Information processing device, information processing method, and program

Abstract

[Problem] To derive a reliable answer from answers to a question received from a plurality of robot devices. [Solution] An information processing device is provided with: a weight management unit that determines weights of answers to a question based on meta-information about the answers, wherein the answers to the question are obtained from each of a plurality of answerer robots that received the question; and a final answer determination unit that derives a final answer to the question from each of the answers obtained from the plurality of answerer robots based on the weights of the answers.

Description

Technical Field

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. Background Art

[0002] In recent years, robot devices, which are mechanical devices that use electrical or magnetic effects to imitate biological behavior, have become popular. For example, robot hand devices and the like are widely used in factories for the purpose of automating production work.

[0003] On the other hand, in various scenes of daily life, the development of robot devices that support human activities as human partners is underway. Unlike industrial robot devices that repeat specific operations, such robot devices are required to have the ability to flexibly adapt to various aspects of the human living environment.

[0004] For example, such a robot device is required to learn how to operate various objects existing in a human living environment by itself. However, autonomous learning by a robot device generally takes time. Therefore, it is considered to improve the learning efficiency of the robot device by sharing information with other robot devices existing around the robot device and increasing the amount of information.

[0005] For example, the following patent document 1 discloses a method of appropriately managing information in a robot device that shares information with other robot devices using a communication function. According to the technology disclosed in patent document 1, the newness of shared information can be determined by counting the number of generations when the information is transmitted.

[0006] Reference List

[0007] Patent Literature

[0008] Patent document 1: JP 2005-279830 A. Summary of the invention

[0009] Technical issues

[0010] Here, in order for a robot to determine how to operate an unknown object based on information obtained from other robot devices through communication, it is important to classify the acquired information with reasonable weighting. However, the technology disclosed in Patent Document 1 cannot determine the reliability of shared information.

[0011] For this reason, when receiving answers to a question from a plurality of robot apparatuses, there is a need for an information processing apparatus that can derive a reliable answer by appropriately weighting the answers to the question.

[0012] Solutions to the Problem

[0013] According to the present disclosure, an information processing device is provided. The information processing device includes: a weight management unit that determines the weight of the answer based on the meta-information of the answer, with respect to the answer to the question obtained from each of a plurality of respondent robots that have received the question and the meta-information of the answer; and a final answer determination unit that derives a final answer to the question from the respective answers from the plurality of respondent robots based on the weight of the answer.

[0014] In addition, according to the present disclosure, there is provided an information processing method. The information processing method includes: determining, by an arithmetic unit, a weight of an answer based on the meta-information of the answer, regarding an answer to a question obtained from each of a plurality of respondent robots that have received the question and meta-information of the answer; and deriving a final answer to the question from the respective answers from the plurality of respondent robots based on the weight of the answer.

[0015] Furthermore, according to the present disclosure, there is provided a program that causes a computer to function as: a weight management unit that determines a weight of an answer to a question obtained from each of a plurality of answerer robots that have received the question and meta-information of the answer based on meta-information of the answer; and a final answer determination unit that derives a final answer to the question from the respective answers from the plurality of answerer robots based on the weight of the answer.

[0016] According to the present disclosure, answers can be weighted based on the similarity between the respondent robot and the questioner robot, the respondent robot's proficiency level regarding the object, or the respondent robot's learning experience amount, etc., by using meta-information of the answers.

[0017] Beneficial effects of the present invention

[0018] As described above, according to the present disclosure, a reliable answer can be derived by appropriately weighting multiple answers to a question.

[0019] Note that the above-described effects are not necessarily limitative, and any one of the effects described in this specification or any other effect that can be obtained from this specification may be exhibited in addition to or instead of the above-described effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is an explanatory diagram showing transmission of a question from the robot 1 to each of the robots 3 .

[0021] Figure 2 1 is an explanatory diagram showing that an answer to a question is transmitted to the robot 1 from each robot 3 to which the question is transmitted.

[0022] Figure 3 It is an explanatory diagram showing the overall flow of the technology according to the embodiment of the present disclosure.

[0023] Figure 4 is a block diagram showing an example of the functional configuration of the information processing device 10 according to the same embodiment.

[0024] Figure 5 : is a flowchart showing the flow of the operation of the information processing device 10 according to the same embodiment.

[0025] Figure 6 is a sequence diagram showing the interaction between the information processing device 10 and the robot device 30 according to the same embodiment.

[0026] Figure 7 is an explanatory diagram showing that an answer from each of the respondent robots 3 is transmitted to the information processing device 11 in the modification.

[0027] Figure 8 1 is an explanatory diagram showing that the robot 12 newly added to the network 40 accesses the information processing device 11 in the modification.

[0028] Fig. 9 is an explanatory diagram showing an example of information stored in the database storage unit 100 of the information processing device 11 in the modification.

[0029] Fig.10 is a block diagram showing an example of the hardware configuration of the information processing device 10 according to the same embodiment. DETAILED DESCRIPTION

[0030] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Note that in this specification and the accompanying drawings, redundant descriptions of configuration elements having substantially the same functional configuration are omitted by providing the same reference numerals.

[0031] The description will be given in the following order.

