JP2020135488A - Mobile body control device and mobile body control program - Google Patents

Abstract

To provide a mobile body control device capable of easily changing a speed pattern.SOLUTION: A track generation device generates a target track for a moving body to move from a start point position to a target position. The track generation device sets a speed pattern for moving on the target track. The speed pattern indicates the change in speed over a movement period. When the speed pattern is integrated over time, the path length of the target track is obtained. The track generation device generates a virtual time corresponding to the actual time used in the speed pattern. The virtual time is a time at which the elapsing speed of the virtual time with respect to the actual time can be reduced, in at least a part of a change period of the movement period of the speed pattern. The track generation device changes the speed pattern during the change period based on the virtual time. The track generation device reduces the speed during the change period and extends the speed pattern during the change period in a time axis direction. Time integration values become equal to each other between the speed patterns before and after the change.SELECTED DRAWING: Figure 1

Description

The techniques disclosed herein relate to mobile control devices and mobile control programs.

Technologies for self-propelling moving objects such as unmanned forklifts have been developed. In order to make the moving body self-propelled, a locus is generated, a speed pattern is designed in advance so as to run on the locus, and the moving body is made to run on the locus according to the speed pattern. Related techniques are disclosed, for example, in Patent Document 1.

Japanese Unexamined Patent Publication No. 2012-173760

When the moving body travels on a predetermined trajectory, the speed may be temporarily suppressed or stopped due to an abnormality or the like. Then, the premise of the previously designed locus and velocity pattern breaks down, and it is necessary to start over from the locus generation and velocity pattern design. The present specification discloses a moving body control device capable of easily changing the speed pattern.

One embodiment of the trajectory generating device disclosed in the present specification is a trajectory generating device that generates a trajectory in which a moving body moves. The trajectory generator includes a trajectory generator that generates a target trajectory for moving from the start point position to the target position. The trajectory generator includes a velocity pattern generator that sets a velocity pattern for moving on the target trajectory. The velocity pattern is a pattern showing a change in velocity over the entire movement period from the start point position to the target position. When the velocity pattern is time-integrated, the path length of the target trajectory is obtained. The trajectory generator includes a virtual time generator that generates a virtual time corresponding to the real time used in the velocity pattern. The virtual time is a time at which the elapsed speed of the virtual time with respect to the real time can be reduced during at least a part of the change period of the moving period of the speed pattern. The trajectory generator includes a speed pattern control unit that changes the speed pattern during the change period based on the virtual time. The speed pattern control unit lowers the speed during the change period from the target speed indicated by the speed pattern before the change, and extends the speed pattern during the change period in the time axis direction. The time integral value of the changed speed pattern becomes equal to the time integral value of the speed pattern before the change.

The virtual time generator can generate a virtual time that corresponds to the real time used in the speed pattern. Then, the speed pattern control unit can change the speed pattern during the change period based on the virtual time. At this time, the speed can be made lower than the target speed indicated by the speed pattern before the change, and the speed pattern during the change period can be extended in the time axis direction. Further, the time integral value of the speed pattern after the change can be made equal to the time integral value of the speed pattern before the change. Therefore, it is possible to accelerate or decelerate the speed during the change period while leaving the speed pattern designed in advance. Even when the speed is temporarily suppressed or stopped, it is not necessary to regenerate the target trajectory or the speed pattern.

The virtual time may be a time integral value of the slope of a straight line indicating the relationship between the real time and the virtual time when the horizontal axis is the real time and the vertical axis is the virtual time. The virtual time generator may reduce the elapsed speed of the virtual time with respect to the real time by reducing the slope of the straight line during the change period. Details of the effect will be described in Examples.

The moving body control device may further include a start time determination unit that determines the start time of the change period. The virtual time generation unit may generate a virtual time in which the elapsed speed with respect to the real time is reduced from the start time. The speed pattern control unit may lower the speed below the target speed after the start time. Details of the effect will be described in Examples.

The moving body control device may further include a detection unit that detects an obstacle on the target trajectory. The start time determination unit may determine the start time of the change period depending on the detection of an obstacle by the detection unit. Details of the effect will be described in Examples.

