WO2024166711A1 - Information processing method, information processing device, and program - Google Patents

Abstract

The information processing method according to the present disclosure receives target values for a robot and a person in a shared facility that is shared by the robot and the person, and creates a robot behavior plan that satisfies the received target values, on the basis of costs including an occupancy cost for the robot to occupy the shared facility.

Description

Information processing method, information processing device, and program

This disclosure relates to an information processing method, an information processing device, and a program.

Many robots are used in factories. In factories, people frequently work in various places for equipment maintenance, inspection, product inspection, etc., making it difficult to carry out manufacturing entirely with robots.

As a result, humans and robots must share and move around in the same space. Efficient control of elevators is particularly difficult when humans and robots are mixed together. Giving priority to robots reduces human work efficiency, and giving priority to humans means that the robots' task plans cannot be carried out effectively.

　Conventionally, there is technology that creates a robot's behavior plan by (1) always giving priority to humans, (2) always giving priority to robots, or (3) using certain rules to strike a balance (for example, if a human has been waiting for more than one minute, it switches to human priority).

For example, Patent Document 1 discloses a system that controls multiple shared elevators, in which when a robot is to use the elevator, an interrupt process runs, and the most suitable elevator is selected, reserved for the robot, and operated.

Japanese Patent Application Publication No. 4-32471

However, prioritizing humans means that robots cannot use equipment, slowing down task completion. Conversely, prioritizing robots means that humans cannot use equipment, affecting human work.

The present disclosure has been made in consideration of the above problems, and provides an information processing method, information processing device, and program that can improve the work efficiency of both humans and robots when there is equipment that is used jointly by both humans and robots.

The information processing method disclosed herein receives target values for the robot and the person in a shared facility shared by the robot and the person, and creates a behavior plan for the robot based on costs that satisfy the received target values and include an occupancy cost that represents a penalty for the robot occupying the shared facility, the target values being target values that include the robot that is not the target of the behavior plan.

FIG. 1 is a diagram illustrating a first example of an entire system according to an embodiment. FIG. 13 is a diagram illustrating a second example of the entire system according to the embodiment. 1 is a diagram for explaining a path plan for a plurality of robots according to an embodiment; FIG. 13 is a diagram for explaining the problem of the movement of a robot in an elevator according to an embodiment. FIG. 13 is a diagram for explaining the problem of the movement of a robot in an elevator according to an embodiment. 10 is a diagram for explaining a flow of parameter adjustment according to an embodiment. FIG. FIG. 13 is a diagram showing an example in which tasks assigned to a robot and a person according to an embodiment have priorities. FIG. 13 is a diagram showing an initial state before parameter adjustment of an evaluation score monitor in the embodiment. 13 is a diagram showing a state after a predetermined time has elapsed from the initial state before parameter adjustment of the evaluation score monitor in the embodiment. FIG. FIG. 13 is a diagram showing the settings and evaluation score monitor of a robot during parameter adjustment by a central server in an embodiment. A figure showing the robot settings and evaluation score monitor after the central server in the embodiment has completed parameter adjustment. 11A and 11B are diagrams for explaining a pattern in which the central server in the embodiment recognizes a robot made by another company as an unknown moving object using a camera. FIG. 2 is a functional block diagram of a central server according to an embodiment. FIG. 13 is a diagram showing an example of a data input source input to an evaluation score data input unit of the central server according to the embodiment. 1 is a diagram for explaining a relationship between a central server and a robot made by another company according to an embodiment. FIG. 1 is a diagram for explaining a relationship between a central server and a robot made by another company according to an embodiment. FIG. 10 is a flowchart illustrating an operation of a central server according to an embodiment. 10 is a flowchart illustrating a flow of a task execution process of a central server according to an embodiment. 13A and 13B are diagrams illustrating a comparison of people's waiting times before and after parameter adjustment of occupancy cost according to an embodiment. FIG. 2 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the central server according to the embodiment.

Below, a preferred embodiment of the present disclosure will be described in detail with reference to the attached drawings. Note that in this specification and drawings, components having substantially the same functional configuration will be assigned the same reference numerals to avoid duplicated explanations. The explanation will be given in the following order.

1. Prerequisite system 2. Overall system configuration 3. Example of robot 103 plan 4. Issues with elevator 101 5. Overview of embodiment 6. Occupancy cost, etc. 6.1. Occupancy cost 6.2. Example of occupancy cost 6.3. Changing the weight of occupancy cost 6.4. Parameter adjustment flow 6.5. Trigger for parameter adjustment 6.6. Status (metrics) confirmation 6.7. Parameter adjustment 6.8. Case where priority exists within a robot 103 task 6.9. Case where priority exists within a person 104 task 6.10. Example where a task has priority 6.11. Example of measuring the task efficiency of a person 104 6.12. Parameter adjustment flow (overall)
7. Functional block diagram of the central server 100 according to the embodiment 7.1. Relationship between the central server 100 according to the embodiment and a robot system 613 made by another company 8. Flowchart for explaining the overall operation of the central server 100 according to the embodiment 9. Flowchart for explaining the operation of the central server 100 according to the embodiment during task execution 10. Settings and UI
11. Use Cases 12. Effects 13. Hardware Configuration

＜1. Prerequisite system＞
The system on which the embodiment is based is a system that uses multiple robots to perform some task, such as transporting items in a factory or transporting goods in a logistics warehouse.

The system can freely operate the robots that are the objects of the system's control. The system can also freely operate the equipment used by the robots. For example, the system can operate elevators and open automatic doors. The system's equipment is used simultaneously by (multiple) people. Human behavior is unknown, and it is not known when the equipment will be used.

2. Overall system configuration
Fig. 1 is a diagram showing a first example of the entire system according to an embodiment. As shown in Fig. 1, the facility includes a plurality of robots 103 that are subject to the control of a central server 100, and an elevator 101 that is a shared facility. A plurality of people 104 are present near the elevator 101. The movements of the people 104 are not managed by the central server 100.

The robot 103 is a robot that performs a task. Multiple robots 103 may each perform a different task. Multiple people 104 may also perform tasks, each performing a different task. The robot 103 transmits its current location to the central server 100. Also, the robot 103 performs actions such as moving based on movement instructions from the central server 100.