[0032] 1. Implementation Method

[0033] 1.1. Summary

[0034] 1.2. Configuration of information processing equipment

[0035] 1.3. Operation of information processing devices

[0036] 2. Modify the plan

[0037] 3. Hardware Configuration

[0038] 4. Summary

[0039] <1. Implementation Method>

[0040] (1.1. Summary)

[0041] First, refer to Figures 1 to 3 An outline of the technology according to the embodiment of the present disclosure is described. Figure 1 3 is an explanatory diagram showing the transmission of a question from the robot 1 to the robot 3 . Figure 2 1 is an explanatory diagram showing that an answer to a question is transmitted to the robot 1 from each robot 3 to which the question is transmitted. Figure 3 It is an explanatory diagram showing the overall flow of the technology according to the present embodiment.

[0042] like Figure 1 As shown, when the robot 1 identifies an object 2 that is unknown to it, the robot 1 (hereinafter, also referred to as the questioner robot 1) sends questions related to the object 2 to other robots 3 (hereinafter, also referred to as the respondent robots 3) present around the robot 1.

[0043] When the questioner robot 1 encounters an object 2 unknown to itself, the questioner robot 1 does not know the name of the object 2 or how to handle the object 2. For this reason, in order to teach the respondent robot 3 that knows the object 2 the name of the object 2 or how to handle the object 2, the questioner robot 1 sends questions related to the object 2 to the respondent robots 3 that exist around the questioner robot 1. Accordingly, the questioner robot 1 can learn the name of the object 2 or the method of handling the object 2 in a shorter time than the time the questioner robot 1 learns how to handle the object 2 by itself or is taught how to handle the object 2 by a user.

[0044] Specifically, in various scenes of daily life, a robot device that supports human activities as a human partner may encounter various unknown objects 2. For example, there are things in the world that have exactly the same appearance but have completely different applications, such as pens and laser pointers. Therefore, when a slender rod-shaped unknown object exists in another person's house, it is difficult for the robot device to determine whether the unknown object is a pen or a laser pointer. In addition, when the robot device moves to an unknown country or region due to the user moving, etc., the robot device may encounter unknown objects whose applications, etc., are difficult to determine from the appearance. In addition, in the user's home, there may be unique objects created by the user, etc. Even when the robot device refers to any database, etc., the robot device cannot know how to handle such a unique object.

[0045] Therefore, the questioner robot 1 sends questions about the object 2 unknown to the questioner robot 1 to a plurality of respondent robots 3 existing nearby to obtain information about the unknown object 2 before arbitrarily handling the unknown object 2 .

[0046] exist Figure 1 and Figure 2 In the example, the object 2 has a shape, but the present embodiment is not limited to this example. The object 2 may not have a shape such as an inner space of a predetermined room. As long as the questioner robot 1 can apply some action, the object 2 may have a shape or may not have a shape.

[0047] Afterwards, if Figure 2 As shown, each respondent robot 3 that has received the question generates an answer to the question related to the object 2 and sends the answer to the questioner robot 1. The questioner robot 1 derives a final answer to the question by weighting each of the multiple answers received. Specifically, the questioner robot 1 determines the most effective or most reliable answer from the multiple answers received based on the weighting, and the determined answer can be regarded as the final answer to the question.

[0048] Here, the questioner robot 1 receives answers to questions from a plurality of respondent robots 3, but the answers from the plurality of respondent robots 3 cannot be the same. This is because the answers from the respondent robots 3 are generated based on the learning experience, etc. of each respondent robot 3. Therefore, the answers from the respondent robots 3 have different possibilities.

[0049] Therefore, the questioner robot 1 can more easily determine an answer with higher validity or reliability by weighting each of the received multiple answers. It should be noted that the weighting of the received multiple answers is performed based on the meta information of the answers received together with the answers.

[0050] For example, the questioner robot 1 can arrive at a final answer to the question by giving a larger weight to the answer from the respondent robot 3 within a certain distance from the object 2. In this case, the meta-information of the answer will include information about the distance between the object 2 and the respondent robot 3. The respondent robot 3 closer to the object 2 is more likely to have detailed information about the object 2 than the respondent robot 3 farther from the object 2. For this reason, the questioner robot 1 can more easily determine an answer with higher reliability from among the received multiple answers by giving a larger weight to the answer from the respondent robot 3 closer to the object 2.

[0051] Furthermore, the questioner robot 1 can arrive at a final answer to the question by giving a larger weight to the answer from the respondent robot 3 having a shape or type similar to that of the questioner robot 1. In this case, the meta-information of the answer will include information on the shape or type of the respondent robot 3. Regarding how to handle the object 2 or the sense of distance from the object 2, the respondent robot 3 similar to the questioner robot 1 is more likely to have information suitable for the questioner robot 1 than the respondent robot 3 different from the questioner robot 1. For this reason, the questioner robot 1 can more easily determine an answer suitable for itself from among the received multiple answers by giving a larger weight to the answer from the respondent robot 3 similar to the questioner robot 1.

[0052] Specifically, according to Figure 3 The technology according to this embodiment is executed according to the process shown.

[0053] First, a question about the object 2 is sent from the questioner robot 1 to the surrounding answerer robots 3 together with meta information including an image of the object 2. The answerer robot 3 inputs the meta information including the image of the object 2 into each classifier (classifier 1 to classifier N) (input mode) to classify the object 2 and generate an answer to the question. Thereafter, the answerer robot 3 sends the generated answer and meta information to the questioner robot 1.

