JP7540280B2 - Autonomous driving device and autonomous driving control device - Google Patents

Description

The present invention relates to an autonomous driving device and an autonomous driving control device.

Conventionally, autonomous driving devices are known that autonomously select a route to travel toward a target location requested by a user based on map information, which is environmental information, and self-location information obtained from a sensor. Autonomous driving devices are used, for example, in factory transport robots, cleaning robots, and guide robots.

The logic for route selection (calculation) in an autonomous driving device is usually such that the route with the shortest travel distance is selected (calculated). In addition, various sensors such as gyro sensors and GPS can be used to obtain self-position information, but since high-precision sensors are expensive, it is desirable to enable autonomous driving even with low-precision sensors.

Map information includes areas where the autonomous driving device can travel and areas where walls and obstacles are placed that make it impossible to travel. When the autonomous driving device travels autonomously, it not only does not enter the impossible area, but also determines a travel route to keep a certain distance from the impossible area so that the device's casing does not come into contact with the impossible area. There are also autonomous driving devices that do not use map information, and the impossible area is identified by, for example, detecting walls and installed objects using sensors.
For example, Patent Documents 1 and 2 describe examples of application to cleaning robots.

JP 2009-95361 A JP 2019-58803 A

Generally, an autonomous driving device is controlled to reach a target position and attitude, but for example, when self-position information is obtained using a low-precision sensor, there is a certain amount of variation in the reached position and attitude compared to the target position and attitude due to errors. For example, when the target position is located near an untravelable area or when the travel route passes near an untravelable area, it is known that the above-mentioned variation and the operation of the device to avoid contact with the untravelable area may cause the device to be unable to move forward or turn, resulting in a state (stuck state) in which effective route generation and command generation are impossible. When the autonomous driving device becomes stuck, not only is it unable to continue operating, but it is also unable to self-return to normal operation, so manual movement of the device and re-setting of the self-position are required to return to normal operation, which is a lot of work.
Therefore, the present invention provides a means for automatically recovering from a stuck state even when the stuck state occurs, and enabling operation to be continued without human intervention.

In order to solve the above problems, one aspect of the autonomous driving device according to the present invention includes a drive unit that drives the vehicle according to a drive instruction, a sensor for obtaining the position of the vehicle, a course determination unit that determines a course to a target position based on the position of the vehicle obtained by the sensor, a drive instruction unit that instructs the drive unit to drive to proceed along the course determined by the course determination unit, a memory unit that stores instructions from the drive instruction unit, a stack determination unit that determines whether the vehicle has become stuck, and a reverse drive instruction unit that reads out the instruction from the memory unit when it is determined that the vehicle has become stuck, and instructs the drive unit to drive in a direction opposite to the drive indicated by the instruction.

According to the above-mentioned autonomous driving device, the drive unit is instructed by the reverse drive instruction unit to drive in a reverse direction along the path (travel route) that led to the stuck state, and the autonomous driving device can return to normal operation from the stuck state. Therefore, even if the autonomous driving device becomes stuck, it can automatically return from the stuck state and continue operating without human intervention.

In the above-mentioned autonomous driving device, it is preferable that the stuck determination unit determines that the vehicle is stuck when a route toward the target position cannot be obtained by combining forward movement and direction changes. Such a stuck determination unit makes it easy to determine whether the vehicle is stuck, and the route that must be followed in reverse to return is a short and simple route, which is expected to result in a more reliable return.

In the above-mentioned autonomous driving device, it is preferable that the stuck determination unit determines that the vehicle is stuck when a part of the circumscribing circle of the vehicle is caught on an obstacle and a course toward the target position cannot be obtained. If a part of the circumscribing circle is caught on an obstacle, free rotation is restricted, so there is a risk of the vehicle becoming seriously stuck if the vehicle continues to travel forcibly. Therefore, by determining that the vehicle is stuck based on the fact that a part of the circumscribing circle is caught on an obstacle, the vehicle can be reliably restored to normal operation.

In addition, in the above-mentioned autonomous driving device, the route determination unit may refer to predetermined map information when determining the route. By referring to the map information, an efficient route can be determined.