The elevator 101 is a shared facility between the robot 103 and the person 104, and has an elevator controller 102. The elevator controller 102 is a device that can operate the elevator 101, etc. The elevator controller 102 controls the operation of the elevator 101 based on instructions from the central server 100.

The central server 100 generates a new plan that is an overall optimization based on the current position of the robot 103, the current plan, and the state of the elevator 101. The plan also includes operations for the elevator 101. For example, elevator operations include when to call which elevator and for which robot. Based on the generated plan, the central server 100 issues movement instructions to the robot 103 and button operation instructions to the elevator controller 102 of the elevator 101. The robot 103 moves according to the operation instructions from the central server 100. In response to the operation instructions from the central server 100, the elevator controller 102 outputs the elevator button pressing status, etc. to the central server 100.

FIG. 2 is a diagram showing a second example of the entire system according to the embodiment. In addition to the configuration shown in FIG. 1, FIG. 2 shows an example in which a human presence sensor 402 is provided. The same parts as in FIG. 1 are given the same reference numerals and their explanations are omitted.

The human presence sensor 402 sends information about people 104, such as the number of people waiting in the elevator 101, to the central server 100. The human presence sensor 402 may be a camera, etc.

3. Example of Plan for Robot 103
Next, an example of a plan for multiple robots 103 will be described.

Each robot 103 has a goal, and wants to reach that goal in the shortest time possible. However, if the shortest route is given without any thought, the robots may collide or deadlock, preventing them from reaching the goal. For this reason, it is necessary to use rules and algorithms to create a route that will avoid collisions and deadlocks. Therefore, the central server 100 creates a plan that can minimize the total time (= cost) it takes for the robots 103 to reach the goal. The plan that minimizes the cost is called the optimal plan.

FIG. 3 is a diagram for explaining path planning for multiple robots 103 according to an embodiment. In FIG. 3, three paths 200a, 200b, and 200c for the robots 103 are shown, each of which is given a reference number, and the central server 100 creates a plan for one of these paths 200a, 200b, and 200c, which has the lowest cost and does not cause collisions or deadlocks between the robots 103.

<4. Issues with Elevator 101>
Next, the problem of the movement of the robot 103 in the elevator 101 will be described.

FIG. 4 is a diagram for explaining the problem of the movement of the robot 103 in the elevator 101 according to the embodiment. For example, as shown in FIG. 4, assume that a person 104 who wants to go to the 2nd floor is waiting for the elevator 101 on the 1st floor. Assume that there are three robots 103 on the 2nd floor who want to go to the 1st floor to perform a task.

If we only consider the efficiency of the robots 103, it is best to load all the robots on the second floor and have them go to the first floor together. However, this means that people 104 on the first floor have to wait a long time. If only the robots 103 near the elevator 101 on the second floor get on the elevator 101 and go to the first floor immediately, people will have to wait less time.

Also, for example, as shown in FIG. 5, suppose that a person 104 who wants to go to the second floor is waiting for the elevator 101 on the first floor. Suppose that there are two robots 103 on the second floor who want to go to the first floor to perform a task.

In this case, conversely, if only the closest robot 103 on the second floor is loaded and moved, the second robot may have to wait a long time in front of the elevator. If the elevator waits a little longer and the second robot can be loaded, the robot's task efficiency may increase significantly. Note that, because the tasks of the robot 103 are controlled by the central server 100, it is clear that it is more efficient to load two robots 103.

In other words, we want to optimize how long the elevator 101 should wait for the robot 103, taking into account both the human and the robot.

5. Overview of the embodiment
As such an adjustment, in this embodiment, the following solution is adopted for creating a behavior plan for the robot 103.

Usually, a robot's action plan is optimized to minimize "cost." Cost can be, for example, the time it takes to reach a goal, the distance to the goal, etc.

In the embodiment, in addition to the main cost (called the normal cost), an occupancy cost is introduced that represents the penalty for the robot 103 occupying equipment shared with the person 104.

The occupancy cost is expressed as a function of the time that the robot 103 occupies the shared facility (= the time that the person 104 cannot use it). For example, the occupancy cost is a (coefficient) x "the time that the elevator 101 waits for the robot 103."

The robot's 103 behavior plan is optimized so that the "normal cost + occupancy cost" is minimized. This means that plans that require the robot 103 to occupy shared equipment for too long will not be executed. The acceptable amount of occupancy can be adjusted by changing the coefficient (parameter) or function of the occupancy cost.

<6. Occupancy costs, etc.>
6.1. Occupancy Cost
Next, the occupancy cost according to the embodiment will be described.

The function used to calculate the occupancy cost can be expressed as a linear function, an exponential function, or a step function, where t is the time that the robot 103 occupies the shared equipment.
Linear function: at (a is the coefficient)
Exponential function: a ^t (used when you want to avoid waiting longer than the linear function)
Step function: 0 (t≦a)
∞(t>a)
Note that the step function is used as the threshold.

Let T be the set of times during which shared facilities are occupied in the plan.
Linear function: amax(T)
Exponential function: a ^(max(T))
Step function: 0 (max(T)≦a)
∞(max(T)>a)
Also, a combination of these functions may be used, for example at+bmax(T).

Next, we will explain the weighting of the normal cost f(t) and the occupancy cost g(t). If the parameter representing the weighting is α, then the total cost can be expressed as f(t) + α × g(t). Therefore, by adjusting α, it is possible to adjust the weighting of the occupancy cost.

Next, an example of the time t during which the robot 103 occupies the shared equipment will be described. In the case of the elevator 101, the time t during which the robot 103 occupies the shared equipment is, for example, as follows:
The time spent waiting for the elevator car to arrive after the button (up or down) is pressed The time spent waiting for the robot 103 to get on after the person 104 gets on the elevator 101 If multiple people 104 are waiting, the waiting times are added up. For example, if three people 104 wait for 20 seconds, 10 seconds, and 5 seconds, respectively, a total of 35 seconds is added up to the time t that the robot 103 occupies the shared facility.