[0054] The questioner robot 1, which receives the answer and meta-information from each of the answerer robots 3, combines each received answer by weighting each answer based on the meta-information of the answer. Next, the questioner robot 1 makes a decision on the final answer to the question from the received answers by considering the weight of each answer. Therefore, the questioner robot 1 can determine what the unknown object 2 is by using the classifier of the answerer robot 3. Therefore, the questioner robot 1 can appropriately handle the object 2 in a shorter time without autonomous learning about the unknown object 2 through trial and error, etc.

[0055] (1.2. Configuration of information processing device)

[0056] Next, we will refer to Figure 4 An example of a functional configuration of the information processing apparatus according to the present embodiment is described. Figure 4 1 is a block diagram showing an example of the functional configuration of the information processing device 10 according to the present embodiment. The information processing device 10 according to the present embodiment can be installed in, for example, Figure 1 and Figure 2 The questioner robot 1 is shown.

[0057] like Figure 4As shown, the information processing device 10 includes a sensor unit 110, a question generator 120, a communication unit 130, an answer acquisition unit 140, a weight management unit 150, a final answer determination unit 160, and a drive control unit 170. The information processing device 10 communicates with the robot device 30 (with the robot device 30) via the network 40 through the communication unit 130. Figure 1 and Figure 2 The responder robot 3 shown communicates accordingly).

[0058] The sensor unit 110 includes various known sensors and measures the state of the external environment including the object 2. The sensor unit 110 may include various cameras, such as an RGB camera, a grayscale camera, a stereo camera, a depth camera, an infrared camera, or a time-of-flight (ToF) camera, or may include various distance sensors, such as a laser imaging detection and ranging (LIDAR) sensor, or a radio detection and ranging (RADAR) sensor, or may include a microphone, an illuminometer, a pressure gauge, a thermometer, a hygrometer, etc.

[0059] The question generator 120 generates questions to be sent to the robot device 30. Specifically, the question generator 120 generates questions related to the object 2 and meta information of the question. For the questioner robot 1 on which the information processing device 10 is mounted, the object 2 is an unknown object. Therefore, the question generator 120 generates questions asking what the object 2 is and how to handle the object 2. For example, the question generator 120 may generate questions asking about any one of the structure, application, name, grasping method, or operation method of the object 2.

[0060] The meta information of the question includes information for allowing the robot device 30 that has received the question to specify the object 2. For example, the meta information of the question may include information about the object 2, such as image information of an image of the object 2, color and shape information of the object 2, position information of the object 2, sound information picked up from the object 2, or odor information measured from the object 2. Such information about the object 2 may be acquired, for example, by the sensor unit 110. The information about the object 2 may be information that can be acquired without contacting the object 2. In this case, the sensor unit 110 may reduce the influence on the object 2 when acquiring the information about the object 2.

[0061] In addition, the meta information of the question may include information about the purpose of the question. The clearer the purpose of the question, the more accurately the robot device 30 to which the question is sent can generate an answer that matches the answer requested by the information processing device 10. Accordingly, the information processing device 10 can receive a more appropriate answer.

[0062] Furthermore, the meta information of the question may include a one-time password. When the meta information of the question includes a one-time password, the information processing device 10 can construct a system in which an answer is transmitted from one robot device 30 only once.

[0063] It is considered that the information processing device 10 can derive a final answer to the question with high validity and reliability by receiving answers from each of a large number of robot devices 30. Therefore, the same answer sent multiple times from one robot device 30 may be a disturbing factor in deriving the final answer to the question. Therefore, it is preferable that the information processing device 10 restricts the answers from the robot device 30 to only one answer using a one-time password or the like. When a robot device 30 that sends the same answer multiple times is found without using a one-time password, the information processing device 10 can exclude the answer from the robot device 30 to derive the final answer to the question.

[0064] The communication unit 130 is a communication interface that sends questions to the robot device 30 and receives answers from the robot device 30 .

[0065] For example, the communication unit 130 may send questions to the robot device 30 and receive answers from the robot device 30 via the network 40. The network 40 may be a public communication network such as the Internet, a satellite communication network, or a telephone line network, or may be a communication network provided in a limited area such as a local area network (LAN) or a wide area network (WAN).

[0066] For example, the communication unit 130 may directly send a question to the robot device 30 and receive an answer from the robot device 30. In this case, the communication unit 130 includes an antenna and a communication circuit, and communicates wirelessly with the robot device 30. For example, the communication unit 130 may perform wireless communication with the robot device 30 by communicating using a signal in a frequency band of several hundred MHz to several GHz (e.g., 920 MHz) such as Wi-Fi (registered trademark), ZigBee (registered trademark), Bluetooth (registered trademark), Bluetooth Low Energy (registered trademark), ANT (registered trademark), ANT+ (registered trademark), or EnOcean Alliance (registered trademark).

[0067] The robot device 30 receives a question from the information processing device 10, and transmits an answer to the received question and meta information of the answer to the information processing device 10. The robot device 30 corresponds to Figure 1 and Figure 2 The robot device 30 generates an answer to the received question and meta-information of the answer based on the information on the object 2 included in the meta-information of the question.

[0068] Specifically, the robot apparatus 30 may input the image information of the object 2 included in the meta information of the question into the classifier or learning device of the robot apparatus 30 to generate an answer to the received question. For example, the robot apparatus 30 may generate an answer to the following question: what is the object 2, or which operation method should be used for the object 2. In particular, when the robot apparatus 30 uses its own learning device to generate an answer to a question, the learning device reflects the data learned by each robot apparatus 30 in the past, making it more likely that different answers will be obtained for each robot apparatus 30.