Furthermore, in the above-mentioned autonomous driving device, it is preferable that the memory unit stores the above-mentioned instructions up to a predetermined amount, and when an instruction exceeding that amount arrives, the oldest instruction is deleted and a new instruction is stored. This preferable memory unit makes it possible to store information necessary for recovery without wasting memory capacity.

It is also preferable that the autonomous driving device further includes an error notification unit that notifies an error if the determination that the vehicle has become stuck occurs a predetermined number of times or more. By including the error notification unit, an error can be promptly notified if the vehicle becomes stuck repeatedly.

In order to solve the above problem, one aspect of the autonomous driving control device according to the present invention includes a course determination unit that determines a course to a target position based on the position of the vehicle, a drive instruction unit that instructs a drive unit that drives the vehicle to drive along the course determined by the course determination unit, a stack determination unit that determines whether the vehicle has become stuck, and a reverse drive instruction unit that reads out an instruction from the drive instruction unit from a storage unit that has stored the instruction, and instructs the drive unit to drive in a direction opposite to the drive indicated by the instruction, when it is determined that the vehicle has become stuck.

According to the above-mentioned autonomous driving control device, the reverse drive instruction unit can instruct the autonomous driving device to drive in the opposite direction to the path that led to the stuck state, thereby enabling the autonomous driving device to return to normal operation from the stuck state.

According to the present invention, even if an autonomous driving device becomes stuck, it can automatically recover from the stuck state, and the autonomous driving device can continue to operate without human intervention.

1 is a side view showing a transfer robot corresponding to an embodiment of an autonomous driving device of the present invention. 1 is a bottom view showing a transport robot corresponding to an embodiment of the autonomous driving device of the present invention. FIG. FIG. 2 is a functional block diagram showing a functional structure of the transport robot. 13 is a flowchart showing the operation of the transport robot. FIG. 13 is a diagram illustrating a situation in which a command value is input to a storage unit. FIG. 13 is a diagram illustrating a situation in which a command value is read out from a storage unit.

Below, an embodiment of the present invention will be described with reference to the drawings. However, to avoid the following description becoming unnecessarily redundant and to facilitate understanding by those skilled in the art, more detailed description than necessary may be omitted. For example, detailed description of already well-known matters and duplicate description of substantially identical configurations may be omitted. Furthermore, elements shown in previously described figures may be appropriately referenced in the description of subsequent figures.

1 and 2 are external views showing a transport robot corresponding to an embodiment of an autonomous mobile device of the present invention, with Fig. 1 showing a side view and Fig. 2 showing a bottom view.
From the outside, the transport robot 1 comprises a main body 10 , drive wheels 20 and support wheels 30 .

The main body 10 has, for example, a square outer shape, and a loading platform 11 is provided on the upper part. In addition, for example, a proximity sensor 12 is provided at each of the four corners of the main body 10. Note that the placement locations of the proximity sensors 12 are not limited to the four corners, and the proximity sensors 12 may be provided at, for example, two diagonally positioned corners, or at, for example, any one corner. The transport robot 1 of this embodiment is a robot that transports cargo such as materials by placing them on the loading platform 11.

The drive wheels 20 are provided, for example, on the left and right sides of the main body 10, and each drive wheel 20 is driven individually by a motor inside the main body 10, allowing the transport robot 1 to move (i.e., move forward and backward) and change direction.

The support wheels 30 are provided, for example, at the bottom of each of the four corners of the main body 10, and support the weight of the entire transport robot 1. The locations at which the support wheels 30 are provided are not limited to the four corners, and the support wheels 30 may be provided, for example, in a pair at the front and rear along the central axis of the main body 10, or may be provided, for example, at two corners at the front and at one location on the central axis at the rear, or conversely, may be provided, for example, at one location on the central axis at the front and at two corners at the rear.
The support wheels 30 are not connected to a motor or the like, and roll in accordance with the movement and direction changes of the transport robot 1 caused by the drive wheels 20. The axles of the support wheels 30 are held by a rotating holder 31, and the support wheels 30 are structured to freely roll in any direction of 360 degrees along the bottom surface of the main body 10 so as not to impede the movement and direction changes of the drive wheels 20.
FIG. 3 is a functional block diagram showing the functional structure of the transfer robot 1.
The transfer robot 1 functionally comprises a controller 40 , a drive unit 50 , a position sensor 60 , a memory unit 70 , and a communication unit 80 .
The controller 40 is responsible for the overall control of the transfer robot 1 .