6.2. Examples of Occupancy Costs
Next, the planning of the robot 103 with respect to costs will be described.
Total cost = normal cost + α × occupancy cost Normal cost: Time it takes for all robots to reach the goal,
Occupancy cost: The time the elevator waits for the robot.

In the state shown in FIG.
Plan to move all three robots 103 to the first floor:
Normal cost = 50
Occupancy cost = 50
When α=1, the total cost is 50+50=100.

Plan to move three robots 103 to the first floor in two batches:
Normal cost = 60
Occupancy cost = 30
When α=1, the total cost is 60+30=90.

Therefore, in this case, it would be better to plan to move three robots 103, which have a lower total cost, to the first floor in two trips.

In the state shown in FIG.
Plan to move two robots 103 together to the first floor:
Normal cost = 30
Occupancy cost = 40
When α=1, the total cost is 30+40=70.

Plan to move two robots 103 to the 1st floor in two trips:
Normal cost = 50
Occupancy cost = 30
When α=1, the total cost is 50+30=80.

Therefore, in this case, it would be better to plan to move the two robots 103 together, which have a smaller total cost, to the first floor.

<6.3. Change in weighting of occupancy costs>
Next, a case where the parameter α representing the weight is changed will be described.

Plan to move two robots 103 together to the first floor:
Normal cost = 50
Occupancy cost = 40
When α=1, the total cost is 50+1×40=90.

Plan to move two robots 103 to the 1st floor in two trips:
Normal cost = 80
Occupancy cost = 25
When α=1, the total cost is 80+1×24=105.

Therefore, when α = 1, it is better to plan to move the two robots 103, which have the smallest total cost, together to the 1st floor.

Next, a case where the parameter α representing the weight is changed from 1 to 3 will be described.
Plan to move two robots 103 together to the first floor:
Normal cost = 50
Occupancy cost = 40
If α=3, the total cost is 50+3×40=170.

Plan to move two robots 103 to the 1st floor in two trips:
Normal cost = 80
Occupancy cost = 25
If α=3, the total cost is 80+3×25=155.

Therefore, when α = 3, it is better to plan to move two robots 103, which have a smaller total cost, to the 1st floor in two trips.

In other words, by changing the parameter α, which indicates the weighting, it is possible to adjust whether to give priority to the robot 103 or the person 104.

In the embodiment, the robot's occupancy time and human waiting time are dynamically adjusted during operation by adjusting parameters such as the weighting of normal cost and occupancy cost, and the occupancy cost formula (coefficient).

In cases where dynamic adjustments are made, for example, adjustments are made to reduce the waiting time of people when the elevator 101 becomes crowded. In addition, adjustments are made to extend the occupancy time of the robot 103 when the robot 103 is not making good progress in its tasks, adjustments are made to extend the occupancy time of the robot 103 during times when people 104 are not using the facility, and adjustments are made to give priority to people 104 temporarily for equipment maintenance.

<6.4. Parameter adjustment flow>
6 is a diagram for explaining a flow of parameter adjustment according to an embodiment. As shown in FIG. 6, input data 301 includes, for example, the number of people waiting captured by a camera, the usage status of the elevator 101, a manual trigger, and the like.

When a trigger to start parameter adjustment occurs based on the input data 301, the parameters are adjusted (step S1).

In parameter adjustment (step S1), the weighting α between normal cost and occupancy cost and the formula (coefficient) for occupancy cost are adjusted.

Next, the situation (metrics) is checked (step S2). After the situation is checked, it is determined whether the situation has reached the target value. If the target value has not been reached, the parameters are adjusted again (step S1). If it is determined that the situation has reached the target value, the parameter adjustment process ends.

The target values are set, for example, to the task efficiency of the person 104, the average waiting time of the person 104, the task efficiency of the robot 103, the average occupancy time of the robot 103, etc.

6.5. Trigger for parameter adjustment
The trigger for starting parameter adjustment may be adjusted depending on the time of day. For example, the waiting time of people is shortened during the day when there are many people 104, and the occupancy time of the robot 103 is lengthened during the middle of the night when there are fewer people 104.

The trigger for starting parameter adjustment may be adjusted based on data obtained from outside. For example, when a sensor detects congestion of people 104, the waiting time of people 104 is temporarily shortened. The system learns how to make adjustments. Congestion is detected by recognizing people with a camera and determining how many people 104 are waiting. For example, adjustments are made to temporarily shorten the waiting time of people 104 depending on the usage status of the elevator 101. In this case, data on the button pressing status of the elevator 101 and the movement status of the car is used.

The trigger for starting parameter adjustment may be adjusted by an external operation. For example, to temporarily prioritize the task of the robot 103, the parameter is adjusted manually from the UI (User Interface).

<6.6. Status (metrics) check>
The user determines target values in advance (before starting parameter adjustment). The target values are, for example, the average waiting time of the person 104, the maximum waiting time of the person 104, the task efficiency of the robot 103 (how many tasks are completed per hour), and the average occupancy time of the equipment by the robot 103.

The waiting time of person 104, for example, in the case of elevator 101, is the time from pressing the button (up or down) until the elevator car arrives, or the time from when a person gets into elevator 101 until the robot 103 gets in. If multiple people 104 are waiting, the waiting times are added up. If person 104A waits 20 seconds, person 104B waits 10 seconds, and person 104C waits 5 seconds, the total is 35 seconds. The waiting time of person 104 may be an indicator of the increase in people's waiting time due to the influence of robot 103.

The status check in step S2 continues parameter adjustment until the target value is reached. The UI can be used to check how close the status is to the target value. If the status has reached the target value, the parameter adjustment process ends, and if the status has not reached the target value, the process returns to the parameter adjustment process in step S1.

6.7. Parameter Adjustment
Next, the concept of parameter adjustment will be described. For example, if you want to reduce the waiting time of the person 104, increase the weight of the occupancy cost, and if you want to increase the task efficiency of the robot 103, decrease the weight of the occupancy cost. Note that PID control and learning are used to determine how much the parameters should be changed in one cycle.

