to 1.

(Minimum curvature radius)
The minimum curvature radius of the movement path candidate is the minimum value among the curvature radii (turning radii) at each position in the entire range of the movement path candidate (from the start point to the end point of the movement path candidate). Specifically, since the radius of curvature at each position of the movement path candidate is the reciprocal 1 / κ (S) of κ (S) described above, it is the minimum value of 1 / κ (S) at each position. is there.

The minimum curvature radius of the moving route candidate is converted into the above-described standard value according to the relationship of FIG. In the example of FIG. 2B, the minimum curvature radius is R, and the standard value Z of the minimum curvature radius is expressed by the following equation (Eq4).

Z = (1 / n) × logR (Eq4)

That is, the standard value is 1 / n of the logarithm with 10 as the base with respect to the minimum radius of curvature R, and n is 10 with respect to the radius of curvature (setting value) that regards the moving path candidate as a straight line. Logarithmic value. When the minimum curvature radius of the moving route candidate exceeds the curvature radius (setting value) that regards the moving route candidate as a straight line, the standard value is set to 1.

(Good road surface)
The road surface goodness degree is calculated based on the height difference of the road surface area along the moving route candidate. The road surface goodness degree is a minimum value of evaluation values calculated for each evaluation area of a moving route candidate as follows. First, when the moving device moves from the start point to the end point of the moving route candidate along the target moving route candidate, each evaluation area spaced along the moving route candidate is to be passed by the moving device. Each time, based on the height of each measurement point in the evaluation area, it is determined whether the evaluation area is unrunnable, a stepped surface, a flat surface, or an uneven surface. In addition, if there is one that is determined to be unmovable in each evaluation area of the moving route candidate, the road surface goodness of the moving route candidate is set to 0, and otherwise, as follows To. That is, for each evaluation area, when it is determined that the surface is a step surface, an evaluation value of the road surface goodness indicating how small the step is calculated, and when it is determined that the surface is a plane, the road surface When the evaluation value of goodness is calculated and judged to be an uneven surface, an evaluation value of road surface goodness indicating how small the unevenness is calculated, and each evaluation of all evaluation areas of the moving route candidate is calculated The smallest value among the values is set as the road surface goodness of the moving route candidate. In addition, since the road surface favorableness of a movement route candidate takes the value of 0-1, it becomes the above-mentioned standard value as it is.

  In the following, the road surface goodness will be described in detail.

  The road surface goodness degree is calculated by the route search device 9 based on the height of each measurement point incorporated in a two-dimensional local map (described in detail later) including the current position of the mobile device. The height of each measurement point (that is, the position coordinate value in the vertical direction) is the height of each corresponding position on the actual road surface, and the obstacle detector 7 emits a laser beam. It may be calculated based on measurement data obtained by detecting the reflected laser beam. For example, the measurement data is the distance between a point on the road surface that reflects the laser beam and the laser light source, and is converted into the height of the point based on the laser beam transfer direction, the attitude of the moving device, and the like. Then, it is associated with the measurement point on the local map and incorporated into the local map.

  The road surface goodness of each moving route candidate is the lowest value among the evaluation values of each evaluation area (indicated by reference sign ER in FIG. 3) in the moving route candidate. For each evaluation area of the movement path candidate, an evaluation value is calculated based on the height of each measurement point of the evaluation area, and the lowest value among the calculated evaluation values of each evaluation area is set to the movement path candidate. The road surface is good. As shown in FIG. 3, each evaluation area ER of a movement route candidate is represented by a black circle in FIG. 3 at predetermined positions on the movement route candidate (indicated by a broken line in FIG. 3) on a two-dimensional local map. This is a linear region that spreads so that the position is centered on both sides (corresponding to both sides in the horizontal direction) intersecting (for example, orthogonal to) the movement path candidate at the position. The length of the linear region ER is preferably about the length corresponding to the width of the moving device (vehicle body) in the two-dimensional local map.

  The evaluation value of each evaluation area is calculated according to the flowchart of FIG. For each movement path candidate, the evaluation value of each evaluation area of the movement path candidate is calculated according to the flowchart of FIG. 4, and the evaluation value of each evaluation area of the movement path candidate that has the lowest value is calculated as the movement path. Candidate road surface quality.

  A method of calculating the evaluation value of the evaluation area according to the flowchart of FIG. 4 will be described with reference to FIG. FIG. 5 is a partially enlarged view showing an arbitrary one of the plurality of linear evaluation regions ER in FIG.

In step ST1, for each of the left and right traveling tires among the plurality of measurement points (indicated by white circles in FIG. 5) that exist on the evaluation region ER shown in FIG. For each of the measurement points P _{m1 to} P _m4 , the height difference between the two measurement points is calculated for each two adjacent measurement points, the gradient between the two measurement points is calculated, and the variance of these height differences is calculated. To do.
The first to fourth measurement points P _{m1 to} P _m4 will be described with reference to FIG. In FIG. 5, broken lines indicate left and right traveling tires of the moving device (vehicle). As shown in FIG. 5, there are two or more measurement points used in this step ST1 within the width of each tire (the width in the horizontal direction in the figure). More specifically, the two parts surrounded by a broken line in FIG. 5 are the two traveling tires on the left and right of the moving device when the target moving route candidate is set on the local map as described later. These traveling tires when passing through the evaluation region ER are shown. As shown in FIG. 5, the first and second measurement points P _m1 and P _m2 sandwich the left and right ends of the traveling tire in a contact area where the traveling tire contacts the road surface when the road surface is a flat surface. . The third and fourth measurement points P _m3 and P _m4 sandwich the left and right other ends of the traveling tire in a contact area where the traveling tire contacts the road surface when the road surface is a flat surface. The first and fourth measurement points P _m1 and P _m4 are outside the contact area, and the second and third measurement points P _m2 and P _m3 are inside the contact area.
The difference in height between the two measurement points described above is the difference in height between the two measurement points, and the gradient between the two measurement points described above is an inclination angle from the horizontal plane, and the horizontal coordinate in the local coordinates. Atan (Y / X), where X is the difference between the position coordinate values of the two measurement points with respect to the direction, and Y is the difference in height between the two measurement points. Atan is a tangent inverse function (arctangent).
The difference in height between two measurement points is the difference in height between the two measurement points, and the gradient between the two measurement points is an inclination angle from a horizontal plane, It is Atan (Y / X) where X is the actual distance corresponding to the distance between the measurement points, and Y is the difference in height between the two measurement points. Atan is a tangent inverse function (arctangent).

In step ST2, regarding the first to fourth measurement points P _{m1 to} P _m4 corresponding to each traveling tire, the difference in height between two adjacent measurement points (P _m1 and P _m2 etc.) is not less than a predetermined threshold value. It is judged whether it is. That is, if there is a height difference calculated in step ST1 that is equal to or greater than a predetermined height difference threshold value, the process proceeds to step ST3. On the other hand, if none of the height differences calculated in step ST1 is greater than the height difference threshold value, the process proceeds to step ST4.

  In step ST3, the evaluation value is set to 0, assuming that the target evaluation area cannot pass (impossible to travel).