The moving body control device may further include an end point determination unit that determines the end point of the change period. The virtual time generator may generate a virtual time in which the elapsed speed with respect to the real time is restored from the end point. The speed pattern control unit may return the speed to the target speed after the end point. Details of the effect will be described in Examples.

The virtual time generator may calculate the centripetal acceleration applied to the moving body on the target locus. The change period may be a period during which the calculated centripetal acceleration is equal to or higher than a predetermined threshold value. Details of the effect will be described in Examples.

The moving body may be provided with a handle for steering. The virtual time generator may calculate the steering speed, which is the rotation speed of the steering wheel. The change period may be a period during which the calculated steering speed becomes equal to or higher than a predetermined threshold value. Details of the effect will be described in Examples.

The virtual time generator may be able to make the elapsed speed of the virtual time negative with respect to the real time in a part of the change period. The moving body may back up on the target trajectory during a period in which the elapsed speed of the virtual time with respect to the real time is negative. Details of the effect will be described in Examples.

The control program for realizing the above information processing device is also new and useful.

The block diagram which shows the structure of a forklift. A flowchart illustrating a self-propelled control method. The figure which shows the setting example of a start point position, a target position, and a target locus. The figure which shows an example of the speed pattern in the initial state. A graph showing the relationship between the real time t and the virtual time e. The figure which shows an example of the speed pattern after a change. A graph showing the relationship between the real time t and the virtual time e. A graph showing the relationship between the real time t and the virtual time e. The figure which shows the setting example of a start point position, a target position, and a target locus. The figure which shows an example of a speed pattern. A graph showing an example of centripetal acceleration. The graph which shows an example of the slope λ. The figure which shows an example of a speed pattern. A graph showing the relationship between the real time t and the virtual time e.

(Structure of forklift 10)
Hereinafter, the forklift 10 of this embodiment will be described with reference to the drawings. The forklift 10 shown in FIG. 1 is an unmanned forklift. The forklift 10 includes a vehicle body 12, a range sensor 26, and an arithmetic unit 30.

The vehicle body 12 includes a steering control unit 21, a drive control unit 22, and a wheel rotation speed measuring instrument 23. The steering control unit 21 and the drive control unit 22 are composed of a microprocessor including a CPU and the like. The steering control unit 21 is communicatively connected to a steering device (not shown) and controls the traveling direction of the forklift 10. The drive control unit 22 is communicatively connected to a drive wheel motor (not shown) and controls the traveling speed of the forklift 10. The wheel rotation speed measuring instrument 23 is connected to a front wheel (not shown), and can detect the amount of rotation of the front wheel. The wheel rotation speed measuring instrument 23 detects the moving amount and the moving direction of the vehicle body 12 based on the movement of the front wheels.

The range sensor 26 is a one-dimensional scanning type range sensor that is installed on the vehicle body 12 and scans the laser beam in one direction (horizontal direction in this embodiment). The range sensor 26 measures the distance between the vehicle body 12 and the object in the set space set in front of the vehicle body 12. As a result, the distance data in front of the forklift 10 is acquired. The distance data acquired by the range sensor 26 is input to the self-position estimation unit 38, the start time determination unit 43, and the end time determination unit 44. Further, the range sensor 26 has a function of detecting an obstacle on the target trajectory TT. The obstacle detection result is input to the start time point determination unit 43 and the end time point determination unit 44.

The arithmetic unit 30 is composed of a microprocessor including a CPU and the like. The arithmetic unit 30 includes a start point position / target position setting unit 35, a moving body control device 40, map information 58, a self-position estimation unit 38, and a control command generation unit 48. The map information 58 is information indicating the position and size of an object (for example, a pillar, a wall, etc.) existing in the space where the forklift 10 moves. The map information 58 can be created, for example, by measuring an object in the area where the forklift 10 moves in advance with the range sensor 26 and based on the measurement result.

The map information 58, the distance data between the peripheral objects measured by the range sensor 26, and the movement amount and movement direction of the vehicle body 12 detected by the wheel rotation speed measuring instrument 23 are input to the self-position estimation unit 38. To. The self-position estimation unit 38 estimates the self-position of the forklift 10 based on these input data. The start point position / target position setting unit 35 sets the start point position PS and the target position PE, which will be described later.