<6.8. When there is a priority within a task of the robot 103>
The multiple robots 103 perform different tasks, and the priorities of the tasks may differ. For example, a transportation task and a cleaning task for manufacturing. The transportation task is prioritized for production planning, but the cleaning task has a low priority because it does not have a large impact even if it does not proceed as planned.

Priority is expressed by how the cost is determined. With normal cost, higher costs are assigned to tasks with higher priority. For example, a low priority task takes 60 seconds and costs 60, while a high priority task takes 60 seconds and costs 120. With occupancy cost, higher priority tasks take less. In other words, costs are determined based on the priority of the task.

<6.9. When there is a priority within the task of person 104>
Like the robot 103, the person 104 also performs various different tasks, and the priority may differ for each task (person). For example, there is a person 104 who goes to fix the alarm of a device or robot and a person 104 who performs regular inspections.

In the case of person 104, priority can be expressed by determining the situation (metrics) used when checking the situation. For example, if a person's waiting time is used as a score, a person with a low priority will receive a score of 10 if they wait 10 seconds, but if a person with a high priority waits 10 seconds, a score of 30 will be recorded. Alternatively, only the waiting time and task efficiency of people performing high priority tasks can be recorded, and the rest of them can be ignored and not recorded as costs regardless of how long they wait. In other words, the evaluation score of person 104 is calculated based on the priority set for person 104.

<6.10. Example of when tasks have priority>
FIG. 7 is a diagram showing an example in which the tasks assigned to the robot 103 and the person 104 according to the embodiment have priorities.

In FIG. 7, an example is shown in which the central server 100 manages robots 103a and 103b. For robot 103a, "task 1, normal cost multiplier x 2.0, occupancy cost multiplier x 0.25, priority 5" is set. For robot 103b, "task 2, normal cost multiplier x 1.0, occupancy cost multiplier x 1.0, priority 1" is set. The central server 100 plans the robots 103 by changing the way costs are assigned for each robot 103's priority.

Furthermore, for person 104a, "Task 1, evaluation score multiplier x 2.0, priority 3" is set, and for person 104b, "Task 2, evaluation score multiplier x 1.0, priority 1" is set. The evaluation score monitor 401 displays the task efficiency score (evaluation score) of person 104 calculated by the central server 100. In other words, the evaluation score of person 104 who is assigned task 1 with a high priority is higher than the evaluation score of person 104 who is assigned task 2 with a lower priority than task 1.

The evaluation score monitor 401 displays the evaluation score calculated by the central server 100. For example, the evaluation score displayed may be a person's task efficiency score, but is not limited to this. It is also possible to display multiple scores. For example, the evaluation score may be displayed as a score for task A, a score for task B, etc. The central server 100 adjusts the parameters so that the multiple evaluation scores approach the target value. The central server 100 also changes the way the evaluation score is assigned for each priority of the person 104. Then, the occupancy score parameters are adjusted based on the evaluation score.

6.11. Example of Measuring Task Efficiency of Person 104
When measuring task efficiency of person 104, for example, the following factors are used:
1. Waiting time in front of the elevator 101 Waiting time information is acquired by a camera, a human presence sensor 402, or the like.
2. Operational status of elevator 11 The operational status of elevator 11 is how many people it has transported, how many times (floors) it has moved, and how many times the doors have been opened and closed.
3. Number of alarms responded to/Response time An alarm goes off when an abnormality occurs in a device or robot. The number of alarms responded to/Response time is the number of responses or the time it takes to respond, and is managed by the production system or robot system.
4. (Factory) Production volume (Factory) production volume is managed by the production system.
5. (Factory) Product Inspection Finished products and products in the process of production are inspected for defects.
(Factory) The number of product inspections is controlled by the production system.

<6.12. Parameter adjustment flow (overall)>
Fig. 8 is a diagram showing an initial state before parameter adjustment of the evaluation score monitor 401 according to the embodiment. Fig. 8 shows an example in which the central server 100 manages robots 103a and 103b. For the robot 103a, "task 1, normal cost multiplier x 2.0, priority 5" is set. For the robot 103b, "task 2, normal cost multiplier x 1.0, priority 1" is set. The central server 100 plans the robots 103 by changing the way costs are assigned for each priority of the robots 103.

Furthermore, for person 104a, "task 1, evaluation score multiplier x 2.0, priority 3" is set, and for person 104b, "task 2, evaluation score multiplier x 1.0, priority 1" is set. The evaluation score monitor 401 displays the task efficiency score of person 104 calculated by the central server 100.

In the initial state, the central server 100 operates without considering the balance with the person 104, and the evaluation score monitor 401 does not display evaluation scores such as the efficiency of task A and the efficiency of task B. Note that the central server 100 may operate with the balance with the person 104 adjusted in advance.

FIG. 9 is a diagram showing the state of the evaluation score monitor 401 according to the embodiment after a predetermined time has elapsed from the initial state before parameter adjustment. As shown in FIG. 9, after a predetermined time (1 hour) has elapsed from the initial state shown in FIG. 8, the evaluation score monitor 401 displays, for example, that the efficiency of task A is 10 (task A was performed 10 times per hour) and the efficiency of task B is 20.

If it is desired to increase the efficiency of task A to 15, the goal of increasing the efficiency of task A to 15 is input to the central server 100. Then, the central server 100 adjusts the occupancy cost parameters.

FIG. 10 shows the settings and evaluation score monitor 401 of the robot 103 during parameter adjustment by the central server 100 according to the embodiment.

As shown in FIG. 10, during parameter adjustment in the central server 100, for example, the following are set for robot 103a: "Task 1, normal cost multiplier x 2.0, occupancy cost (parameter a), priority 5." For robot 103b, the following are set: "Task 2, normal cost multiplier x 1.0, occupancy cost (parameter a), priority 1."

At this time, the evaluation score monitor 401 displays that the efficiency of task A is 10 (task A was performed 10 times per hour) and the efficiency of task B is 20. However, in the state shown in FIG. 10, the efficiency of task A is not 15, so the goal has not been achieved.

FIG. 11 is a diagram showing the settings of the robot 103 and the evaluation score monitor 401 after the central server 100 according to the embodiment has completed parameter adjustment. FIG. 11 shows a case where one hour has passed since the central server 100 started parameter adjustment, and the efficiency of task A has reached 15, achieving the goal.