[0069] The robot apparatus 30 generates the meta information of the answer. The meta information of the answer includes at least one of information about the robot apparatus 30 and information about the relationship between the robot apparatus 30 and the object 2.

[0070] For example, the meta information of the answer may include information about the structure, development year, model, arm or leg structure, stay period, manufacturer, operating system (OS) version, or software version of the robot apparatus 30 as the information about the robot apparatus 30. When the meta information of the answer includes the information about the robot apparatus 30, the information processing device 10 can determine the similarity between the questioner robot 1 on which the information processing device 10 is mounted and the robot apparatus 30. Based on this, the information processing device 10 can determine whether the answer received from the robot apparatus 30 is an answer suitable for itself.

[0071] In addition, the meta-information of the answer may include information about the last use date, the last viewing date, the number of uses, or the frequency of use of the object 2 by the robot apparatus 30, or information about the distance between the robot apparatus 30 and the object 2, as information about the relationship between the robot apparatus 30 and the object 2. When the meta-information of the answer includes information about the relationship between the robot apparatus 30 and the object 2, the information processing apparatus 10 can determine the frequency, etc., of the processing of the object 2 by the robot apparatus 30. Based on this, the information processing apparatus 10 can determine whether the answer received from the robot apparatus 30 is highly reliable.

[0072] In addition, the meta-information of the answer may include: the frequency of the user asking the robot apparatus 30 about the object 2, the frequency of another questioner robot 1 asking the robot apparatus 30 about the object 2, the possibility of the answer expected by the robot apparatus 30, information about the communication path between the information processing apparatus 10 and the robot apparatus 30, etc. The more kinds of information included in the meta-information of the answer, the more detailed weighting can be performed by the weight management unit 150 described later.

[0073] The meta information of the answer preferably includes information that makes the difference of each robot apparatus 30 that sends the answer obvious. The weight management unit 150 described later uses the meta information of the answer to determine the weight of each answer. Therefore, the more the meta information of the answer includes information that makes the difference of each robot apparatus 30 obvious, the weight management unit 150 described later can give more different weights to the answer of each robot apparatus 30.

[0074] Note that, when the information is insufficient to generate an answer to a question, the robot apparatus 30 may request the information processing apparatus 10 or the questioner robot 1 to additionally send meta information.

[0075] The answer acquisition unit 140 acquires an answer to the question and meta information of the answer from each of the plurality of robot devices 30 via the communication unit 130. The answer acquisition unit 140 outputs the meta information of the answer to the weight management unit 150 to output the answer to the question to the final answer determination unit 160. Thus, the information processing device 10 can derive a final answer to the question by collecting the answers to the question while considering the weight determined based on the meta information of the answer.

[0076] The weight management unit 150 determines the weight of the answer based on the meta information of the answer. For example, the weight management unit 150 may determine the weight of the answer based on one type of information in the meta information of the answer.

[0077] Specifically, the weight management unit 150 may determine the weight of each answer received from the robot device 30 based on the distance between the object 2 and each robot device 30. For example, the weight management unit 150 may determine the weighting so that the answer received from the robot device 30 that is closer to the object 2 has a larger weight. This is because the robot device 30 that is closer to the object 2 is likely to grasp the object 2 more accurately, or handle the object 2 more frequently, and the received answer is considered to be highly reliable. Accordingly, the weight management unit 150 may determine the weighting of the answer so that the higher the reliability, the greater the weight the answer has.

[0078] In addition, the weight management unit 150 can determine the weight of each answer received from the robot device 30 based on the degree of structural similarity between the questioner robot 1 on which the information processing device 10 is installed and the robot device 30. For example, the weight management unit 150 can determine the weighting so that the answer received from the robot device 30 having a higher degree of structural similarity with the questioner robot 1 has a larger weight. This is because when the structures of the robot devices are different, it is likely that similar behaviors cannot be achieved. For example, it is possible that although some robot devices can grasp the object 2 using multiple arms, other robot devices cannot grasp the object 2 because they have only one arm. Therefore, by giving a larger weight to the answer received from the robot device 30 having a higher degree of structural similarity with the questioner robot 1, the information processing device 10 can make the final answer more suitable for the questioner robot 1.

[0079] In addition, the weight management unit 150 may determine the weight of each answer received from the robot device 30 based on the degree of software matching between the questioner robot 1 on which the information processing device 10 is installed and the robot device 30. For example, the weight management unit 150 may determine the weighting so that the answer received from the robot device 30 having a higher degree of software matching with the questioner robot 1 has a larger weight. This is because when the software of the robot devices is different, even if the structure is similar, it is possible that the same action or the action with the same accuracy cannot be performed. For example, it is possible that the precise movement of the object 2 that can be performed by one robot device cannot be performed by another robot device with the same accuracy. Therefore, the answer received from the robot device 30 having a higher degree of software matching with the questioner robot 1 is given a larger weight so that the information processing device 10 can make the final answer more suitable for the questioner robot 1.

[0080] In addition, the weight management unit 150 may determine the weight of each answer received from the robot device 30 based on the length of the stay period of the robot device 30 in the area where the object 2 exists. For example, the weight management unit 150 may determine the weighting so that the answer received from the robot device 30 that has stayed in the area where the object 2 exists for a longer time has a larger weight. This is because it is likely that the robot device 30 that stays in the area where the object 2 exists for a longer time grasps the object 2 more accurately or handles the object 2 more frequently, and the received answer is considered to be highly reliable. Accordingly, the weight management unit 150 may determine the weighting of the answer so that the higher the reliability, the greater the weight the answer has.