The moving body control device 40 includes a locus generation unit 41, a speed pattern generation unit 42, a start time point determination unit 43, an end time point determination unit 44, a virtual time generation unit 45, and a speed pattern control unit 46. The contents of the locus generation unit 41 to the speed pattern control unit 46 will be described later.

(Self-propelled control of forklift 10)
The self-propelled control method of the forklift 10 will be described with reference to the flowchart of FIG. In step S1, the start point position / target position setting unit 35 sets the start point position PS and the target position PE. In addition, the starting point posture and the target posture are set. The start point posture and the target posture are the traveling directions of the forklift 10 at each of the start point position PS and the target position PE. FIG. 3 shows a setting example of the start point position PS and the target position PE. FIG. 3 describes a case where the xy coordinates of the start point position PS and the target position PE are the same at (0,0). That is, in the example of FIG. 3, the start position and the goal position are the same.

In step S2, the locus generation unit 41 generates a target locus TT for moving from the start point position PS to the target position PE. It is possible to generate the target locus TT by calculating the total curve length from the start point position PS to the target position PE and the coordinates of the points having a predetermined curve length from the start point position PS. There may be various ways to obtain the curve length. The length of the approximate curve including a slight error with respect to the real curve may be used as the approximate value of the curve length. In the example of FIG. 3, a target locus TT is generated starting from the start point position PS and reaching the target position PE via the transit positions P1 to P5 in order.

In step S3, the velocity pattern generation unit 42 sets a velocity pattern for moving on the target locus TT. The velocity pattern is a pattern showing a change in velocity over the entire movement period from the start point position PS to the target position PE. The velocity pattern is generated so that the velocity and the angular velocity become zero at the target position PE. An example of the velocity pattern VP designed for the target locus TT of FIG. 3 is shown in FIG. The vertical axis is the moving speed of the forklift 10, and the horizontal axis is the real time. The integrated value (shaded portion) of the velocity pattern VP is the path length from the start point position PS to the target position PE.

In step S4, the virtual time generation unit 45 generates a virtual time corresponding to the real time. The virtual time is always generated corresponding to the real time. The virtual time is a time integral value of the slope of a straight line indicating the relationship between the real time and the virtual time when the horizontal axis is the real time and the vertical axis is the virtual time.

The real time and the virtual time match in the period until the start command (S6) of the change period described later is output. In other words, as shown in the graph of FIG. 5, the relationship between the real time t and the virtual time e is represented by a straight line having a slope of 1. Since the virtual time is the integral value of the slope of the straight line between the real time t and the virtual time e, it is expressed by the following equation (1).
Since the equation (1) is an equation in the period before the start command (S6) of the change period is issued, λ (t) is 1. The speed target value du / dt at this time is obtained by the following equation (2).
Here, u is the curve length or an approximate value thereof. The curve length u and the points (x, y) representing the curve have a relationship represented by a polynomial such as the following equation (3).
Here, m is a scaling parameter for similarity transformation. In this embodiment, m is 1.

Since du / de is defined by the following equation, du / de can be obtained by an algebraic expression (an expression of only addition, subtraction, multiplication, division and exponentiation) if the above relationship is established.
The angular velocity target value can be obtained by the following equation (4).
Where κ is the curvature. The speed target value obtained by the equation (2) and the angular velocity target value obtained by the equation (4) can be used as feed forward values when controlling the movement of the forklift 10.

In step S5, the movement of the vehicle body 12 is started. Specifically, the speed pattern control unit 46 outputs the speed pattern VP to the control command generation unit 48. In the period until the start command (S6) of the change period described later is output, the speed pattern VP in the initial state that has not been changed is output. The control command generation unit 48 sets the target locus TT generated by the locus generation unit 41, the speed pattern VP input from the speed pattern control unit 46, and the current position of the forklift 10 calculated by the self-position estimation unit 38. Based on this, the control command value for moving the vehicle body 12 to the target position PE is calculated. The generated control command value is input to the steering control unit 21 and the drive control unit 22.