As shown in FIG. 11, after the central server 100 completes parameter adjustment, parameter a is changed to a', and the robot 103a is set to "task 1, normal cost multiplier x 2.0, occupancy cost (parameter a'), priority 5." The robot 103b is set to "task 2, normal cost multiplier x 1.0, occupancy cost (parameter a'), priority 1."

At this time, the evaluation score monitor 401 displays that the efficiency of task A is 15, and the efficiency of task B is 25. Therefore, in the state shown in FIG. 11, the efficiency of task A is 15, so the goal has been achieved and parameter adjustment is completed.

The part that has been described so far as a person 104 can be any unknown moving object that cannot be managed by the central server 100. For example, it could be a robot 105 made by another company that is managed by another company's system (i.e. not managed by the central server 100).

FIG. 12 is a diagram for explaining a pattern in which the central server 100 according to the embodiment recognizes a robot 105 made by another company as an unknown moving object using the camera 106. In FIG. 12, the camera 106 recognizes the number of people 104 and robots 105 made by other companies in front of and inside the elevator 101 (=crowding situation), and transmits this to the central server 100.

The central server 100 recognizes the number of people 104 and robots 105 made by other companies that have been sent, and reflects this in the evaluation score. In FIG. 12, the settings for the people 104 are "task 1, evaluation score multiplier x 2.0, priority 3," and the settings for the robots 105 made by other companies are "task 2, evaluation score multiplier x 1.0, priority 1." The settings for the robot 103a are "task 1, normal cost multiplier x 2.0, occupancy cost parameter a', priority 5." The settings for the robot 103b are "task 2, normal cost multiplier x 1.0, occupancy cost parameter a', priority 1."

7. Functional block diagram of central server 100 according to the embodiment
Fig. 13 is a functional block diagram of the central server 100 according to the embodiment. As shown in Fig. 13, the central server 100 includes a parameter adjustment unit 601, an evaluation score calculation unit 602, a task planning unit 603, a robot control unit 604, and an elevator control unit 605.

Also, outside the central server 100, there are provided an external evaluation score target value input unit 500, a task input unit 501, an evaluation score display unit 502, an evaluation score data input unit 504, and communication units 503_1 and 503_2. The robot control unit 604 of the central server 100 is connected to the robot 103 via the communication unit 503_1. The elevator control unit 605 of the central server 100 is connected to the elevator controller 102 of the elevator 101 via the communication unit 503_2.

The evaluation score target value input unit 500 is a terminal or the like for an operator, into which the target value of the evaluation score is input. The task input unit 501 inputs the tasks of the robot 103 from a production system or the like.

The parameter adjustment unit 601 receives the target value input by the evaluation score target value input unit 500. The parameter adjustment unit 601 adjusts the parameters of the occupancy cost so that the evaluation score reaches the target value received from the evaluation score target value input unit 500. The evaluation score data input unit 504 inputs the number of units manufactured from the manufacturing equipment or production system, the waiting time of people from the camera in front of the elevator 101, etc.

The task planning unit 603 plans tasks using the occupancy cost after parameter adjustment. For example, the task planning unit 603 plans the task allocation and route of the robot 103, the timing of elevator control, and the like. The task planning unit 603 creates an action plan for the robot 103 based on costs that satisfy the received target value and include an occupancy cost that represents a penalty for the robot 103 occupying shared equipment. The target value is a target value that includes robots 105 made by other companies that are not subject to the action plan (see Figure 12).

The robot control unit 604 controls the robot 103 via the communication unit 503_1 based on the plan formulated by the task planning unit 603. The elevator control unit 605 controls the elevator controller 102 of the elevator 101 via the communication unit 503_2 based on the plan formulated by the task planning unit 603.

The evaluation score calculation unit 602 calculates an evaluation score based on the data input from the evaluation score data input unit 504, and outputs it to the evaluation score display unit 502 via the communication unit. The evaluation score display unit 502 is a terminal for an operator, etc., and displays the current evaluation score calculated by the evaluation score calculation unit 602. The evaluation score display unit 502 is, for example, the evaluation score monitor 401 shown in FIG. 7.

FIG. 14 is a diagram showing an example of a data input source input to the evaluation score data input unit 504 of the central server 100 according to an embodiment. In FIG. 14, the same parts as those in FIG. 13 are given the same reference numerals, and their description will be omitted.

As shown in FIG. 14, data is input to the evaluation score data input unit 504 from the production system/manufacturing equipment 606, the camera 106, the human presence sensor 106, and the elevator 101. For example, the production system/manufacturing equipment 606 inputs information related to the tasks of the people 104/other company's robot 105, such as the number of productions and inspections, the number of alarm responses, and the response time. For example, the camera 106 and the human presence sensor 106 inputs information related to the people 104/other company's robot 105 waiting for the elevator 101, and the number and duration of the people 104/other company's robot 105 inside the elevator 101. For example, the elevator 101 inputs information related to the number of times the elevator 101 is operated, the number of times the doors are opened and closed, and the time the doors are opened and closed.

<7.1 Relationship between the central server 100 and the robot system 613 manufactured by another company according to the embodiment>
FIG. 15 is a diagram for explaining the relationship between the central server 100 according to the embodiment and a robot system 613 made by another company.

For example, in a factory use case, there is a production system 611 that manages all tasks in the factory, and each company's robot system is linked to it. Figure 15 shows a robot system 612 that uses an embodiment and a robot system 613 made by another company. If information such as the progress of tasks in the robot system 613 made by another company can be obtained from a system that manages the entire system, such as the production system 611, this information can be used in the evaluation score.

The production system 611 issues task instructions for production process A to a robot system 612 using an embodiment, and receives the task progress status and task efficiency of production process A. The production system 611 also issues task instructions for production process B to a robot system 613 made by another company, and receives the task progress status and task efficiency of production process B. In the embodiment, the production system 611 transmits the task efficiency (for calculating an evaluation score) of production process B to the robot system 612 using an embodiment.