In step ST4, it is determined whether the target evaluation area is a stepped surface. That is, with respect to the first to fourth measurement points P _{m1 to} P _m4 corresponding to each traveling tire, the gradient between the first and second measurement points P _m1 and P _m2 calculated in step ST1, and in step ST1 If any of the calculated gradients between the third and fourth measurement points P _m3 and P _m4 is equal to or less than the upper limit value that can be regarded as a plane, the target evaluation area is determined to be a step surface, and the process proceeds to step ST5. Advance (this upper limit value is smaller than the gradient threshold value). On the other hand, for the first to fourth measurement points P _{m1 to} P _m4 corresponding to at least one of the traveling tires, the gradient between the first and second measurement points P _m1 and P _m2 calculated in step ST1, and If any of the gradients between the first and second measurement points P _m3 and P _m4 calculated in step ST1 is not less than or equal to the upper limit value that can be regarded as a plane, the process proceeds to step ST6.
For example, in the case of FIG. 6A, it is determined that the target evaluation area is a step surface, and the process proceeds to step ST5. In the graphs of FIGS. 6A to 6C, the horizontal axis indicates the horizontal position coordinates of the measurement points P _{m1 to} P _m4 corresponding to the traveling tire, and the vertical axis indicates the measurement points P _{m1 to} P _m4. The white circles indicate the measurement points P _{m1 to} P _m4 .

In step ST5, an evaluation value is calculated according to the larger one of the height differences between the second and third measurement points P _m2 and P _m3 corresponding to each traveling tire calculated in step ST1. For example, the evaluation value is calculated according to the relationship shown in FIG. Graph of FIG. 7 (A) shows the relationship between the size and the evaluation value of the height difference between the measurement point P _{m @ 2,} the measuring point towards the running tire the magnitude of the height difference P _m3 P _m2, P _m3 .

In step ST6, it is determined whether or not the target evaluation area can be regarded as a plane. In other words, if the variance of the height difference calculated in step ST1 is equal to or less than the upper limit value that can be regarded as a plane, the target evaluation area is assumed to be a plane, and the process proceeds to step ST7. Here, the variance is calculated based on the height difference calculated in step ST1. That is, for each height difference calculated in step ST1, the difference between the height difference and the average value of the height difference calculated in step ST1 is squared, and the average value of these squared values is the above-described variance. is there. On the other hand, if the variance of the height difference calculated in step ST1 is not less than or equal to the upper limit value that can be regarded as a plane, the process proceeds to step ST8.
For example, in the case of FIG. 6B, assuming that the target evaluation area is a plane, the process proceeds to step ST7.

  In step ST7, an evaluation value is calculated according to the variance calculated in step ST6. For example, the evaluation value is calculated according to the relationship shown in FIG. The graph in FIG. 7B shows the relationship between the variance and the evaluation value.

In step ST8, it is assumed that the target evaluation area is an uneven surface. For example, in the case of FIG. 6C, it is assumed that the target evaluation region is an uneven surface.
Further, in step ST8, an evaluation value is calculated according to the maximum height difference (referred to as the maximum height difference) calculated in step ST1. For example, the evaluation value is calculated according to the relationship shown in FIG. The graph in FIG. 7C shows the relationship between the maximum height difference and the evaluation value.

(Proximity between current travel route and travel route candidate)
This proximity is calculated using the distance between the currently used / set travel route (hereinafter referred to as the current travel route) and the travel route candidate.

  This distance will be described with reference to FIG. FIG. 8 is a diagram showing moving route candidates on the local map. The distance between the current travel route and the travel route candidate is calculated by the route search device 9 as the sum of the distances L1 to LP between the corresponding points in FIG. First, a plurality of sample points P1 to Pn on the movement path candidate (on the broken line in FIG. 8) are taken, and a point on the current movement path (on the solid line in FIG. 8) closest to the sample point is used as a corresponding point. Ask. Next, for each sample point, the distance between the sample point and its corresponding point is calculated as the distance between corresponding points. Thereafter, the sum of the corresponding point distances L1 to LP of each sample point is calculated as the distance between the current movement route and the movement route candidate.

  Note that the method shown in FIG. 8 may be used as a method of obtaining sample points for each moving route candidate. In other words, a point that has virtually traveled a certain distance along the movement path candidate from the current position of the movement distance is set as the first sample point P1, and further progresses virtually a certain distance along the movement path candidate from this sample point P1. A plurality of sample points P1 to Pn are set as the next sample point P2, and the point that has further advanced virtually a certain distance along the movement path candidate from this sample point P2 as the next sample point P3. You can take it. The last sample point Pn that is farthest along the movement path candidate from the current position of the movement distance may be a point on the movement path candidate that is closest to the end point of the current movement path.

  The distance between the current movement route and the movement route candidate calculated as described above is converted into the proximity (that is, the standard value) according to the relationship shown in FIG. In the example of FIG. 2C, when the calculated distance exceeds the set value, the standard value is set to zero.

  According to the present embodiment, the route search device 9 performs a plurality of movement route candidates generated and found under a predetermined route search condition or under a route search condition changed as described below. As described above, by calculating the high-speed driving possibility for each of these moving route candidates and comparing these high-speed driving possibilities with each other, the movement with the highest high-speed driving possibility among the plurality of moving route candidates is performed. A route candidate is selected, and the selected movement route candidate is set as a movement route to be actually used.

When the route search device 9 cannot find the moving route candidate under the route searching condition, the route searching device 9 changes the route searching condition and searches for the moving route candidate again under the changed route searching condition.
The route search condition (that is, the initial route search condition) includes an interference condition that the obstacle detected by the obstacle detection device 7 does not interfere with the moving route candidate, and the speed of the mobile device is controlled based on the command speed value. An obstacle detection condition that the obstacle detection device performs obstacle detection in a state where the obstacle is detected. In this case, when the route search device 9 cannot find the moving route candidate under the route search condition, the command speed value (used for the speed control in the obstacle detection condition of the route search condition ( The command speed value output from the speed command device is adjusted so that the moving device moves at a lower speed than the previous command speed value (that is, the current command speed value is larger than the previous command speed value). In this case, the route search condition is changed by the obstacle detection device detecting the obstacle while the mobile device is speed-controlled based on the adjusted command speed value. The travel route candidate is searched again under the changed route search condition.

  The route search condition (that is, the initial route search condition) may include a curvature condition in which the curvature of the moving route candidate is equal to or less than a predetermined curvature reference value over the entire range of the moving route candidate.

  The speed control related to the curvature of the moving route candidate will be described. The speed command device outputs a command speed value based on a curvature at a location immediately before the moving device passes on a moving route set as described later. Specifically, the speed command device provides a command speed value corresponding to the curvature (for example, a smaller command speed value as the curvature is larger) so that the moving device can proceed through the part without skidding and rollover. Output. The output command speed value is input to the speed control device, and the speed control device controls the speed of the mobile device so that the mobile device moves at the input command speed value (that is, the speed indicated by this value). To do. Accordingly, the moving device moves at a higher speed as it passes through a portion having a small curvature in the set moving route.

  In the above description, the route search condition is changed by adjusting the command speed value. However, this change may not be performed. For example, the route search condition includes an interference condition and a curvature condition, omitting the obstacle detection condition, and the route search device 9 does not find the moving route candidate under the route search condition. The route search condition may be changed by increasing the curvature reference value.

  Further, the route search device 9 repeats the process of changing the route search condition and searching again for the travel route candidate under the changed route search condition until a travel route candidate satisfying the route search condition is found.

  Next, the search / generation of the movement route by the route generation device 10 will be described in detail. The following explanation corresponds to both the case where the route search condition includes an interference condition, an obstacle detection condition and a curvature condition, and the case where the route search condition includes an interference condition and a curvature condition while omitting the obstacle detection condition. Yes.

  FIG. 9 is a flowchart illustrating a route generation method by the route generation device 10.