In step S6, the virtual time generation unit 45 determines whether or not the start command of the change period CT has been received from the start time determination unit 43. The change period CT is a period during which the speed pattern VP is changed. The change period CT start command is a command output in response to the start time determination unit 43 determining the change start of the speed pattern VP. For example, when an abnormality occurs (eg, when the range sensor 26 notifies that an obstacle has been detected), it may be decided to start changing the speed pattern VP. If a negative judgment is made (S6: NO), the process proceeds to step S10, and if a positive judgment is made (S6: YES), the process proceeds to step S7.

In step S7, the virtual time generation unit 45 changes the virtual time. Specifically, by reducing the slope of the straight line indicating the relationship between the real time t and the virtual time e, the elapsed speed of the virtual time e with respect to the real time t is reduced. The speed pattern control unit 46 changes the speed pattern based on the changed virtual time, and outputs the changed speed pattern to the control command generation unit 48.

A specific example will be described. The solid line in FIG. 6 shows the speed pattern VP in the initial state, which is the same pattern as in FIG. The dotted line in FIG. 6 shows the changed speed pattern VPa. In the speed pattern VP in the initial state, the case where the “change period start command” for increasing the speed by 0.3 times is generated at the actual time t1 (13.5 s), which is the start time of the change period CT, will be described. As shown in FIG. 7, the relationship between the real time t and the virtual time e after the real time t1 is changed to a straight line LL1 having a slope of 0.3. That is, the elapsed speed of the virtual time e with respect to the real time t can be reduced. Therefore, the virtual time corresponding to the current time can be calculated by the following equation (5).
Here, λ (t) is 0.3. Using the obtained virtual time, the velocity target value and the angular velocity target value are obtained by the equations (2) and (4).

As a result, the speed pattern control unit 46 changes the speed pattern based on the virtual time. Specifically, as shown in the period TE1 from the real time t1 to t2 in FIG. 6, the speed pattern VPa (dotted line) after the change is higher than the target speed shown by the speed pattern VP (solid line) before the change. Can be reduced.

In step S8, the virtual time generation unit 45 determines whether or not the end command of the change period CT has been received from the end time determination unit 44. The end command of the change period CT is a command output in response to the decision of the end time determination unit 44 to restore the speed pattern VP. For example, when the abnormality is resolved (eg, when the range sensor 26 notifies that the obstacle is no longer detected), the end of changing the speed pattern VP may be determined. If a negative judgment is made (S8: NO), the process returns to step S8 and waits, and if a positive judgment is made (S8: YES), the process proceeds to step S9.

In step S9, the virtual time generation unit 45 restores the virtual time. Specifically, by returning the slope of the straight line indicating the relationship between the real time t and the virtual time e to "1", the elapsed speed of the virtual time e with respect to the real time t is made the same. The speed pattern control unit 46 restores the speed pattern based on the virtual time, and outputs the returned speed pattern to the control command generation unit 48.

A specific example will be described. In the changed speed pattern VPa (dotted line) of FIG. 6, the case where the “end command of the change period” for returning the speed to the 1x speed is generated at the actual time t2 (16s) will be described. As shown in FIG. 8, the relationship between the real time t and the virtual time e after the real time t2 is changed to the straight line LL2 having the slope of 1. Therefore, the virtual time corresponding to the current time can be calculated by the following equation (6).
Here, λ (t) is 1. Using the obtained virtual time, the velocity target value and the angular velocity target value are obtained by the equations (2) and (4).

As a result, the speed pattern control unit 46 restores the speed pattern based on the virtual time. Specifically, in the changed speed pattern VPa (dotted line) in FIG. 6, the speed pattern shape in the period TE2 after the real time t3 is the same as the speed pattern VP (solid line) before the change. That is, as shown in FIG. 6, by inserting the period in which the speed pattern is changed (change period CT) into the speed pattern VP (solid line) in the initial state, the subsequent speed pattern VP (solid line) becomes the virtual time. The velocity pattern VPa (dotted line) after the change is obtained by shifting in the time axis direction Y1. The changed speed pattern and the speed pattern shifted to the time axis direction Y1 which is the future direction are always connected at the connection point CP at the real time t3. This is because the speed pattern is changed by adjusting the time axis. Further, the time integral value of the changed speed pattern VPa (shaded portion in FIG. 6) and the time integrated value of the speed pattern VP in the initial state (shaded portion in FIG. 4) are equal to each other.