The drive unit 50 includes the drive wheels 20, a motor and a power source (not shown), and is responsible for driving the transport robot 1. The drive unit 50 corresponds to an example of a drive unit that drives its own machine according to a drive command, as referred to in the present invention.

The position sensor 60 includes the proximity sensor 12 and a rotation sensor (not shown) of the drive wheel 20, and is responsible for detecting the position and orientation of the transport robot 1. The position sensor 60 may be capable of detecting the absolute position of the transport robot 1, such as a GPS, or may be capable of confirming the relative position by checking the detected value against map information, or may be capable of estimating the position by touch, so to speak, from the detected values of the surroundings without having map information. In this embodiment, the position is confirmed by checking the detected value of the position sensor 60 against map information. The position sensor 60 corresponds to an example of a sensor referred to in the present invention.
The memory unit 70 stores map information 71 and the latest driving instructions 72 .

The communication unit 80 is capable of, for example, wireless communication, and is responsible for receiving setting information of the target position, transmitting error information when travel is impossible, etc. In this embodiment, the target position is given to the transport robot 1 by, for example, communication, but the target position may be one selected by the transport robot 1 from a plurality of candidate positions depending on the situation, etc., or may be an intermediate target position calculated by the transport robot 1 from map information, etc., in order to move toward the final target position.
The controller 40 includes a driving command calculation unit 41 and a drive control unit 42, and corresponds to one embodiment of the autonomous driving control device of the present invention.

The driving command calculation unit 41 confirms the position by comparing the detection value of the position sensor 60 with the map information 71, and calculates a driving route based on the confirmed position and the target position. The driving command calculation unit 41 may only calculate the path as the direction to move forward, but in the transport robot 1 of this embodiment, the driving route as the path that the transport robot 1 should follow is also calculated by the driving command calculation unit 41. By calculating the driving route in this manner, it is possible to appropriately avoid areas where it is impossible to move through, such as walls or installed objects. The driving command calculation unit 41 corresponds to an example of a route determination unit as referred to in the present invention.

The driving command calculation unit 41 inputs a driving command indicating the driving of the drive unit 50 to follow the calculated driving route to the drive control unit 42, and also sends the driving command to the storage unit 70 to store it as the latest driving command 72. The driving of the drive unit 50 indicated by the driving command may be, for example, the rotation speed, the rotation acceleration, or the number of revolutions of the drive wheels 20. The driving command may also indicate the rotation speed of the motor that drives the drive wheels 20, etc.
As will be described later, the travel command calculation unit 41 also determines whether the vehicle is stuck. In other words, the travel command calculation unit 41 also corresponds to an example of the stuck determination unit of the present invention.

The drive control unit 42 controls the drive unit 50 in accordance with the travel command input from the travel command calculation unit 41 to travel the transport robot 1. The drive control unit 42 corresponds to an example of a drive instruction unit according to the present invention.
Next, the operation of the transport robot 1 will be described.
FIG. 4 is a flowchart showing the operation of the transfer robot 1.

When the operation of the transport robot 1 is started, first in step S101 it is determined whether the operation should be continued or ended. If the transport robot 1 has reached the final target position, the travel command calculation unit 41 determines that the operation should be ended, and if the transport robot 1 has not reached the final target position, the travel command calculation unit 41 determines that the operation should be continued.