In step S1, the route search device 9 generates the two-dimensional local map. That is, the route search device 9 extracts map information of a range including the current position of the mobile device detected by the position detection device 5 and the obstacle search area by the obstacle detection device 7 from the storage device 3, and Generate a local map. In this local map, direction information to the next via position may be incorporated. The direction information may be, for example, the direction of a straight line connecting the immediately preceding route position and the next route position, or may be the direction of the next route position viewed from the current position of the mobile device. In this local map, the next via position itself may be incorporated.
Note that the current position used in step S1 and the obstacle detection result information (obstacle position and range) by the obstacle detection device 7 are respectively the position detection device 5 and the obstacle each time step S1 is started. Acquired by the detection device 7 and incorporated in the local map.

In step S2, a movement route candidate is searched. Specifically, the route search device 9 performs the following processing (A) or (B) using the above equation κ (S).
(A) For κ (S) obtained by giving predetermined values to the plurality of coefficients a _{1 to an + 1} and the upper limit value of S, a moving route candidate determined by κ (S) is the route search condition. It is determined whether (interference condition and curvature condition) are satisfied.
(B) When the process of (A) is performed a plurality of times, each time the process of (A) is performed, at least one value of a ₁ to an _{+ 1} is changed and the process of (A) is performed. Note that the process (A) may be performed a plurality of times while keeping the upper limit value of the S constant, or the upper limit value of the S may be changed in at least one of the processes of the (A). Also good.
If the route search device 9 cannot find a moving route candidate that satisfies the route search condition by the process (A) or (B), the process proceeds to step S3, and a moving route candidate that satisfies the route search condition is found. If YES, go to step S4.

In step S3, the route search condition is changed and the process returns to step S1. Specifically, in step S3, the route search device 9 adjusts the command speed value as described above, and in step S1, the speed of the moving device is controlled based on the adjusted command speed value. Then, the obstacle detection device 7 performs obstacle detection again. The local map is generated based on the obstacle detection result. At this time, the current position of the mobile device may also be detected again by the position detection device 5, and a local map may be generated based on the current position and the obstacle detection result performed again.
In step S3, the route search apparatus 9 may adjust the command speed value as described above and increase the curvature reference value.
In step S3, the command speed value may not be adjusted. That is, the route search condition may be changed simply by increasing the curvature reference value. In this case, unlike FIG. 9, the process may return from step S3 to step S2 without going through step S1, or as shown in FIG. 9, the process returns from step S3 to step S1 to find a new current position and obstacle. A local map may be generated again based on the detection result.

In step S4, for the plurality of movement route candidates found in step S2, as described above, the high-speed driving possibility is calculated for each of the movement route candidates, and the high-speed driving possibilities are compared with each other, thereby calculating the plurality of moving route candidates. The travel route candidate having the highest possibility of traveling at high speed is selected from the travel route candidates, and the selected travel route candidate is set as a travel route to be actually used.
If there is only one moving route candidate found in step S2, step S4 is omitted and the process proceeds from step S2 to step S5.

  In step S5, the movement route candidate selected in step S4 is set as a movement route to be actually used, and the movement route is updated. That is, the set travel route is updated by replacing the set travel route that has been used for the progress control of the mobile device with the travel route candidate selected in step S4. The set movement route is stored in the storage device 3. If it updates, it will progress to step S6. When step S4 is omitted, only one moving route candidate found in step S2 is set as a moving route, and the moving route is updated.

In step S6, the route search device 9 determines whether or not a predetermined time (for example, 100 ms) has elapsed since the start of generation of the local map in step S1, by measuring the time of a timer built in the route search device 9. . If it is determined that the predetermined time has elapsed, the process returns to step S1, and if it is determined that the predetermined time has not elapsed, the determination in step S6 is repeated.
Thus, a local map is generated and a route search is performed every predetermined time. In response to this, the position detection device 5 detects the current position of the moving device every time the predetermined time elapses, and the obstacle detection device 7 detects the obstacle every time the predetermined time elapses.
When returning from step S6 to step S1 and returning to step S2, the route search in step S2 is not the changed route search condition but the initial route search condition (for example, the initial interference condition, obstacle detection condition, and curvature condition). ).

  During the processes of steps S1 to S4 and S6, the control device of the moving device controls the moving device so that the moving device moves along the set moving path stored in the storage device 3.

[Effects of First Embodiment]
According to the first embodiment described above, the route search device 9 generates a plurality of moving route candidates, calculates the high-speed driving possibility for each of these moving route candidates, and compares these high-speed driving possibilities with each other. Then, the travel route candidate having the highest possibility of traveling at high speed is selected from the plurality of travel route candidates, and the selected travel route candidate is set as the travel route to be actually used. A high travel route is set. Thereby, the moving device can be moved at high speed.

In addition, there is a case where an obstacle is erroneously detected because the moving device swings and vibrates due to the moving device moving at a high speed as a reason why the moving path candidate cannot be found. For example, the detection direction of the obstacle detection device may be changed to the lower side due to swinging / vibration, and the road surface itself may be detected as an obstacle.
On the other hand, in the first embodiment, when the moving route candidate cannot be found, the command speed value is adjusted as described above, and the obstacle detecting device 7 is in a state where the moving device is moving at a lower speed. Since object detection is performed, the possibility that a moving path candidate cannot be found due to an erroneous detection of an obstacle can be effectively reduced.

  In addition, an obstacle detected by the obstacle detection device 7 (for example, a stopped vehicle, a hole, a rock, a wall, etc.) and a moving route candidate under a route search condition including an interference condition that the moving route candidate does not interfere with the obstacle move. There are cases where a route candidate cannot be found. In this case, by increasing the curvature reference value of the curvature of the moving path candidate, it is possible to find a moving path candidate that does not interfere with the obstacle, that is, a moving path candidate having a relatively large curvature / curve so as to avoid the obstacle. The potential increases effectively. Accordingly, it is possible to effectively increase the possibility that the moving route candidate can be found and the movement of the moving device can be continued.

[Example of the first embodiment]
Next, examples of the first embodiment will be described. In this embodiment, the mobile device may be a vehicle.

  The route search device 9 enlarges the obstacle range actually detected by the obstacle detection device 7 by a predetermined size and recognizes it on the local map. That is, the range of the obstacle actually detected by the obstacle detection device 7 is enlarged in all directions by a predetermined size (for example, half the vehicle width + one pixel on the local map) Recognized and set on the map.

According to the present embodiment, the curvature of the moving route candidate is κ (S) as described above, and κ (S) is a third-order polynomial of S. That is, κ (S) is expressed by the following equation (Eq5).

κ (S) = a ₁ + a ₂ S + a ₃ S ² + a ₄ S ³ (Eq5)

In this formula:
S indicates a path length along the moving path candidate from the current position of the mobile device to an arbitrary position on the moving path candidate, and the upper limit value of S is the end position of the moving path candidate from the current position Is the travel path length along the travel path candidate to
a _{1 to} a ₄ are parameters (coefficients).

A moving route candidate can be represented by the above-described κ (S). _That, a by giving a predetermined value to 1 ~a _4, in other _words, a 1 and the value of ~a ₄ is determined determined movement route candidate. In this embodiment, by giving the following three boundary conditions, for example, a ₁ , a ₃ , and a ₄ can be expressed by a function of a ₂ .

boundary condition:
(1) The next boundary condition (Eq6) for the curvature at the starting point of the moving path candidate, that is, the current position of the moving device

κ (0) = κ ₀ = a ₁ (Eq6)

(2) The following boundary condition (Eq7) for the curvature at the end point S _f (ie, the upper limit value of S) of the travel route candidate to be searched and generated

κ (S _f ) = κ _f = a ₁ + a ₂ S _f + a ₃ S _f ² + a ₄ S _f ³ (Eq7)

(3) movement path direction at the end point S _f of the movement route candidate of the search-generation target following boundary conditions for (tangential movement path) (EQ8)

_{θ (S f) = θ f} = a 1 S f + a 2 S f 2/2 + a 3 S f 3/3 + a 4 S f 4/4
... (Eq8)

Note that κ ₀ , κ _f , and θ _f are already known. Preferably, κ _f is zero, and θ _f is in the same direction as a straight line connecting the end point of the moving route candidate and the next waypoint.