Explain the effect. It is possible to accelerate or decelerate the speed during the change period CT while leaving the speed pattern VP (solid line) in the initial state designed in advance. It is possible to temporarily suppress or stop the speed without having to regenerate the target trajectory TT or the speed pattern.

In step S10, the control command generation unit 48 determines whether or not the target position PE has been reached. If a negative judgment is made (S10: NO), the process returns to step S6, and if a positive judgment is made (S10: YES), the flow ends.

In the first embodiment, a mode in which the speed pattern is changed (S7) in response to the start command (S6) of the change period has been described. In the second embodiment, a mode in which the speed pattern is changed in advance at a place where there is a risk of falling or the like (eg, a sharp curve portion) will be described using a pre-designed target locus TT.

The parts different from the flowchart of the first embodiment (FIG. 2) in the second embodiment will be described below. In step S1, the start point position / target position setting unit 35 sets the start point position PS and the target position PE. In step S2, the locus generation unit 41 generates the target locus TT. FIG. 9 shows a setting example of the start point position PS, the target position PE, and the target locus TT. FIG. 9 describes a case where the xy coordinate of the start point position PS is (0,0) and the xy coordinate of the target position PE is (9,3).

In step S3, the speed pattern generation unit 42 sets the speed pattern in the initial state for moving on the target locus TT. FIG. 10 shows a specific example of the velocity pattern designed for the target locus TT of FIG. The solid line in FIG. 10 shows the speed pattern VP1 in the initial state, and the broken line shows the speed pattern VP1a after the change.

In step S4, the virtual time generation unit 45 generates a virtual time corresponding to the real time. At this time, in the second embodiment, the virtual time is generated so as to limit the centrifugal force on the sharp curve. This will be described below.

In two dimensions in the xy plane, the curvature of the path is κ and the radius of curvature 1 / κ = r. Since the velocity v is tangential to the path, the centripetal acceleration a is orthogonal to the velocity v. Since the velocity v = ds / dt, the centripetal acceleration a can be obtained by the following equation (7).

It is assumed that the maximum value a _max of the absolute value of the centripetal acceleration a is given by a user or an external device (eg, a load determination device). The maximum value a _max may be determined from the viewpoint of slip prevention and load collapse prevention. In this case, the value of the slope λ can be obtained as shown in the following equation (8).

When the polynomial of the equation (3) is used, ds / du is almost 1, so it can be ignored. Therefore, the equation (8) can be simplified as in the following equation (9).

Curvatures κ and ds / du can be considered smooth enough for virtual time e. Also, since du / de is also a speed pattern designed by the user, it can be regarded as sufficiently smooth. Therefore, the slope λ also changes smoothly with time.

FIG. 11 shows a graph of the centripetal acceleration a calculated using the equation (7). The solid line in FIG. 11 shows the centripetal acceleration pattern AP corresponding to the velocity pattern VP1 (FIG. 10) in the initial state. The dotted line in FIG. 11 shows the centripetal acceleration pattern APa corresponding to the changed velocity pattern VP1a. Further, FIG. 11 describes a case where the maximum value a _max of the centripetal acceleration is 0.8. In the centripetal acceleration pattern AP corresponding to the velocity pattern VP1 in the initial state, the centripetal acceleration a exceeds the maximum value a _max in the period TR1 (left curve period in FIG. 9) and the period TR2 (right curve period). You can see that.