A system that oversees the entire system, such as a production system in a factory, also exists in warehouse use cases. When a system that oversees the entire system exists, the status of the tasks performed by the robot system 613 made by another company can be input to the central server 100 via the system and used to calculate the evaluation score. As a result, the task efficiency of the robot system 612 using the embodiment and the robot system 613 made by another company can be optimally balanced by adjusting the parameters.

FIG. 16 is a diagram for explaining the relationship between the central server 100 and a robot system 613 made by another company according to an embodiment. In FIG. 16, the progress and task efficiency of tasks performed by a robot system 613a made by another company are input to the overall management system 614. In addition, the progress and task efficiency of tasks performed by a robot system 613b made by another company are input to the overall management system 614.

The overall management system 614 is, for example, a production system in the case of a factory, or a warehouse management system in the case of a warehouse. The overall management system 614 inputs the task efficiencies of the other company's robot system 613a and the other company's robot system 613b to the evaluation score data input unit 504. The evaluation score data input unit 504 transmits the task efficiencies to the evaluation score calculation unit 602. The task efficiencies are used for the evaluation score in the evaluation score calculation unit 602.

8. Flowchart for explaining the overall operation of the central server 100 according to the embodiment
FIG. 17 is a flowchart for explaining the operation of the central server 100 according to the embodiment.

As shown in FIG. 17, task execution is started without occupancy cost or with default parameters (step S11). In step S11, task input, task planning, and robot 103/elevator 101 control are performed.

Next, data is acquired from the evaluation score data input unit 504, and task execution is performed for a while while calculating the evaluation score (step S12). In the states of steps S11 and S12, the tasks of the person 104 and the robot 103 are not balanced.

Next, it is determined whether the evaluation score has reached the target value (step S13). If the evaluation score has reached the target value (YES in step S13), the control process of the central server 100 ends. If the evaluation score has reached the target value (NO in step S13), the parameter adjustment unit 601 changes the occupancy cost parameters so that the evaluation score approaches the target value (step S14).

After the occupancy cost parameters are changed in the parameter adjustment unit 601, the changed parameters are applied to the task planning unit 603 (step S15). Then, data is obtained from the evaluation score data input unit 504, and the task is executed for a while while calculating the evaluation score (step S16).

In steps S14 to S16, the balance of tasks between the person 104 and the robot 103 is gradually optimized.

The evaluation score target value can be input from the evaluation score target value input unit 500 at any time. It can also be input before the start. If it is input after the end, the flow starts again from the process of determining whether the evaluation score has reached the target value in step S503.

9. Flowchart for explaining the operation of the central server 100 according to the embodiment during task execution
18 is a flowchart for explaining the flow of task execution processing by the central server 100 according to the embodiment. In FIG. 18, new tasks are input one after another from a production system or the like, and therefore the process is expressed as a loop.

A new task is input from the task input unit 501 (step S21). The task planning unit 603 assigns the input new task to the robot 103 (step S22). Then, a path for the assigned robot 103 is planned (step S23). Here, when planning the path, the occupancy cost is used as described above.

The robot control unit 604 issues a movement command to the robot 103 based on the planned route (step S31). The elevator control unit 605 moves the elevator 101 at an appropriate time based on the created plan (step S32). The robot control unit 604 starts the robot 103's work when the robot 103 reaches the destination (step S33). The work may be, for example, loading and unloading of goods. Note that the processing of steps S31 to S33 may be executed asynchronously.

10. Settings and UI
Next, the settings and UI of the central server 100 according to the embodiment will be described.

In the embodiment, the central server 100 sets parameters related to occupancy costs. Changing the parameters through parameter setting changes the occupancy time of the robot 103 and the waiting time of the person 104. When a target average waiting time of people is given, the parameters are adjusted and the waiting time of the person changes. When a target task efficiency is given, the parameters are adjusted and the task efficiency of the robot 103 changes.

In the embodiment, there is a UI for adjusting parameters related to occupancy cost, a UI for setting the target average waiting time of people, a UI for setting the target task efficiency of the robot, and a UI for displaying the status during parameter adjustment.

The UI that displays the status during parameter adjustment includes the parameter value itself, changes in the average waiting time of people, changes in the task efficiency of the robot 103, changes in the time that the robot 103 occupies shared equipment, etc.

<11. Use Cases>
Next, a use case of the system according to the embodiment will be illustrated.

Use case 1
The system according to the embodiment can be used for inter-floor transportation in a factory or warehouse. In inter-floor transportation in a factory or warehouse, a robot 103 transports an object to another floor by using an elevator 101. Both the robot 103 and a person 104 use the same elevator 101 to move between floors.

The elevator 101 is treated as a shared facility, and the occupancy cost is used to optimize how long the elevator should wait for the robot 103. In addition to the elevator 101, control similar to that of the elevator 101 can be performed for narrow passages where only one robot 103 can pass, or for equipment that uses people 104.

Use case 2
The system according to the embodiment can be used as a guide robot in an airport or a commercial building. When the robot 103 is used as a guide robot, the robot 103 guides a person 104. The person 104 guided by the robot 103 gets into the elevator 101 together with the robot 103. A user who has not been guided may also use the elevator 101 at the same time.

Elevator 101 is used as a shared facility. Robot 103's plan uses occupancy cost to optimize how long elevator 101 should wait for the guide robot. Escalators and narrow passageways can be controlled in the same way as elevator 101.

Use case 3
The system according to the embodiment can be used for parking and charging unmanned taxis.
In the case of an unmanned taxi that is remotely controlled, if the unmanned taxi occupies a parking lot or charging space for too long, the person 104 will be unable to use the parking lot or charging space.

In such cases, the system according to the embodiment uses occupancy costs to make optimal plans to prevent the autonomous vehicle from occupying the space for too long.

<12. Effects>
Therefore, when there is equipment that is used jointly by a person 104 and a robot 103, the central server 100 according to the embodiment can improve the work efficiency of both parties and can apply an optimal plan overall to the robot 103.

While the optimal plan differs depending on the environment and circumstances, the method of the embodiment allows for parameter adjustments to adjust the balance between the waiting time of the person 104 and the work efficiency of the robot 103, thereby achieving the optimal balance desired by the user.