If it is determined in step S101 that the operation should be continued, an autonomous driving operation is executed in step S102. In the autonomous driving operation, the driving command calculation unit 41 performs the above-mentioned position confirmation and calculation of the driving route to output a driving command, the drive control unit 42 controls the drive unit 50 according to the driving command, and the transport robot 1 travels along the driving route. In this embodiment, the driving error of the drive wheels 20 is small, and the transport robot 1 travels along the driving route almost accurately. However, as a result of small errors accumulating, the driving route may deviate from the original driving route that reaches the original target position or posture. In addition, the driving route may deviate from the target position due to unpredictable unevenness or slippage caused by, for example, dust. Therefore, in this embodiment, the position sensor 60 includes a proximity sensor 12, and the position is appropriately corrected based on the detection value of the wall by the proximity sensor 12, by collating it with the above-mentioned map information, etc.
The command value of the driving command output in the autonomous driving operation in step S102 is stored as the most recent driving command 72 in the memory unit 70 in step S103.

Next, in step S104, the travel command calculation unit 41 determines whether the transport robot 1 is stuck or not. If it is determined that the transport robot 1 is not stuck and is in a normal state, the operations in steps S101 to S103 are repeated, and the determination of the stuck state in step S104 is also repeated.

In this embodiment, if a route toward the target position cannot be obtained by combining forward movement and direction changes, the travel command calculation unit 41 determines that the vehicle is stuck (i.e., stuck).

As described above, in this embodiment, the accumulated error caused by the driving unit 50 is appropriately corrected based on the detection value by the proximity sensor 12, but when the target position is set near the impassable area, there is a risk that the transport robot 1 will get too close to the impassable area while traveling due to a deviation in the traveling path. If the transport robot 1 gets too close to the impassable area, it is dangerous to continue following the traveling path as it is. Therefore, the detection of significant proximity to a wall or the like by the proximity sensor 12 is used as a trigger to recalculate the traveling path by the traveling command calculation unit 41.
2, if a part of the circumscribing circle R of the transport robot 1 is caught by an obstacle such as a wall or an installation, the free rotation of the transport robot 1 is restricted, and even if the travel route is recalculated, a path toward the target position may not be obtained. In such a case, in this embodiment, the travel command calculation unit 41 determines that the robot is stuck, and the process proceeds to step S105, where a return operation is performed.

In the return operation in step S105, a drive operation for retracing the travel path taken by the transport robot 1 up to the stuck state is executed by the drive unit 50. Specifically, the most recent travel command 72 stored in the memory unit 70 is read out by the travel command calculation unit 41, and the travel command calculation unit 41 instructs the drive control unit 42 to perform reverse driving of the drive indicated by the command value of the travel command 72. Therefore, the travel command calculation unit 41 also corresponds to an example of the reverse drive instruction unit referred to in the present invention.
Here, a method for storing command values in the storage unit 70 will be described.

Figures 5 and 6 are diagrams showing the method of storing command values in the memory unit 70. Figure 5 shows the situation in which command values are input to the memory unit 70, and Figure 6 shows the situation in which command values are read out from the memory unit 70.

The storage unit 70 is a so-called first-in, first-out type memory, and a total of, for example, n+1 command values 73 are stored in the storage unit 70 as the most recent driving commands 72 shown in Fig. 3. The command value 73 output from the driving command calculation unit 41 at the latest time t is input (pushed) to the bottom row of the storage unit 70 as shown in Fig. 5, and each command value 73 before time t-1 is moved up one by one to the upper row (i.e., the older side). Then, when the command values 73 have piled up to the top row (i.e., up to time t-n), the oldest command value 73 in the top row is deleted from the storage unit 70 as a new command value 73 is input to the bottom row.
In this manner, the memory unit 70 stores the command values 73 at each time from the latest time t up to time t−n in chronological order.

During the return operation, the command value 73 for the latest time t is read (popped) from the bottom row of the memory unit 70, and each command value 73 before time t-1 is moved down one by one (i.e. to the newer side). Then, the command values 73 for each time up to a maximum of time t-n are read from the memory unit 70 in order of most recent. Each time a command value 73 is read, a drive opposite to the drive represented by each command value 73 is executed, causing the transport robot 1 to retrace its travel path in the opposite direction.