FIG. 12 shows a graph of the slope λ calculated using the equation (9). The periods TR1a and TR2a in FIG. 12 correspond to the periods TR1 and TR2 in FIG. 11 and the modification periods CT1 and CT2 in FIG. In the period TR1a and the period TR2a of FIG. 12, the slope λ is changed so as to be smaller than 1. As described above in the first embodiment, the elapsed speed of the virtual time e with respect to the real time t can be reduced as the slope λ becomes smaller than 1. Therefore, as shown in the changed speed pattern VP1a (dotted line in FIG. 10), the speed pattern can be changed so as to reduce the speed during the change periods CT1 and CT2. As a result, in the centripetal acceleration pattern APa (dotted line in FIG. 11) corresponding to the changed velocity pattern VP1a, the centripetal acceleration a can be suppressed so as not to exceed the maximum value a _max . Since the slope λ changes continuously in FIG. 12, the speed also changes continuously during the change periods CT1 and CT2 of the speed pattern VP1a. Therefore, the control command generation unit 48 can sufficiently follow the speed pattern VP1a.

(effect)
Since the target trajectory TT is designed in advance, it is possible to grasp the sharp curve portion where there is a risk of falling or the load collapse as prior information. In the technique of the second embodiment, a speed pattern that reduces the speed at a sharp curve portion can be created in advance using the target locus TT. This makes it possible to correct the velocity pattern without having to regenerate the target trajectory TT or the velocity pattern.

In the third embodiment, a mode in which the speed pattern is changed in advance at a place where the steering speed becomes high (example: a place where the lane is changed to take the target object (pallet)) is described by using the target locus TT designed in advance. .. The forklift 10 is provided for steering (not shown). The steering speed is the rotation speed of the steering wheel.

The parts that differ from the second embodiment will be described below. In step S4, the virtual time generation unit 45 generates a virtual time corresponding to the real time. At this time, in the third embodiment, the virtual time is generated so as to limit the steering speed. This will be described below.

Let Φ be the angle of the steering wheel and κ be the curvature of the path. The origin is the center of the non-steering wheels (front wheels of the forklift 10). When the forklift 10 is viewed from above, the traveling direction of the vehicle is the x-axis, and the left side with respect to the traveling direction is the y-axis. The position vector of the center of the steering wheel and _{(l x, l y, 0} ). In this case, the steering speed dΦ / dt is obtained by the following equation (10).

It is assumed that the maximum value (dΦ / dt) _max of the absolute value of the steering speed is given by the user or an external device. In this case, the value of the slope λ can be obtained as shown in the following equation (11). As described above, this slope λ changes smoothly.

(effect)
Since the target trajectory TT is designed in advance, it is possible to grasp in advance the portion where the steering speed exceeds the upper limit. In the technique of the third embodiment, a speed pattern in which the steering speed does not exceed the upper limit at the grasped portion can be created in advance using the target locus TT. This makes it possible to correct the velocity pattern without having to regenerate the target trajectory TT or the velocity pattern.

In the fourth embodiment, a mode in which the back running is possible by setting the slope of the straight line between the real time t and the virtual time e to be negative will be described. The parts different from the flowchart of the first embodiment (FIG. 2) in the fourth embodiment will be described below. In step S1, the start point position PS and the target position PE are set. In step S2, the target locus TT is generated. In the fourth embodiment, a case where the start point position PS, the target position PE, and the target locus TT as shown in FIG. 9 are set will be described.

In step S3, the speed pattern generation unit 42 sets the speed pattern in the initial state for moving on the target locus TT. FIG. 13 shows a specific example of the velocity pattern designed for the target locus TT of FIG. The solid line in FIG. 13 shows the speed pattern VP2 in the initial state, and the broken line shows the speed pattern VP2a after the change.

In step S6, when the virtual time generation unit 45 receives the "change period start command" instructing the vehicle to travel backward, the virtual time generation unit 45 in step S7 has a relationship between the real time t and the virtual time e. Change the slope λ of the straight line indicating that to “negative”. As a result, the elapsed speed of the virtual time e with respect to the real time t can be made "negative" (that is, the time is reversed). During the period when the elapsed speed of the virtual time e with respect to the real time t is “negative”, the vehicle body 12 can be back-traveled on the target trajectory TT.

A specific example will be described. The relationship between the real time t and the virtual time e corresponding to the speed pattern of FIG. 13 is shown in FIG. In the period from time t10 to t11, which is the start time, the virtual time and the real time match. Therefore, the slope λ in FIG. 14 is 1, and in FIG. 13, the velocity patterns VP2 and VP2a coincide with each other.