From the above boundary conditions, the following equations (Eq9), (Eq10)

a ₃ S _f ² + a ₄ S _f ³ = κ _f −κ ₀ −a ₂ S _f (Eq9)

^{_{a 3 S f 3/3 +}} a 4 S f 4/4 = θ f -κ 0 S f -a 2 S f 2/2 ··· (Eq10)


Therefore, the following relational expression 1 is obtained from the above formulas (Eq9), (Eq10), and the like.
[Relational expression 1]

a ₁ = κ <sub>0

a 3 = (- 3κ f -9κ</sub> 0) / S f 2 -3a 2 / S f + 12θ f / S f 3

a ₄ = 4 (κ _f + 2κ ₀ ) / S _f ³ + 2a ₂ / S _f ² −12θ _f / S _f ⁴

Therefore, a moving route candidate (trajectory) can be uniquely obtained using a ₂ as a parameter. The above relational expression 1 between each of a ₁ , a ₃ and a ₄ and a ₂ , κ ₀ , κ _f , and S _f is stored in the storage device 3 and used when searching for a route. Note that by setting the boundary conditions as described above, given the value of a _2, can be obtained in a short time of operation of _{a 1, a 1, a 4} . As a result, a moving route candidate can be specified in a short time by κ (S).

  In this embodiment, the route search condition includes an interference condition that the obstacle detected by the obstacle detection device 7 does not interfere with the moving route candidate, and a state in which the speed of the moving device is controlled based on the commanded speed value. One having an obstacle detection condition that the obstacle detection device performs obstacle detection, a curvature condition that the curvature of the moving path candidate is equal to or less than a curvature reference value over the entire range of the moving path candidate, and a set length And a generation route selection condition of selecting, from the above movement route candidates, a condition that satisfies the interference condition as a generation movement route candidate.

(About the curvature condition)
When the mobile device is a device that travels (for example, a vehicle), the upper limit value of the curvature κ (S) is, for example, the smallest one of the following κturn, κslip, and κrollover.
(1) A curvature κturn of a moving path candidate indicating the minimum turnable radius of the vehicle.
(2) The limit curvature κslip that does not cause skidding when the vehicle travels at a predetermined speed. In other words, when the vehicle travels at a predetermined speed, the vehicle skids at a location where the curvature κ (S) is larger than the curvature κ slip on the moving route candidate.
(3) The limit curvature κrollover that does not cause rollover when the vehicle travels at the predetermined speed. That is, when the vehicle travels at the predetermined speed, the vehicle rolls over at a location having a curvature κ (S) larger than the curvature κ rollover on the moving route candidate.
In general, the values of the curvature κ slip and the curvature κ rollover are smaller than the value of the curvature κ turn. Further, the values of the curvature κ slip and the curvature κ rollover vary depending on the predetermined speed of the vehicle, and generally decrease as the predetermined speed of the vehicle increases. The value of the curvature κslip also depends on the coefficient of dynamic friction between the vehicle and its running surface.
Therefore, when κ (S) satisfies the following magnitude relationship, the curvature condition is satisfied.

κ (S) ≦ κturn, κ (S) ≦ κslip, and κ (S) ≦ κrollover,

Such curvature κslip and curvature κrollover can be obtained based on characteristics of a mobile device (for example, a vehicle). For reference, the curvature κ slip may be, for example, Equation (2) described in Non-Patent Document 2, and the curvature κ rollover may be Equation (4) described in Non-Patent Document 2. Good.

(About generation route selection conditions)
When searching for a moving route candidate determined by κ (S), the route search device 9 fixes the upper limit value S _f of S to a set length value and gives the value of the parameter a ₂ , The values of a ₁ , a ₃ , and a ₄ are determined from the relational expressions 1 to determine whether the determined moving path candidate satisfies the interference condition and the curvature condition.

(About setting values)
The route search device 9 performs a route search by giving predetermined values to parameters and coefficients used for the route search. For this purpose, the storage device 3 stores the preset values of these parameters and coefficients. Specifically, the storage device 3 stores set values such as upper limit values S _f , a ₂ , κ ₀ , κ _f , and θ _f of the path length S.
Further, in order to generate a plurality of travel route candidates, setting the plurality of values of a _2. For example, the maximum value _a 2 max the minimum value _a 2 min and _{a 2} of _{a 2,} may be set the increment da ₂ of _{a 2} from the minimum value, the route search unit _9, a 2 min, _{a 2 min + da 2, a} 2 min + 2da 2, a 2 + 3da 2 ···, keep to the movement route candidate with each of a 2 max is determined. Thereby, a plurality (for example, 21) of movement route candidates can be generated.
The storage device 3 stores the curvature reference value of the curvature κ (S).
Further, when κslip and κrollover depend on variables such as the speed of the mobile device, the storage unit 3 also stores the relational expression between these variables and κslip and the relational expression between these variables and κrollover. As a result, the route search device 9 calculates κslip and κrollover using the relational expression based on the values of these variables at any time, compares the magnitudes of κturn, κslip, and κrollover, and determines the smallest one of them. Is specified as the curvature reference value. Note that κturn may be a fixed value stored in the storage device 3.
Furthermore, the storage device 3 stores the predetermined threshold value related to the unknown area, a command speed value decrease amount / decrease rate, which will be described later, and a path length increase amount / increase rate.

  Next, the movement route search method according to the present embodiment will be described. This method is the same as the flowchart shown in FIG. 9, but this embodiment will be described in more detail as an example of the method of FIG. The points not described above will be described in detail below, but the points not described below may be the same as described above.

  In step S1, a local map is generated as described above.

In step S2, a movement route candidate is searched. Specifically, each of a plurality of a ₂ set as described above is applied to the equation for κ (S). The value of _{a 1,} a 3, a ₄ are, _{a 2} as equation 1 for each set value of _{a 2, κ 0, κ f} , S f, obtained by using the theta _f. In this way, a plurality of movement route candidates are generated.
It is determined whether these movement path candidates satisfy the interference condition and the curvature condition. The curvature condition, for each movement path candidates, kappa (S) is, over the entire range of 0 ≦ S ≦ S _f, to determine whether it is less than the curvature reference value described above kappa (S). When it is below the curvature reference value, the curvature condition is satisfied.
In this way, if there is no moving path candidate that satisfies the interference condition and the curvature condition, that is, if none of the moving path candidates satisfy both the interference condition and the curvature condition, the moving path candidate is not viewed. Since it was not issued, it progresses to step S3. On the other hand, if there is a moving path candidate that satisfies the interference condition and the curvature condition, the moving path candidate is found, and the process proceeds to step S4.

In step S3, the route search condition is changed. In this embodiment, in addition to the adjustment of the command speed value described above, to increase the curvature reference value, and the value of the upper limit value S _f of path length S may be to increase, changing the route search conditions. Incidentally, an increase in the curvature reference value is carried out based on the curvature reference value larger volume and magnification stored in the storage device 3, an increase in the value of the upper limit value S _f is stored in the storage unit 3 and the path length increment・ Based on the rate of increase. If the route search conditions are changed, the process returns to step S2 to perform a route search under the changed route search conditions. This re-route search process is the same as the process performed in the previous step S2 except that the route search conditions are changed.
By doing in this way, even if there is an obstacle on the moving route candidate in the previous route search, the values of a ₃ and a ₄ are also changed by the relational expression 1 by increasing S _f. In addition, the route of the moving route candidate determined by κ (S) is also different from the previous time. Moreover, the total length of the generated travel route candidates is also increased. Thereby, for example, as schematically illustrated in FIG. 10, the possibility that a moving route candidate extending so as to avoid an obstacle is effectively increased. In FIG. 10, a plurality of thin curves indicate previously generated movement route candidates, and a thick curve indicates a movement route candidate generated this time (in FIG. 10, only one is shown for simplicity).