The transport robot 1 can easily get out of a stuck state by reversing its travel path. In this embodiment, the stuck state is determined based on the criterion that a path to the target position cannot be obtained by combining forward movement and direction changes, so reverse driving is performed while the robot is still only slightly stuck, without any turning around, and the robot can be reliably returned to its original state.

When the return operation of step S105 in FIG. 4 is completed, the operation returns to step S101, and the above-mentioned operations are repeated, allowing the transport robot 1 to move toward the target position.

However, depending on the travel path leading to the stuck state, the command value 73 used in the return operation up to time t-n may not be sufficient, and the robot may again be stuck even with a newly calculated travel path. For this reason, in the judgment of step S104, if the robot is repeatedly judged to be stuck a predetermined number of times near the location where the robot is judged to be stuck, an error is judged, and error information is sent from the travel command calculation unit 41 to the communication unit 80, and the error information is sent from the communication unit 80 to the manager or the like. The manager or the like can know from the error information that the transport robot 1 has reached a serious stuck state, and can attempt to recover quickly. The combination of the travel command calculation unit 41 and the communication unit 80 corresponds to an example of the error notification unit referred to in the present invention.
In addition, when the driving route is recalculated in step S102 after recovery from being stuck, it is preferable to set the periphery of the stuck location as a non-drivable area and calculate the driving route in order to prevent re-stuck. Specifically, the map information is updated with the position coordinates where the stuck occurred and an area of a certain range centered on that coordinate as the non-drivable area, and the driving route is calculated based on the updated map information.

In the above explanation, an application example to a transport robot is shown, but the autonomous driving device and autonomous driving control device of the present invention may also be applied to cleaning robots, guide robots, etc.

1 ... transport robot, 10 ... main body, 20 ... drive wheel, 30 ... support wheel, 40 ... controller,
41: Travel command calculation unit, 42: Drive control unit, 50: Drive unit, 60: Position sensor,
70: memory unit, 71: map information, 72: most recent driving command, 73: command value, 80: communication unit

Claims (6)

A drive unit that drives the player's aircraft according to drive instructions;
a sensor for obtaining the position of said aircraft;
a route determination unit that determines a route to a target position based on the position of the aircraft obtained by the sensor;
a drive instruction unit that instructs the drive unit to drive in order to proceed along the path determined by the path determination unit;
A storage unit that stores an instruction from the drive instruction unit;
a stuck determination unit that determines whether the player's own machine has become stuck;
a reverse drive instruction unit that reads out the instruction from the storage unit when it is determined that the stuck state has occurred, and instructs the drive unit to perform a drive reverse to the drive indicated by the instruction ,
The stuck determination unit determines that the vehicle is stuck when a path toward a target position cannot be obtained by a combination of forward movement and direction changes. The autonomous driving device according to claim 1, characterized in that the stuck determination unit determines that the vehicle is stuck when a part of the circumscribing circle of the vehicle is caught on an obstacle and a path toward the target position cannot be obtained. The autonomous driving device according to claim 1 or 2, characterized in that the route determination unit determines the route by referring to predetermined map information. The autonomous driving device according to any one of claims 1 to 3, characterized in that the storage unit stores the instructions up to a predetermined amount, and when an instruction exceeding the amount is received, the oldest instruction is deleted and a new instruction is stored. The autonomous driving device according to any one of claims 1 to 4, further comprising an error notification unit that notifies an error when the determination that the device has become stuck occurs a predetermined number of times or more. a route determination unit that determines a route to a target position based on the position of the aircraft;
a drive instruction unit that instructs a drive unit that drives the own aircraft to drive the own aircraft along the course determined by the course determination unit;
a stuck determination unit that determines whether the player's own machine has become stuck;
a reverse drive instruction unit that, when it is determined that the stuck state has occurred, reads out an instruction from the drive instruction unit from a storage unit that has stored the instruction, and instructs the drive unit to perform a drive reverse to the drive indicated by the instruction ,
The stuck determination unit determines that the vehicle is stuck when a path toward a target position cannot be obtained by a combination of forward movement and direction changes.

Images