[0081] In addition, the weight management unit 150 can determine the weight of the answer for multiple types of information in the meta information of the answer, and calculate the weights of the determined multiple types to determine the weight of the answer. For example, the weight management unit 150 can weight the answer based on each piece of information shown in Table 1 below, and calculate the weighted weight for each answer to determine the weight of the answer. In Table 1 below, "-" indicates that there is no corresponding state.

[0082] Table 1

[0083]

[0084]

[0085] According to this, since the weight management unit 150 can evaluate the weight of the answer from multiple viewpoints, the weighting can be determined more finely for each robot apparatus 30 that has sent the answer.

[0086] At this time, the weight management unit 150 can change the size of the weight between the important information used for weighting and the less important information used as a reference for the meta information of the answer. Specifically, the weight management unit 150 can make the change amount of the weight for the important information larger to derive a final answer with high reliability from the received answer, and the weight management unit 150 can make the change amount of the weight for the less important information smaller. Accordingly, since the weight management unit 150 can give a larger weight to the answer with higher reliability or validity, the final answer determination unit 160 can obtain a final answer with higher reliability or validity.

[0087] In addition, the weight management unit 150 may determine the weight of each answer received from the robot apparatus 30 based on external data. Specifically, the weight management unit 150 may determine the weight of each answer received from the robot apparatus 30 based on the amount of knowledge of the robot apparatus 30 or external data on the reliability of the action. For example, the weight management unit 150 may determine the weight based on information such as a database collecting the degree of learning progress of the robot apparatus 30, a database collecting the operation time of the robot apparatus 30, or a database collecting the task success rate of the robot apparatus 30, so that answers with higher reliability received from the robot apparatus 30 have a larger weight. By weighting the answers using such external data, the weight management unit 150 improves the reliability of the final answer, and can improve the effectiveness of the final answer for the questioner robot 1.

[0088] The final answer determination unit 160 derives a final answer to the question from the received answers based on the weighting of the answers by the weight management unit 150. Specifically, the final answer determination unit 160 may derive a final answer by collecting answers to the same content and making a majority decision in consideration of the weighting.

[0089] In addition, the final answer determination unit 160 may derive a final answer to the question using only the answers received from the robot apparatus 30 existing within a certain distance from the object 2. In addition, instead of the distance from the object 2, the final answer determination unit 160 may derive a final answer to the question using only the answers received from the robot apparatus 30 existing within a certain range (e.g., in the same building, on the same floor, or in the same store area).

[0090] In addition, the final answer determination unit 160 may derive a final answer to the question using only the answers received from the robot apparatus 30 having the same or similar structure. Specifically, the traveling performance of the robot apparatus, the shape of the hand, the movable range of the arm, and the like are important factors for manipulating the object 2. For this reason, the final answer determination unit 160 may determine information more suitable for the questioner robot 1 as the final answer by using only the answers received from the robot apparatus 30 having the same or similar structure.

[0091] The driving control unit 170 controls the behavior of the questioner robot 1 on which the information processing device 10 is mounted. Specifically, the driving control unit 170 controls the driving of each motor provided in each joint of the questioner robot 1 to control the overall behavior of the questioner robot 1. In addition, the driving control unit 170 can control the driving of various modules such as a speaker, a projector, a display, or a light emitter (e.g., a light bulb, an LED, a laser, etc.) provided in the questioner robot 1 to output to the user. For example, the driving control unit 170 can control the behavior of the questioner robot 1 so as to operate an object 2, wherein the object 2 is known or the processing method of the object 2 is known through the final answer to the question.

[0092] (1.3. Operation of information processing device)

[0093] Next, we will refer to Figure 5 and Figure 6 The operation of the information processing apparatus 10 according to the present embodiment is described.

[0094] First, refer to Figure 5 The flow of the operation of the information processing apparatus 10 is described. Figure 5 : is a flowchart showing the flow of the operation of the information processing device 10 according to the present embodiment.

[0095] like Figure 5 As shown, first, the information processing device 10 recognizes the unknown object 2 through the sensor unit 110 (S110). Subsequently, the question generator 120 generates a question related to the unknown object 2 (S120). At this time, the question generator 120 can generate meta information of the question, and the meta information of the question includes information that allows the robot device 30 that has received the question to recognize the object 2.

[0096] Next, the question generated by the question generator 120 is sent to the surrounding robot devices 30 together with the meta information (S130). Thereafter, an answer to the question is returned from each of the surrounding robot devices 30 to which the question was sent, and the information processing device 10 receives the answer to the question and the meta information of the answer from each of the robot devices 30 (S140).

[0097] Subsequently, the weight management unit 150 determines the weight of each answer based on the meta-information of the answer (S150). Specifically, the weight management unit 150 weights each answer to reflect its reliability or validity based on information about the robot device 30 included in the meta-information of the answer, information about the relationship between the robot device 30 and the object 2, etc. Next, the final answer determination unit 160 derives the final answer to the question based on the weight of each answer (S160). Specifically, the final answer determination unit 160 can derive the final answer by considering the weight of each answer and collecting multiple answers with the same content to make a majority decision. Accordingly, the questioner robot 1 on which the information processing device 10 is installed can identify the unknown object 2 as a known object based on the final answer, so that the object 2 can be correctly operated (S170).