  The process in step S4 is the same as described above.

  In step S5, the set movement route is updated as described above.

  In step S6, it is determined whether a predetermined time (for example, 100 ms) has elapsed since the generation of the local map was started as described above. If it is determined that the predetermined time has elapsed, the process returns to step S1, and if it is determined that the predetermined time has not elapsed, the determination in step S6 is repeated.

[Second Embodiment]
A second embodiment of the present invention will be described. The route generation apparatus and method according to the second embodiment are different from the first embodiment in the points described below. Other contents of the route generation device and method according to the second embodiment may be the same as those of the first embodiment.

According to the second embodiment, the route search device 9 performs the following processes (a) and (b) when searching for a route.
(A) The route search device 9 determines whether the one or more moving route candidates satisfy an interference condition.
(B) When it is determined that none of the movement path candidates satisfies the interference condition, the one or more movement path candidates are changed, and the changed one or more movement path candidates satisfy the interference condition. Judge whether to do.

  The change of the one or more moving route candidates is the increase in the set length, and the route from the current position to the arbitrary intermediate position in each of the changed one or more moving route candidates This is different from any one of the one or more moving route candidates.

According to the second embodiment, the process of step S2 described in the first embodiment is performed as follows unlike the first embodiment.
That is, the route search apparatus 9 performs the following processing (A) or (B) using the above-described equation κ (S) = a ₁ + a ₂ S + a ₃ S ² +... + _{An +} S ⁿ .
(A) For κ (S) obtained by giving predetermined values to the upper limit value of S and the plurality of coefficients a _{1 to} a _{n + 1} , the moving path candidate determined by κ (S) satisfies the interference condition. Judge whether to meet.
(B) When the process (A) is performed a plurality of times, each time the process (A) is performed, at least one value of a ₁ to an _{+ 1} is changed, and the process (A) is performed. Thereby, it is determined whether an interference condition is satisfied for the plurality of movement route candidates obtained by a plurality of processes.

Further, according to the second embodiment, if it is determined in step S2 that none of the movement path candidates satisfies the interference condition in the process (A) or (B), the process proceeds to step S3.
In the second embodiment, in step S3, the upper limit value of S is increased, and at least _one of the coefficients a _{1 to an + 1} is set to the corresponding coefficient in the above (A) or (B). By changing from the value given in the process, the one or more moving route candidates are changed, and the process (A) or (B) is performed again.

[Third Embodiment]
Next, a route generation device 10 and a route generation method according to a third embodiment of the present invention will be described. The point that the route generation device 10 and the route generation method according to the third embodiment are not described below may be the same as those in the first embodiment or the second embodiment, or may be appropriately selected from the first embodiment or the second embodiment. It may be changed.

  In the above-described step S1, the route search device 9 determines the local map near the current position X based on the current position X of the mobile device and the obstacle detected by the obstacle detection device 7 provided in the mobile device. Is generated. FIG. 11A is an example of this local map. In FIG. 11A, the upper side is the direction of the passing position (next passing position) that passes next. In addition, the route search device 9 further specifies a travelable region 13 and an unmeasured region 15 described later in the local map based on known terrain data indicating a region where an obstacle exists and a region where no obstacle exists. May be. This terrain data may be stored in the memory of the route search device 9 or may be received by the route search device 9 by wireless communication.

  Step S2 described above is different in the third embodiment from the case of the first embodiment or the second embodiment as follows.

  Step S2 includes steps S21 to S28 as in the flowchart of FIG.

In step S21, the route search device 9 determines whether the left and right obstacles 11a and 11b are present in the two-dimensional local map when the obstacles 11a and 11b exist on the left and right sides with respect to the direction toward the next passing position. The travelable area 13 of the mobile device existing in between is extracted. The travelable area 13 is an area where no obstacle exists, and has a boundary in contact with the left and right obstacles 11a and 11b. In FIG. 11A, the stitch portions are obstacles 11a and 11b.
In the example of FIG. 11A, the obstacles 11a and 11b and the travelable area 13 are detected by the obstacle detection device 7, or by the obstacle detection device 7 and the terrain data. In FIG. 11A, the other part is an undetected area 15 where the presence or absence of the obstacles 11a and 11b is not detected by the obstacle detection device 7. Note that, as shown in FIG. 11A, the travelable region 13 detected by the obstacle detection device 7 may be discontinuous due to the shaking of the mobile device during travel.

In step S <b> 22, the route search device 9 generates a plurality of inscribed circles that are located in the travelable area 13 and touch the boundary of the travelable area 13 in the local map. These inscribed circles are generated at an interval from the current position X of the moving device to the side of the next passing position.
FIG. 11B shows the inscribed circle generated for the case of FIG. Inscribed circles in contact with both the left and right obstacles 11a and 11b are generated in order from the current position X side.
The center of the next inscribed circle (referred to as the target inscribed circle) is the center of the inscribed circle (referred to as the previous inscribed circle) generated immediately before this target inscribed circle, and the one before this. Is located on a straight line (referred to as an orthogonal straight line) orthogonal to a straight line passing through the center of the inscribed circle (referred to as the inscribed circle two before). The inscribed circle generated next is the largest inscribed circle whose center is located on the orthogonal straight line.
Note that the orthogonal straight line is a fixed distance from the center of the inscribed circle two immediately before the inscribed circle to be generated next, regardless of the number of the inscribed circle to be generated next. It may be generated at a position that is only a distance away. The first inscribed circle is generated in the vicinity of the current position X, and the second inscribed circle is positioned on a straight line orthogonal to a straight line passing through the current position X and the center of the first inscribed circle. May be generated.

  In step S23, the route search device 9 generates a road line 17 that passes through the center of each inscribed circle in the local map. FIG. 11C shows a road line 17 generated for the case of FIG. The road line 17 may be a curve obtained by approximating a broken line generated by connecting the centers of adjacent inscribed circles of each pair by an appropriate method (for example, the least square method or the RANSAC method).

In step S24, a large number of linear search lines are extended radially from the current position X. FIG. 11D shows a large number of search lines generated for the case of FIG.
A search adjacent to one end search line La among a plurality of search lines (indicated by symbol G in FIG. 11D) that are adjacent to each other and reach the undetected region 15 without colliding with the obstacles 11a and 11b. The line Lb identifies the collision point that collides with the left obstacle 11a as the first collision point Pa, and among the plurality of search lines (indicated by reference sign G in FIG. 11D), the other search line. A search line Ld adjacent to Lc identifies a collision point that collides with the right obstacle 11b as the first collision point Pb. Preferably, the angle formed by the adjacent search lines is the same for any adjacent search line.

  In step S25, the first collision point Pa and the second collision point Pb are compared, and the one closer to the current position X is specified as the target collision point. In the example of FIG. 11D, the second collision point Pb is the target collision point.

  In step S26, the left and right obstacles 11a and 11b extending from the target collision point, orthogonal to the road line 17 and including the first collision point Pa and the second collision point Pb, respectively, are on the opposite side of the target collision point. A line segment 19 extending to the obstacle is generated. FIG. 13A shows a line segment 19 generated for the case of FIG.

  In step S27, the center position of the generated line segment 19 is set as the via point Pi. This center position is a point equidistant from both ends of the line segment 19. FIG. 13B shows via points Pi set for the case of FIG.