At time t11 (FIG. 14), when the "start command of the change period" instructing the inclination λ to be -1 (that is, running backward) is input, the inclination λ starts to decrease. The rate of change of the slope λ can be determined by a constant given in advance. Then, since the slope λ becomes “negative” after the time t12, the vehicle travels backward after the time t12 as shown in the speed pattern VP2a (FIG. 13).

When the "start command of the change period" instructing to set the slope λ to 0 (that is, to stop) is input at the time t13, the slope λ becomes “0” after the time t14 as shown in FIG. Become. Therefore, as shown in the speed pattern VP2a (FIG. 13), the vehicle stops after the time t14.

At time t15, when the “change period start command” instructing the inclination λ to be 1 (that is, to move forward) is input, the inclination λ starts to increase as shown in FIG. Therefore, as shown in the velocity pattern VP2a (FIG. 13), the vehicle starts moving forward at time t15. The slope λ (FIG. 14) becomes “1” after time t16. Therefore, as shown in FIG. 13, the shape of the speed pattern VP2a (dotted line) after the time t16 is the same as the latter half of the speed pattern VP2 (solid line).

(effect)
By setting the slope λ of the straight line between the real time t and the virtual time e to be “negative”, the virtual time e can be reversed. As a result, it is possible to drive back on the target locus TT without having to regenerate the target locus TT and the speed pattern.

Although the examples of the techniques disclosed in the present specification have been described in detail above, these are merely examples and do not limit the scope of claims. The techniques described in the claims include various modifications and modifications of the specific examples illustrated above.

(Modification example)
The techniques described herein can be applied to the overall task of "running on a predetermined trajectory and zeroing speed at the end point". Therefore, the technique of the present specification can be applied to, for example, the technique of automatic parking by an autonomous vehicle. In this case, as an example of the trigger for changing the speed pattern, there is an emergency stop or deceleration for collision prevention. The technique of the present specification can also be applied to, for example, the technique of an autonomous driving bus. In this case, as an example of the trigger for changing the speed pattern, there is a stop instruction when the signal turns red, a stop instruction at the bus stop when there is a person waiting at the bus stop, and the like.

In this embodiment, the case where the speed pattern is changed so as to be lower than the target speed has been described, but the present invention is not limited to this mode, and the speed may be higher than the target speed. For example, in the technology of an autonomous driving bus, there is a form in which a vehicle passes through an intersection by accelerating when the signal turns yellow.

In the first embodiment, the case where the start command (S6) of the change period is issued once has been described, but the present invention is not limited to this mode. The techniques herein are applicable even when the start command of the change period is issued multiple times.

It is said that the moving body control device 40 is provided in the arithmetic unit 30 of the forklift 10, but the present invention is not limited to this form. When the control program is read by a computer (CPU) included in the information processing device such as a server, the information processing device may exhibit the same function as the mobile control device 40. Then, the forklift 10 may be driven by the information processing device communicating with the forklift 10.

In this embodiment, a one-dimensional scanning type range sensor is used as the range sensor 26, but various sensors may be used. For example, the range sensor 26 may be a 2D LiDAR, a 3D LiDAR, a stereo camera, a monocular camera, a distance image sensor, or the like. The self-position estimation unit 38 may estimate the self-position by means using satellite positioning.

Equation (3) is an example of a parametric curve. In the technique of this embodiment, it is possible to use various types of parametric curves.

The techniques described herein are not limited to techniques for self-propelling moving objects without guides, but are also applicable to self-propelled moving objects with guides (eg, magnetic guides, rails). In particular, even with a guide, it is effective when there is a positioning application.

In step S2, the target locus TT generated by the locus generation unit 41 is not limited to the locus generated on the software for guideless use. The target locus TT may be a physically installed locus such as a magnetic guide.

The technical elements described in the present specification or drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the techniques illustrated in the present specification or drawings achieve a plurality of objectives at the same time, and achieving one of the objectives itself has technical usefulness.