[0098] Next, refer to Figure 6 , the interaction between the information processing device 10 and the robot device 30 according to the present embodiment will be described. Figure 6 : is a sequence diagram showing the interaction between the information processing device 10 and the robot device 30 according to the present embodiment.

[0099] like Figure 6 As shown, first, the information processing device 10 identifies the unknown object 2 through the sensor unit 110 (S201). Subsequently, the question generator 120 generates questions related to the unknown object 2 (S202). At this time, the question generator 120 can generate meta information including information for identifying the object 2. Thereafter, the question generated by the question generator 120 and the meta information are sent to the surrounding robot devices 30 (S203).

[0100] Next, the robot device 30 that has received the question generates an answer to the received question using a classifier, etc. (S205). Subsequently, the robot device 30 generates meta information of the answer, which is used to weight the answer in the information processing device 10 (S207). Thereafter, the answer to the question and the meta information are sent to the information processing device 10 (S209).

[0101] Subsequently, in the information processing device 10 having received the answer from each of the plurality of robot devices 30, the weight management unit 150 determines the weight of each answer based on the meta information of each answer (S211). Next, the final answer determination unit 160 determines the final answer to the question based on the weight of each answer (S213). Accordingly, the questioner robot 1 equipped with the information processing device 10 can recognize the object 2 as a known object based on the final answer, so that the object 2 can be correctly operated (S215).

[0102] According to the above operation example, the information processing device 10 can determine what the unknown object 2 is and how to handle the unknown object 2 based on the answers from the surrounding robot devices 30 .

[0103] <2. Modification plan>

[0104] Then, refer to Figures 7 to 9 Modifications of the information processing apparatus according to the present embodiment are described. Figure 7 1 is an explanatory diagram showing that the answer from the respondent robot 3 is transmitted to the information processing device 11 . Figure 8 1 is an explanatory diagram showing that the robot 12 newly joined the network 40 accesses the information processing device 11 . Fig. 9 is an explanatory diagram showing an example of information stored in the database storage unit 100 of the information processing device 11 .

[0105] like Figure 7 As shown in FIG. 1 , the information processing device 11 according to the present modification includes a database storage unit 100 that stores questions and final answers to the questions in association with each other. The information processing device 11 is provided independently of the questioner robot 1 or the answerer robot 3. That is, the information processing device 11 according to the present modification can be freely accessed from a robot device connected to the network 40, and can be connected to the network 40. Figure 3 The information processing device 10 described in is different in that the information processing device 11 is not installed on the robot device but is provided independently.

[0106] Specifically, the information processing device 11 may receive a question and a final answer from each questioner robot 1 , the final answer being derived based on the answer received from the respondent robot 3 , and the information processing device 11 may store the question and the final answer in the database storage unit 100 .

[0107] Alternatively, when a question is sent from the questioner robot 1 to the respondent robot 3, the information processing device 11 may receive the answer and meta information of the answer to generate a final answer to the question. Thereafter, the information processing device 11 may store the question and the final answer in the database storage unit 100, and may send the generated final answer to the questioner robot 1.

[0108] The information processing device 11 according to the present modification centrally manages each question sent by the questioner robot 1 and the final answer to each question, so that the questions and the final answers to the questions can be shared among the robots connected to the network 40 .

[0109] For example, when an unknown object 2 is recognized, the questioner robot 1 sends a question related to the object 2 to the information processing device 11. When the final answer to the question sent by the questioner robot 1 is stored in the information processing device 11, the information processing device 11 sends the final answer to the received question to the questioner robot 1. Accordingly, the information processing device 11 can prevent the same question from being repeatedly sent from different questioner robots 1.

[0110] On the other hand, when the final answer to the question sent by the questioner robot 1 is not stored in the information processing device 11, the questioner robot 1 sends the question to the surrounding respondent robots 3 to generate the final answer based on the answer from the respondent robot 3. Thereafter, the questioner robot 1 can update the database of the information processing device 11 by sending the question and the final answer to the information processing device 11. That is, the information processing device 11 can update the accumulated information by accepting the latest information at any time. For example, the information processing device 11 can store a combination of a new question and a final answer in the database storage unit 100 to update the stored information.

[0111] Furthermore, the information processing device 11 may delete a combination of a question and a final answer determined to have low reliability from the database storage unit 100. For example, the reliability of the information may be determined by receiving feedback regarding the usefulness of the information from the questioner robots 1 that have referred to the information. For example, when feedback indicating that the usefulness of the information is low is received from a plurality of questioner robots 1 that have referred to the information, the information processing device 11 may delete the information.

[0112] Here, if Figure 8As shown, a case where the robot 12 newly joins the network 40 will be described. In the case where the existence and address of the information processing device 11 are answered when sending a question to the surrounding robot devices, the robot 12 recognizes the existence and address of the information processing device 11. Since the information processing device 11 can be freely accessed via the network 40, the robot 12 can access the information processing device 11 to use the information stored in the database storage unit 100 of the information processing device 11.

[0113] However, in the early stage of participation, the information processing device 11 allows the robot 12 newly joining the network 40 to access only information with a low security level, and restricts the robot 12 from accessing information with a high security level such as a password. Thereafter, as the robot 12's stay period increases, the information processing device 11 may allow the robot 12 to gradually access information with a high security level.