  In step S28, a plurality of movement route candidates reaching the line segment 19 generated in step S26 (that is, a point on the line segment 19) are generated. Preferably, these movement route candidates are routes that face the same direction as the road line 17 at the via point Pi. That is, the tangent line of the movement path candidate at the intersection of each of these movement path candidates and the line segment 19 is parallel to the tangent of the road line 17 at the intersection of the road line 17 and the line segment 19. In step S28, the method for generating the plurality of movement path candidates is plural in step S2 of the first embodiment or the second embodiment described above except that the movement path candidates reaching the line segment 19 are generated in this way. This method may be the same as the method of generating the moving route candidates. For example, a plurality of moving route candidates that further satisfy the route search condition described above may be generated.

  When a plurality of movement route candidates reaching the line segment 19 are generated in step S28, the process proceeds to step S4 described above, and the route search device 9 calculates the high-speed driving possibility for each of these movement route candidates.

In step S4, the distance from the waypoint Pi is further included as an element of the high-speed driving possibility. The distance from the transit point Pi is the distance between the intersection of the movement route candidate generated in step S28 and the line segment 19 and the transit point Pi. The standard value of the distance from the via point Pi is 1 when the distance is zero, gradually decreases as the distance increases from zero, and the distance is reduced from ½ of the length of the line segment 19. When the subtracted value becomes 1/2 or less of the width of the moving device, it becomes zero. In this way, when passing through the route point Pi, the standard value is maximized as the highest speed traveling is possible.
Therefore, the route search device 9 considers not only the standard value of each element of the first embodiment or the second embodiment but also the standard value of the distance from the via point Pi (for example, standard values of all elements). Calculate the high-speed driving possibility.

  According to the third embodiment, since the road line 17 can be generated using the inscribed circle as described above, an obstacle is provided for a portion where the width of the travelable region 13 is abruptly narrowed as shown in FIG. It is possible to generate a safer and faster traveling route so as to reduce the possibility of contact with the vehicle.

[Fourth Embodiment]
Next, a route generation device 10 and a route generation method according to a fourth embodiment of the present invention will be described. The route generation device 10 and the route generation method according to the fourth embodiment are modifications of the third embodiment, and the points not described below are the same as those of the third embodiment.

  In the fourth embodiment, steps S22 and S23 described above are omitted. That is, step S24 is performed after step S21.

In the fourth embodiment, step S26 is performed as follows.
That is, in step S26, the distance from the target collision point is the longest at the boundary of the obstacle on the side opposite to the target collision point among the obstacles 11a and 11b including the first collision point Pa and the second collision point Pb. A point to be shortened is set, and a line segment 21 extending from the point to the target collision point is generated. FIG. 14A shows a line segment 21 generated for the case of FIG.

  In step S27, the center position of the line segment 21 generated in step S26 is set as the via point Pi. FIG. 14B shows the via point Pi set for the case of FIG.

  In step S28, a plurality of movement path candidates reaching the line segment 21 (that is, reaching a point on the line segment 21) are generated. Preferably, these movement route candidates are routes that are orthogonal to the line segment 21 generated in step S25 at the via point Pi.

  In step S4 performed after step S28, the distance from the via point Pi is further included as an element of the high-speed driving possibility as in the third embodiment. The distance from the via point Pi is the distance between the intersection of the moving route candidate generated in step S28 with the line segment 21 and the via point Pi. The standard value of the distance from the via point Pi is 1 when the distance is zero, gradually decreases as the distance increases from zero, and the distance is reduced from ½ of the length of the line segment 23. When the subtracted value becomes 1/2 or less of the width of the moving device, it becomes zero. In this way, when passing through the route point Pi, the standard value is maximized as the highest speed traveling is possible.

  According to the fourth embodiment, as described above, the travel route candidate is generated and selected based on the line segment 21 and the via point Pi, so that the width of the travelable region 13 is abruptly narrowed as shown in FIG. On the other hand, it is possible to generate a safer and faster traveling route so as to reduce the possibility of contact with an obstacle.

[Fifth Embodiment]
Next, a route generation device 10 and a route generation method according to a fifth embodiment of the present invention will be described. The route generation device 10 and the route generation method according to the fifth embodiment are modification examples of the third embodiment or the fourth embodiment, and the points not described below are the same as those of the third embodiment or the fourth embodiment. .

  The fifth embodiment relates to a case where the travelable area 13 is branched. For example, the fifth embodiment relates to a case where the local map generated in step S1 is as shown in FIG.

  First, in the fifth embodiment, the first collision point and the second collision point are specified in the above-described step S24 as in the third embodiment. FIG. 15B shows a pair of the first collision point Pa0 and the second collision point Pb0 specified as described above with respect to the case of FIG.

Furthermore, in the fifth embodiment, the following first and second collision points are also specified in the above-described step S24.
That is, when the difference between the lengths of two search lines that are adjacent to each other and reach the obstacle 11 is larger than the threshold value, the first search line becomes a point at which each of the two search lines collides with the obstacle 11. The collision point and the second collision point are also specified. FIG. 15B shows a pair of the first collision point Pa1 and the second collision point Pb1 specified as described above and another pair of the first collision point Pa2 and the case of FIG. 15A. A second collision point Pb2 is shown.

  The threshold value used in step S24 may be a value corresponding to the width of the moving device. That is, this threshold value is the same as the width of the moving device when converted to the actual length. The search line used in step S24 is the same as the search line described in the third embodiment.

In the fifth embodiment, when a plurality of pairs of first collision points and second collision points are specified in step S24, it is considered that the travelable region 13 is branched. In the example of FIG. 15B, three pairs of the first collision point and the second collision point are specified.
Accordingly, it is determined which of the plurality of travelable areas 13 considered to be branched is to be advanced.

  In a position where there are a plurality of travelable areas 13, when a command indicating which travelable area 13 should be passed is held in the control device of the mobile device, the travel indicated by the command is possible. Proceed to region 13. That is, the route search device 9 is assumed to travel between a pair of the first collision point and the second collision point corresponding to the travelable area 13 indicated by the command, and with respect to the first collision point and the second collision point, Steps S25 to S28 described above are performed in the third embodiment or the fourth embodiment.

  On the other hand, in a case where such a plurality of travelable areas 13 are present, if a command indicating which travelable area 13 should be passed is not held in the control device of the mobile device, the third embodiment The first collision point Pa0 and the second collision point Pb0 specified by the method according to the above are assumed to travel between the first collision point Pa0 and the second collision point Pb0, and the above-described according to the third embodiment or the fourth embodiment. Steps S25 to S28 are performed.

  Note that steps S25 to S28 according to the fourth embodiment are performed for the first collision point and the second collision point that are not specified by the method according to the third embodiment but are specified by the method according to the fifth embodiment.

  FIG. 15C generates a line segment 21 by performing the above-described steps S25 and S26 according to the fourth embodiment at the first collision point Pa1 and the second collision point Pb1 with respect to the case of FIG. 15B. Shows the case. FIG. 15D shows a case where the via point Pi is set by performing the above-described step S27 according to the fourth embodiment with respect to the case of FIG. 15C.

[Sixth Embodiment]
Next, a route generation device 10 and a route generation method according to a sixth embodiment of the present invention will be described. In the sixth embodiment, points not described below are the same as those in the first embodiment, the second embodiment, the third embodiment, the fourth embodiment, or the fifth embodiment described above.

  In the sixth embodiment, the moving device is a vehicle that travels on the road surface by traveling wheels that are rotationally driven. For example, the traveling wheel may be rotationally driven while being in contact with the road surface.