The central server 100 can also adjust the occupancy time of the robot 103 and the waiting time of the person 104. This allows for more optimal control than simply "preventing occupancy for longer than a specified time" or "always giving priority to the person 104 (or robot 103)." The central server 100 of the embodiment can save cases where the task efficiency of the robot 103 can be significantly improved by simply extending the occupancy time a little. Depending on how the occupancy cost is expressed, it is also possible to simply "prevent occupancy for longer than a specified time" for the robot 103.

Furthermore, according to the embodiment, the balance can be dynamically adjusted during operation. For example, an operator can adjust parameters when there is temporary congestion or when prioritizing the work of the robot 103.

FIG. 19 is a diagram showing a comparison of the waiting time of person 104 before and after parameter adjustment of occupancy cost in the embodiment. As shown in FIG. 19, the portion before parameter adjustment 701_1 is shown by dots, and the portion after parameter adjustment 701_2 is shown by diagonal lines. As shown in FIG. 19, the waiting time after parameter adjustment 701_1 is longer than the portion before parameter adjustment 701_2 in the first minute, but over time the waiting time after parameter adjustment 701_1 becomes shorter than the portion before parameter adjustment 701_2.

Furthermore, according to the embodiment, the occupancy cost parameters can be dynamically adjusted. How the parameters should be adjusted varies depending on the environment and situation, so more optimal control can be achieved if they can be dynamically adjusted during operation.

According to the embodiment, by adding an "occupancy cost" to the cost when planning the behavior of the robot 103, it becomes possible to make a plan that takes into account the penalty for the robot 103 occupying equipment. By adjusting the parameters of the occupancy cost, it is possible to adjust the balance between the waiting time when a person uses equipment and the task efficiency of the robot.

According to the embodiment, by incorporating occupancy costs into the plan, control can be performed with consideration of the overall efficiency including the robot 103 and the person 104, rather than simply prioritizing either the robot 103 or the person 104 as in conventional technology. For example, if the task efficiency of the robot 103 is greatly improved, the person 104 is made to wait a little longer. If not, the robot 103 is moved so as to shorten the waiting time of the person.

13. Hardware Configuration
FIG. 20 is a hardware configuration diagram showing an example of the central server 100 according to the embodiment.
A computer that realizes the arithmetic device 102 of the information processing device 100 according to each of the above-described embodiments is realized by a computer 1000 having a configuration as shown in Fig. 20, for example. Fig. 20 is a hardware configuration diagram showing an example of the computer 1000 that realizes the functions of the arithmetic device 102. The computer 1000 has a CPU 1100, a RAM 1200, a ROM (Read Only Memory) 1300, a HDD (Hard Disk Drive) 1400, a communication interface 1500, and an input/output interface 1600. Each unit of the computer 1000 is connected by a bus 1050.

The CPU 1100 operates based on the programs stored in the ROM 1300 or the HDD 1400 and controls each part. For example, the CPU 1100 loads the programs stored in the ROM 1300 or the HDD 1400 into the RAM 1200 and executes processes corresponding to the various programs.

The ROM 1300 stores boot programs such as the BIOS (Basic Input Output System) that is executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the hardware of the computer 1000.

HDD 1400 is a computer-readable recording medium that non-temporarily records programs executed by CPU 1100 and data used by such programs. Specifically, HDD 1400 is a recording medium that records application programs related to the present disclosure, which are an example of program data 1450.

The communication interface 1500 is an interface for connecting the computer 1000 to an external network 1550 (e.g., the Internet). For example, the CPU 1100 receives data from other devices and transmits data generated by the CPU 1100 to other devices via the communication interface 1500.

The input/output interface 1600 is an interface for connecting the input/output device 1650 and the computer 1000. For example, the CPU 1100 receives data from an input device such as a keyboard or a mouse via the input/output interface 1600. The CPU 1100 also transmits data to an output device such as a display, a speaker, or a printer via the input/output interface 1600. The input/output interface 1600 may also function as a media interface that reads programs and the like recorded on a specific recording medium. Examples of media include optical recording media such as DVDs (Digital Versatile Discs) and PDs (Phase change rewritable Discs), magneto-optical recording media such as MOs (Magneto-Optical Disks), tape media, magnetic recording media, and semiconductor memories.

For example, when the computer 1000 functions as the arithmetic device 102 according to the first embodiment, the CPU 1100 of the computer 1000 executes an image processing program loaded onto the RAM 1200 to realize functions such as the reissue request unit 43. The HDD 1400 also stores the SE management processing program according to the present disclosure and data in the content storage unit 121. The CPU 1100 reads and executes the program data 1450 from the HDD 1400, but as another example, it may also obtain these programs from other devices via the external network 1550.

The above describes in detail preferred embodiments of the present disclosure with reference to the attached drawings, but the technical scope of the present disclosure is not limited to such examples. It is clear that a person with ordinary knowledge in the technical field of the present disclosure can conceive of various modified or revised examples within the scope of the technical ideas described in the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

Furthermore, the effects described in this specification are merely descriptive or exemplary and are not limiting. In other words, the technology disclosed herein may achieve other effects that are apparent to a person skilled in the art from the description in this specification, in addition to or in place of the above effects.

This technology can also be configured as follows:

(1)
receiving target values for the robot and the person in a shared facility shared by the robot and the person;
creating an action plan for the robot based on costs that satisfy the received target values and include an occupancy cost of the robot occupying the shared facility;
Information processing methods.
(2)
The information processing method according to (1), wherein the target value includes at least one of an average waiting time of the person, a maximum waiting time of the person, a task efficiency of the robot, and an average occupancy time of the shared equipment by the robot.
(3)
The information processing method according to (1) or (2), further comprising displaying an evaluation score for the received target value.
(4)
Creating an action plan for the robot includes:
calculating said occupancy cost;
The occupancy cost is expressed as a function of the time that the robot occupies the shared equipment.
An information processing method according to any one of (1) to (3).
(5)
The robot is a plurality of robots,
The costs include a normal cost representing a time required for the plurality of robots to reach a goal.
An information processing method according to any one of (1) to (4).
(6)
The action plan is generated to minimize the cost.
An information processing method according to (5).
(7)
The normal cost is f(t),
The occupancy cost is g(t),
The cost is expressed as f(t) + α × g(t),
α is a parameter indicating the ratio between the normal cost f(t) and the occupancy cost g(t),
An information processing method according to (5) or (6).
(8)
The function is a linear function, an exponential function, a step function, or a combination of the linear function, the exponential function, and the step function of the time that the robot occupies the shared equipment.
An information processing method according to (4).
(9)
Creating an action plan for the robot includes:
The information processing method according to (7), further comprising re-adjusting the parameters α and g(t) if the target value is not reached.
(10)
The parameter is adjusted until the target value is reached.
An information processing method according to (9).
(11)
The parameters are adjusted according to the time of day.
An information processing method according to (9).
(12)
The parameters are adjusted based on data obtained from an external source.
An information processing method according to (9).
(13)
The parameters are adjusted by external manipulation.
An information processing method according to (9).
(14)
Priorities are set for the tasks of the plurality of robots;
The cost is determined based on the priority.
An information processing method according to (5).
(15)
a first priority set for a first task of a first robot among the plurality of robots is higher than a second priority set for a second task of a second robot among the plurality of robots;
The normal cost of the first task is higher than the normal cost of the second task.
An information processing method according to (14).
(16)
an occupancy cost of the first task is lower than the occupancy cost of the second task;
An information processing method according to (15).
(17)
The person is a plurality of people,
The priority is set for each person,
The evaluation score of the person is calculated based on the priority set for the person.
An information processing method according to (16).
(18)
a third priority set for a third task of a first person of the plurality of people is higher than a fourth priority set for a fourth task of a second person of the plurality of people;
The evaluation score of the first person is higher than the evaluation score of the second person.
An information processing method according to (17).
(19)
a receiving unit that receives target values for the robot and the person in a shared facility shared by the robot and the person;
and a planning unit that creates a behavior plan for the robot based on costs including an occupancy cost of the robot occupying the shared facility while satisfying the received target value.
(20)
In the information processing device,
receiving target values for the robot and the person in a shared facility shared by the robot and the person;
creating an action plan for the robot based on costs including an occupancy cost of the robot occupying the shared facility while satisfying the received target value;
program.

100 Central server 101 Elevator 102 Elevator controller 103 Robot 104 Person 105 Robot made by another company 106 Camera 200 Route 301 Input data 401 Evaluation score monitor 402 Human presence sensor
403 Apparatus 500 Evaluation score target value input unit 501 Task input unit 502 Evaluation score display unit 503_1, 503_2 Communication unit 504 Evaluation score data input unit 601 Parameter adjustment unit 602 Evaluation score calculation unit 603 Task planning unit 604 Robot control unit 605 Elevator control unit 606 Production system/manufacturing apparatus 611 Production system 612 Robot system using the embodiment 613 Robot system made by another company 614 Overall management system 701_1, 701_2 Waiting time

Claims (20)

1. receiving target values for the robot and the person in a shared facility shared by the robot and the person;
   creating an action plan for the robot based on costs that satisfy the received target values and include an occupancy cost of the robot occupying the shared facility;
   Information processing methods.
2. The target value includes at least one of an average waiting time of the person, a maximum waiting time of the person, a task efficiency of the robot, and an average occupancy time of the shared equipment by the robot.
   The information processing method according to claim 1 .
3. The information processing method according to claim 1 , further comprising displaying an evaluation score for the received target value.
4. Creating an action plan for the robot includes:
   calculating said occupancy cost;
   The occupancy cost is expressed as a function of the time that the robot occupies the shared equipment.
   The information processing method according to claim 1 .
5. The robot is a plurality of robots,
   The costs include a normal cost representing a time required for the plurality of robots to reach a goal.
   The information processing method according to claim 1 .
6. The action plan is generated to minimize the cost.
   The information processing method according to claim 5.
7. The normal cost is f(t),
   The occupancy cost is g(t),
   The cost is expressed as f(t) + α × g(t),
   α is a parameter indicating the ratio between the normal cost f(t) and the occupancy cost g(t),
   The information processing method according to claim 5.
8. The function is a linear function, an exponential function, a step function, or a combination of the linear function, the exponential function, and the step function of the time that the robot occupies the shared equipment.
   The information processing method according to claim 4.
9. Creating an action plan for the robot includes:
   and if the target value is not reached, re-adjusting the α and g(t) parameters.
   The information processing method according to claim 7.
10. The parameter is adjusted until the target value is reached.
    The information processing method according to claim 9.
11. The parameters are adjusted according to the time of day.
    The information processing method according to claim 9.
12. The parameters are adjusted based on externally obtained data.
    The information processing method according to claim 9.
13. The parameters are adjusted by external manipulation.
    The information processing method according to claim 9.
14. Priorities are set for the tasks of the plurality of robots;
    The cost is determined based on the priority.
    The information processing method according to claim 5.
15. a first priority set for a first task of a first robot among the plurality of robots is higher than a second priority set for a second task of a second robot among the plurality of robots;
    The normal cost of the first task is higher than the normal cost of the second task.
    The information processing method according to claim 14.
16. an occupancy cost of the first task is lower than the occupancy cost of the second task;
    The information processing method according to claim 15.
17. The person is a plurality of people,
    The priority is set for each person,
    The evaluation score of the person is calculated based on the priority set for the person.
    The information processing method according to claim 16.
18. a third priority set for a third task of a first person of the plurality of people is higher than a fourth priority set for a fourth task of a second person of the plurality of people;
    The evaluation score of the first person is higher than the evaluation score of the second person.
    18. The information processing method according to claim 17.
19. a receiving unit that receives target values for the robot and the person in a shared facility shared by the robot and the person;
    a planning unit that creates a behavior plan for the robot based on costs including an occupancy cost of the robot occupying the shared facility while satisfying the received target value.
    Information processing device.
20. In the information processing device,
    receiving target values for the robot and the person in a shared facility shared by the robot and the person;
    creating a behavior plan for the robot based on costs including an occupancy cost of the robot occupying the shared facility while satisfying the received target value;
    program.

Images