[0114] Furthermore, in the early stage of participation, the information processing device 11 allows the robot 12 newly joining the network 40 to only read information to the information processing device 11, and may not allow the combination of questions and final answers to be added or updated to the information processing device 11. Thereafter, as the stay period of the robot 12 increases, the information processing device 11 may allow the robot 12 to add the combination of questions and final answers.

[0115] The information processing device 11 according to the present modification may store a database in which identification information of each object 2 is accumulated by accumulating questions and final answers about each object 2. For example, Fig. 9 As shown, the database storage unit 100 may store map information indicating the positions of the objects 2 indoors. In addition, the database storage unit 100 may store information including the name, grip shape, position posture, and approachable direction of each object 2 in association with each position information of the object 2. The information processing device 11 stores such map information of the object arrangement in the database storage unit 100, thereby storing questions and final answers about the objects 2 in a format that is easy for other robot devices to use.

[0116] <3. Hardware Configuration>

[0117] In addition, reference will be made to Fig.10 The hardware configuration of the information processing device 10 according to the present embodiment is described. Fig.10 : is a block diagram showing a hardware configuration example of the information processing device 10 according to the present embodiment.

[0118] like Fig.10As shown, the information processing device 10 includes a central processing unit (CPU) 901, a read-only memory (ROM) 902, a random access memory (RAM) 903, a bridge 907, internal buses 905 and 906, an interface 908, an input device 911, an output device 912, a storage device 913, a drive 914, a connection port 915 and a communication device 916.

[0119] The CPU 901 functions as an arithmetic processing device, and controls the overall operation of the information processing device 10 according to various programs stored in the ROM 902 and the like. The ROM 902 stores programs and arithmetic parameters used by the CPU 901, and the RAM 903 temporarily stores programs used in the execution of the CPU 901, parameters that change appropriately in the execution, and the like. For example, the CPU 901 can perform the functions of the question generator 120, the answer acquisition unit 140, the weight management unit 150, and the final answer determination unit 160.

[0120] The CPU 901, ROM 902, and RAM 903 are connected to each other via a bridge 907, internal buses 905 and 906, etc. The CPU 901, ROM 902, and RAM 903 are also connected to an input device 911, an output device 912, a storage device 913, a drive 914, a connection port 915, and a communication device 916 via an interface 908.

[0121] The input device 911 includes an input device for inputting information such as a touch panel, keyboard, mouse, button, microphone, switch or joystick. The input device 911 also includes an input control circuit that generates an input signal based on the input information and outputs the input signal to the CPU 901.

[0122] The output device 912 includes a display device such as, for example, a cathode ray tube (CRT) display device, a liquid crystal display device, or an organic electroluminescent (EL) display device. In addition, the output device 912 may include an audio output device such as a speaker or headphones.

[0123] The storage device 913 is a storage device for storing data of the information processing device 10. The storage device 913 may include a storage medium, a storage device that stores data in the storage medium, a reading device that reads data from the storage medium, and a deleting device that deletes the stored data.

[0124] The drive 914 is a reader / writer for a storage medium, and is built in the information processing apparatus 10 or externally attached to the information processing apparatus 10. For example, the drive 914 reads out information stored in a removable storage medium such as a mounted magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory to output the information to the RAM 903. The drive 914 can also write information to a removable storage medium.

[0125] The connection port 915 is a connection interface including, for example, a connection port for connecting an external connection device such as a universal serial bus (USB) port, an Ethernet (registered trademark) port, an IEEE 802.11 standard port, an optical audio terminal.

[0126] The communication device 916 is a communication interface including, for example, a communication device for connecting to the network 40. The communication device 916 may be a wired or wireless LAN compatible communication device or a cable communication device that performs cable communication through a wire. The communication device 916 may perform the functions of the communication unit 130, for example.

[0127] It is also possible to create a computer program that causes hardware such as a CPU, ROM, and RAM built in the information processing device 10 to exert the same functions as each configuration of the control device according to the above-described embodiment. A storage medium storing the computer program may also be provided.

[0128] <4. Conclusion>

[0129] As described above based on the information processing device 10 according to the present embodiment, effective actions on the unknown object 2 can be determined in a short time by sending questions related to the unknown object 2 to the surrounding robot devices 30 and receiving answers to the questions from the surrounding robot devices 30.

[0130] Furthermore, based on the information processing device 10 according to the present embodiment, a final answer having higher reliability or validity can be derived from the received answers by weighting each answer based on the meta-information of the answer.

[0131] The preferred embodiments of the present disclosure have been described in detail with reference to the accompanying drawings, but the technical scope of the present disclosure is not limited to such examples. It is obvious that those skilled in the art in the technical field of the present disclosure can find various changes and modifications within the scope of the technical concept described in the claims, and it should be understood that these changes and modifications will naturally fall within the technical scope of the present disclosure.

[0132] For example, the information processing device 10 may use machine learning to perform weighting on each answer and derive a final answer based on the answer. Specifically, the information processing device 10 may include a machine learning classifier that inputs answers received from multiple robot devices 30 and meta information of the answers to output a final answer to the question.

[0133] In addition, the effects described in this specification are merely illustrative or exemplary effects, and are not restrictive. That is, in addition to or in place of the above effects, the technology according to the present disclosure may also achieve other effects that are obvious to those skilled in the art based on the description of this specification.

[0134] Note that the following configurations also belong to the technical scope of the present disclosure.