10: Forklift 12: Body 40: Moving body control device 41: Trajectory generation unit 42: Speed pattern generation unit 43: Start time determination unit 44: End time determination unit 45: Virtual time generation unit 46: Speed pattern control unit 48: Control Command generator e: Virtual time t: Real time

Claims (9)

It is a moving body control device that controls the movement of the moving body.
A locus generator that generates a target locus for moving from the start point position to the target position,
A speed pattern generator that sets a speed pattern for moving on the target trajectory.
The velocity pattern is a pattern showing a change in velocity over the entire movement period from the start point position to the target position, and when the velocity pattern is time-integrated, the path length of the target locus is obtained. Generation part and
A virtual time generator that generates a virtual time corresponding to the real time used in the speed pattern.
The virtual time is a time at which the elapsed speed of the virtual time with respect to the real time can be reduced in at least a part of the change period of the movement period of the speed pattern.
A speed pattern control unit that changes the speed pattern during the change period based on the virtual time.
The speed pattern control unit, which reduces the speed during the change period from the target speed indicated by the speed pattern before the change and extends the speed pattern during the change period in the time axis direction.
With
A moving body control device in which the time integral value of the speed pattern after the change becomes equal to the time integral value of the speed pattern before the change. The virtual time is a time integral value of the slope of a straight line indicating the relationship between the real time and the virtual time when the horizontal axis is the real time and the vertical axis is the virtual time.
The moving body control device according to claim 1, wherein the virtual time generation unit reduces the elapsed speed of the virtual time with respect to the real time by reducing the inclination of the straight line during the change period. The moving body control device further includes a start time determination unit for determining the start time of the change period.
The virtual time generation unit generates the virtual time in which the elapsed speed with respect to the real time is reduced from the start time.
The moving body control device according to claim 1 or 2, wherein the speed pattern control unit reduces the speed to be lower than the target speed after the start time. The moving body control device further includes a detection unit that detects an obstacle on the target trajectory.
The moving body control device according to claim 3, wherein the start time determination unit determines the start time of the change period in response to the detection of an obstacle by the detection unit. The moving body control device further includes an end time determination unit for determining the end time of the change period.
The virtual time generation unit generates the virtual time in which the elapsed speed with respect to the real time is restored from the end time.
The moving body control device according to any one of claims 1 to 4, wherein the speed pattern control unit returns the speed to the target speed after the end time. The virtual time generation unit calculates the centripetal acceleration applied to the moving body on the target locus, and claims that the period during which the calculated centripetal acceleration becomes equal to or higher than a predetermined threshold value is defined as the change period. Item 2. The moving body control device according to any one of Items 1 to 5. The moving body is provided with a handle for steering.
The virtual time generation unit calculates the steering speed, which is the rotation speed of the handle, and sets the period during which the calculated steering speed becomes equal to or higher than a predetermined threshold value as the change period, according to claims 1 to 6. The moving body control device according to any one item. The virtual time generator can make the elapsed speed of the virtual time negative with respect to the real time in a part of the change period.
The moving body control device according to any one of claims 1 to 7, wherein the moving body backs up on the target locus during a period in which the elapsed speed of the virtual time with respect to the real time is negative. A control program that controls the movement of a moving object that is read into the computer of an information processing device.
The control program
A locus generation means for generating a target locus for moving from a start point position to a target position,
A speed pattern generating means for setting a speed pattern for moving on the target trajectory.
The velocity pattern is a pattern showing a change in velocity over the entire movement period from the start point position to the target position, and when the velocity pattern is time-integrated, the path length of the target locus is obtained. Generation means and
A virtual time generation means that generates a virtual time corresponding to the real time used in the speed pattern.
The virtual time generation means and the virtual time generation means, wherein the virtual time is a time at which the elapsed speed of the virtual time with respect to the real time can be reduced in at least a part of the change period of the movement period of the speed pattern.
A speed pattern control means for changing the speed pattern during the change period based on the virtual time.
The speed pattern control means and the speed pattern control means for lowering the speed during the change period from the target speed indicated by the speed pattern before the change and extending the speed pattern during the change period in the time axis direction.
To make the information processing device function
A control program in which the time integral value of the changed speed pattern is equal to the time integral value of the speed pattern before the change.