  According to the sixth embodiment, it is determined whether the road surface is an unpaved road surface or a paved road surface based on the vibration generated in the vehicle during traveling. If it is determined that the road surface is an unpaved road surface, A moving route candidate having a high road surface quality based on the difference is easily selected.

  In the sixth embodiment, the vehicle is equipped with a vibration sensor 23 (see FIG. 16) that measures the vibration of the vehicle when the vehicle is traveling. In this embodiment, the vibration sensor 23 is a vertical gyro that measures the acceleration of the vehicle in the vertical direction as the vibration.

  As shown in FIG. 16, the route generation device 10 according to the sixth embodiment is based on the vibration measured by the vibration sensor 23, and the road surface on which the vehicle is currently traveling is either an unpaved road surface or a paved road surface. A road surface determination device 25 for determining whether or not there is provided.

  The road surface determination device 25 determines that the road surface is an unpaved road surface when the magnitude of vibration measured by the vibration sensor 23 exceeds a predetermined threshold, and otherwise, the road surface is a paved road surface. You may determine that there is.

  A specific example of this determination method will be described.

  On the paved road surface, the magnitude of the vibration does not depend on the traveling speed of the vehicle and usually tends to be a predetermined value or less. On the other hand, on the unpaved road surface, the magnitude of the vibration tends to increase as the traveling speed of the vehicle increases.

Therefore, the road surface determination device 25 is such that the vehicle traveling speed measured by the speed sensor mounted on the vehicle exceeds the speed threshold, and the magnitude of the vibration measured by the vibration sensor 23 is When it is below the vibration threshold value, it is determined that the road surface is a paved road surface, and when it is not, it is determined that the road surface is an unpaved road surface.
The speed threshold is a lower limit value of the traveling speed of the vehicle in which when the vehicle travels at the traveling speed of the value, there is a difference in the magnitude of vibration of the vehicle between a paved road surface and an unpaved road surface. The vibration threshold is the magnitude of the vibration of the vehicle that is difficult to appear when the vehicle travels on a paved road surface at a traveling speed, and the vehicle travels on an unpaved road surface at a traveling speed that exceeds the speed threshold value. Is the lower limit of the magnitude of the vehicle vibration appearing at For example, since the running speed and the average value of vibration amplitude (vibration magnitude) tend to be shown in the graph of FIG. 17, the speed threshold value and the vibration threshold value may be set as shown in FIG. .

  When the vehicle has a suspension spring that absorbs vibration generated in the vehicle, damping vibration is generated in the vehicle due to acceleration / deceleration of the vehicle by the suspension spring. Therefore, in this case, a high-pass filter having a cutoff frequency (for example, 3 Hz) larger than the frequency of such damping vibration is used. That is, the road surface determination device 25 extracts only the vibration component having a frequency equal to or higher than the cutoff frequency from the vibrations measured by the vibration sensor 23, and performs the above-described determination based on the extracted vibration component.

  The road surface determination device 25 performs the above-described determination based on the vibration amplitude measured by the vibration sensor 23 over a predetermined time when the traveling speed measured by the speed sensor exceeds the speed threshold value. The measured vibration amplitude is measured at each time point within the predetermined time. The road surface determination device 25 calculates the average value (for example, the mean square value) of the amplitude of vibration at each time point (for example, the magnitude of the vibration component extracted by the high-pass filter) as the magnitude of the vibration, When the average value is less than or equal to the vibration threshold value, it is determined that the road surface is a paved road surface, and otherwise, it is determined that the road surface is an unpaved road surface.

  The high-speed travel possibility includes not only the road surface goodness but also one or both of the route length and the minimum curvature radius. Preferably, the high-speed traveling possibility has, as its elements, the above-described route length, minimum curvature, road surface goodness, and proximity.

  The route search device 9 calculates the value of each element of the high-speed driving possibility by converting it into a standard value (for example, having a value in the range of 0 to 1 as described above), and calculates the sum of the standard values of each element. Calculated as the high-speed driving possibility.

  When it is determined that the road surface is an unpaved road surface with respect to the same moving route candidate, the route search device 9 is configured so that the route length and the minimum are larger than when the road surface is determined to be a paved road surface. Calculate the standard value of the radius of curvature small.

  A specific example of the calculation method of the standard value of the path length according to the sixth embodiment will be described. As specific examples, calculation methods 1 and 2 will be described.

(Calculation method 1)
When the road surface determination device 25 determines that the road surface is a paved road surface, the route search device 9 calculates the standard value of the route length by the same method as in the first embodiment described above. On the other hand, when the road surface determination device 25 determines that the road surface is an unpaved road, the route search device 9 sets the route length standard value calculated by the same method as in the first embodiment to zero. The value obtained by multiplying the above values less than 1 is calculated as the standard value of the path length.

(Calculation method 2)
The route search device 9 calculates the standard value of the route length as the standard value of the stop distance of the vehicle. The stop distance of the vehicle is expressed by the following formula.

Stop distance = V × t + V ² / (2 × μ × G)

Here, V is the upper limit value of the traveling speed of the vehicle (the maximum speed that the vehicle can output). t is an idle running distance of the vehicle to be described later. μ has a unit of the dynamic friction coefficient of the road surface. For example, when the road surface determination device 25 determines that the road surface is a paved road surface, it is 0.8, and the road surface determination device 25 determines that the road surface is an unpaved road surface. If it is determined that there is, it is assumed to be 0.5. G is the gravitational acceleration.
The idle running distance is the time from when the vehicle control device outputs a command signal to stop the vehicle when the vehicle is running until the vehicle braking device (brake) starts operating.
An example of the standard value calculated in this way is shown in FIG.

  A specific example of a method for calculating the standard value of the minimum radius of curvature according to the sixth embodiment will be described. As specific examples, calculation methods 1 and 2 will be described.

(Calculation method 1)
When the road surface determination device 25 determines that the road surface is a paved road surface, the route search device 9 calculates the standard value of the minimum curvature radius by the same method as in the first embodiment. On the other hand, when it is determined by the road surface determination device 25 that the road surface is an unpaved road, the route search device 9 sets the standard value of the minimum curvature radius calculated by the same method as in the first embodiment to zero. The value obtained by multiplying the above values less than 1 is calculated as the standard value of the minimum curvature radius.

(Calculation method 2)
When the vehicle is traveling at the traveling speed V, the minimum turning radius at which the vehicle can turn without skidding is expressed by the following equation.

Minimum turning radius = V ² / (μ × G)

Here, V is the traveling speed of the vehicle. μ is the dynamic friction coefficient of the road surface. G is the gravitational acceleration.
The value of μ is smaller for an unpaved road surface than for a paved road surface. Accordingly, the standard value of the minimum radius of curvature is expressed by the above-mentioned standard value (1 / n) × logR when the road surface is a paved road surface as shown in FIG. 18B, but when the road surface is an unpaved road surface. May be expressed as a standard value obtained by multiplying this standard value by a value greater than zero and less than one. In FIG. 18B, when the minimum radius of curvature is equal to or less than the minimum turning radius R0 at which the vehicle can turn when the traveling speed is substantially zero, the standard value is set to zero.

  According to the sixth embodiment, when it is determined that the road surface is an unpaved road surface, the standard values of the route length and the minimum curvature radius are reduced, so that the degree that the road surface goodness contributes to the high-speed traveling possibility increases. Accordingly, when it is determined that the road surface is an unpaved road surface, it is easy to select a moving route candidate having a high road surface quality, and on an unpaved road surface, it is possible to perform control for traveling on a moving route having a high road surface quality. .

[Mobile device including route generation device]
The autonomously moving mobile device includes the route generation device 10 according to the first embodiment or the second embodiment, and a control device that controls the mobile device to move along the movement route generated by the route generation device 10. And comprising.
The control device includes the current position detected by the position detection device 5, the generated movement route (that is, the movement route set in the storage device 3), the command speed value output from the speed command device, the movement It has the above-mentioned speed control device that controls the speed of the mobile device based on the current speed of the device and the direction of travel. The moving device also includes a current position detected by the position detecting device 5, the generated moving route, a commanded speed value output from the speed commanding device (for example, a constant designated speed value), a moving device. And a direction control device for controlling the traveling direction of the moving device based on the current speed and the traveling direction. When the route search device 9 adjusts the command speed value as described above, the speed control device performs speed control of the moving device based on the adjusted command speed value.
Note that the speed command device outputs a command speed value based on the curvature of a portion of the set travel route immediately before the mobile device passes. In this case, the speed command device may output a smaller speed command value as the curvature increases.
Since this moving device can find the moving route more reliably as described above, the moving device can continue to move autonomously.

  The present invention is not limited to the above-described embodiment, and various changes can be made without departing from the scope of the present invention.

  The moving device is a vehicle in the above-described embodiment, but the present invention is not limited to this. For example, the mobile device may be a robot that autonomously travels or autonomously walks. In this case, the above-described embodiment may be changed as appropriate.

  The high-speed driving possibility includes, as its elements, the route length of the moving route candidate, the minimum curvature radius of the moving route candidate, the road surface goodness, and the proximity of the currently used moving route and the moving route candidate. , Or any one of them, or any two or more of them.

  Further, according to the present invention, a route search condition may be obtained by omitting one or more of the above-described interference condition, curvature condition, obstacle detection condition, and generation route selection condition. For example, the interference condition may be omitted if it is known in advance that no obstacle exists, the curvature condition may be omitted as necessary, and the obstacle detection condition is a highly accurate obstacle. When detection is not required, the generation route selection condition may be omitted as necessary.

  Further, according to the present invention, a plurality of moving route candidates may be generated without using the route search condition. That is, in the above-described step S2, a plurality of movement route candidates are generated without considering the route search condition (for example, generated by the above method using the equation (Eq1)), and the generated movement route candidates are However, step S4 described above may be performed. In this case, step S3 described above is omitted.

3 storage device, 5 position detection device, 7 obstacle detection device, 9 route search device, 10 route generation device, 11a, 11b obstacle, 13 travelable region, 15 undetected region, 17 road line, 19, 21 line segment, 23 vibration sensor, 25 road surface judging device

Claims (9)

A route generation device that generates a movement route of a mobile device that moves autonomously,
A storage device that stores a plurality of via positions that are set between the movement start position and the target arrival position and serve as an index for generating the movement path;
A position detecting device for detecting a current position of the moving device;
A route search device that searches and generates a movement route for the mobile device to go to the next passing position among the plurality of via positions based on the current position during movement of the mobile device, and
The route search device
While generating a plurality of moving route candidates, calculating the high-speed driving possibility for each of these moving route candidates,
By comparing these high-speed driving possibilities with each other, a moving route candidate having the highest high-speed driving possibility is selected from the plurality of moving route candidates.
A route generation device, wherein the selected movement route candidate is set as a movement route to be actually used. The high-speed driving possibility includes, as its elements, the route length of the moving route candidate, the minimum radius of curvature of the moving route candidate, the road surface quality based on the height difference of the road surface area along the moving route candidate, and the current use. Including at least one of the proximity of the moving route and the moving route candidate,
The route generation device according to claim 1, wherein the route search device calculates the high-speed driving possibility according to a value of each element.   The route generation device according to claim 2, wherein the high-speed driving possibility is higher as the value of each element is larger. The route search device generates a local map near the current position based on the current position of the mobile device and the obstacle detected by the obstacle detection device provided in the mobile device,
In the local map, the route search device extends a large number of linear search lines from the current position to an undetected area or an obstacle where no obstacle is detected, thereby causing obstacles on both sides. A travelable part of a mobile device sandwiched between objects is identified, the plurality of travel route candidates reaching the travelable part are generated, and a high speed travel possibility is calculated for each of the travel route candidates. The route generation device according to claim 1, 2, or 3. The route search device is a local map,
(A) When there are obstacles on both the left and right sides in the direction toward the passing position that passes next, extract the travelable area of the mobile device that exists between the left and right obstacles;
(B) generating a plurality of inscribed circles located in the travelable area and in contact with the boundaries of the travelable area;
(C) A road line is generated based on a line passing through the center of each inscribed circle,
(D) A plurality of linear search lines extending radially from the current position, and one end search line among a plurality of search lines that are continuously adjacent to each other and reach the undetected region without colliding with an obstacle. The search line adjacent to the left obstacle identifies the collision point as the first collision point, and the search line adjacent to the other search line of the plurality of search lines collides with the right obstacle. The collision point to be identified as the second collision point,
(E) The first collision point and the second collision point are compared, and the one closer to the current position among the two collision points is set as the target collision point,
(F) A line segment extending from the target collision point, orthogonal to the road line, and extending to the obstacle on the opposite side of the target collision point is generated as the travelable location,
(G) The route generation device according to claim 4, wherein the plurality of movement route candidates reaching the line segment are generated, and a high-speed driving possibility is calculated for each of the movement route candidates. The route search device is a local map,
(A) A plurality of linear search lines extending radially from the current position, and one end search line among a plurality of search lines that are adjacent to each other and reach the undetected region without colliding with an obstacle; An adjacent search line identifies a collision point that collides with an obstacle on the left as a first collision point, and a search line adjacent to the other search line of the plurality of search lines collides with an obstacle on the right side. Identify the collision point as the second collision point,
(B) The first collision point and the second collision point are compared, and the one closer to the current position among the two collision points is set as the target collision point,
(C) In the obstacle on the side opposite to the target collision point, a point where the distance from the target collision point is the shortest is set, and a line segment extending from the point to the target collision point is generated as the travelable location. ,
(D) The route generation device according to claim 4, wherein the plurality of movement route candidates reaching the line segment are generated, and a high-speed driving possibility is calculated for each of the movement route candidates. The moving device is a vehicle that travels on a road surface by traveling wheels that are rotationally driven,
The vehicle is equipped with a vibration sensor that measures the vibration of the vehicle when traveling,
A road surface determination device that determines whether the road surface on which the vehicle is currently traveling is an unpaved road surface or a paved road surface based on vibrations measured by the vibration sensor;
The high-speed driving possibility includes not only the road surface goodness as an element thereof but also one or both of the route length and the minimum curvature radius,
The route search device calculates the value of each element by converting it to a standard value, calculates the high-speed driving possibility reflecting the standard value of each element,
The route search device, when it is determined that the road surface is an unpaved road surface for the same moving route candidate, than the case where the road surface is determined to be a paved road surface and the minimum curvature. The route generation device according to claim 1, wherein the standard value of the radius is calculated to be small.   A route generation device according to any one of claims 1 to 7, and a control device that controls the movement device so as to move along the movement route generated by the route generation device. Mobile device to do. A route generation method for generating a movement route of a mobile device that moves autonomously,
A storage device that stores a plurality of via positions that are set between a movement start position and a target arrival position and serves as an index for generating the movement route, and a position detection device that detects a current position of the movement device,
While moving a mobile device, based on the current position, when searching and generating a movement route for the mobile device to go to the next via position among the plurality of via positions,
While generating a plurality of moving route candidates, calculating the high-speed driving possibility for each of these moving route candidates,
By comparing these high-speed driving possibilities with each other, a moving route candidate having the highest high-speed driving possibility is selected from the plurality of moving route candidates.
A route generation method characterized in that the selected travel route candidate is set as a travel route to be actually used.