[0135] (1) An information processing device comprising:

[0136] a weight management unit that determines, with respect to an answer to a question obtained from each of a plurality of respondent robots that have received the question and meta-information of the answer, a weight of the answer based on the meta-information of the answer; and

[0137] A final answer determination unit that derives a final answer to the question from the respective answers from the plurality of respondent robots based on the weightings of the answers.

[0138] (2) The information processing device according to (1), wherein the question is transmitted from the questioner robot to each of the plurality of respondent robots together with meta information of the question.

[0139] (3) The information processing device according to (2), wherein the question is a question related to an object recognized by the questioner robot.

[0140] (4) The information processing device according to (3), wherein the meta-information of the question includes at least information about the object.

[0141] (5) The information processing device according to (4), wherein the information about the object is information obtained without contacting the object.

[0142] (6) The information processing device according to any one of (3) to (5), wherein the answer to the question is acquired one by one from each of the plurality of respondent robots.

[0143] (7) The information processing device according to any one of (3) to (6), wherein the answer to the question is generated by each of the plurality of respondent robots using a classifier.

[0144] (8) An information processing device according to any one of (3) to (7), wherein the meta-information of the answer includes at least one of the following information: information about each of the respondent robots, and information about the relationship between each of the respondent robots and the object.

[0145] (9) The information processing device according to any one of (3) to (8), wherein the weight management unit determines the weight of each of the answers based on a distance between the object and each of the respondent robots.

[0146] (10) An information processing device according to any one of (3) to (9), wherein the weight management unit determines the weight of each of the answers based on a structural similarity between the questioner robot and each of the respondent robots.

[0147] (11) An information processing device according to any one of (3) to (10), wherein the weight management unit determines the weight of each of the answers based on the degree of software matching between the questioner robot and each of the respondent robots.

[0148] (12) An information processing device according to any one of (3) to (11), wherein the weight management unit determines the weight of each of the answers based on the length of the stay period of each of the respondent robots in the area where the object exists.

[0149] (13) An information processing device according to any one of (1) to (12), wherein the final answer determination unit determines based on the weighting of each of the answers: among the multiple answers, the answer that has high validity as an answer to the question is the final answer.

[0150] (14) The information processing device according to any one of (3) to (12), wherein the questioner robot is a robot capable of operating the object.

[0151] (15) The information processing device according to (14), wherein the final answer is used for controlling the operation of the object by the questioner robot.

[0152] (16) The information processing device according to (14), wherein the information processing device is mounted on the questioner robot.

[0153] (17) The information processing device according to any one of (1) to (16) further includes a database storage unit, which stores the questions and the final answers to the questions in association with each other, wherein the database storage unit is configured to be accessible from another robot.

[0154] (18) The information processing device according to (17), wherein the final answer stored in the database storage unit is retained or deleted based on feedback from the other robot that has referred to the final answer.

[0155] (19) An information processing method comprising:

[0156] Through the arithmetic unit

[0157] With respect to an answer to a question obtained from each of a plurality of respondent robots that have received the question and meta-information of the answer, determining a weight of the answer based on the meta-information of the answer; and

[0158] Based on the weighting of the answers, a final answer to the question is derived from the respective answers from the plurality of respondent robots.

[0159] (20) A program for causing a computer to:

[0160] a weight management unit that determines, with respect to an answer to a question obtained from each of a plurality of respondent robots that have received the question and meta-information of the answer, a weight of the answer based on the meta-information of the answer; and

[0161] A final answer determination unit that derives a final answer to the question from the respective answers from the plurality of respondent robots based on weights of the answers.

[0162] Reference numerals list

[0163] 1 Questioner Robot

[0164] 2 Objects

[0165] 3. Answer Robot

[0166] 10 Information processing device

[0167] 30 Robotic Device

[0168] 40 Network

[0169] 100 database storage unit

[0170] 110 Sensor unit

[0171] 120 Question Generator

[0172] 130 Communication Unit

[0173] 140 Question acquisition unit

[0174] 150 Weight Management Unit

[0175] 160 Final answer confirmation unit

[0176] 170 Drive control unit

Claims

1. An information processing device, comprising: a weight management unit that determines, with respect to an answer to a question obtained from each of a plurality of answerer robot apparatuses and meta-information of the answer, a weight of the answer based on the meta-information of the answer, wherein the plurality of answerer robot apparatuses have received the question from the questioner robot apparatus; as well as a final answer determination unit that derives a final answer to the question from the respective answers from the plurality of answerer robot apparatuses based on weights of the answers, wherein the question is a question about an object identified by the questioner robot device, the object having a shape and being unknown to the questioner robot device, and The meta-information of the answer includes: information about the last use date, last viewing date, number of uses or frequency of use of the object by each of the respondent robot devices, or information about the distance between each of the respondent robot devices and the object.

2. The information processing device according to claim 1, wherein: The question is transmitted from the questioner robot to each of the plurality of answerer robot devices together with meta information of the question.

3. The information processing device according to claim 1, wherein: The meta information of the question includes at least information about the object.

4. The information processing device according to claim 3, wherein: The information about the object is information obtained without contacting the object.

5. The information processing device according to claim 1, wherein: Answers to the questions are obtained one by one from each of the plurality of respondent robotic devices.

6. The information processing device according to claim 1, wherein: A response to the question is generated by each of the plurality of responder robots using a classifier.

7. The information processing device according to claim 1, wherein: The answering meta-information also includes information about each of the respondent robotic devices.

8. The information processing device according to claim 1, wherein: The weight management unit determines a weight of each of the answers based on a structural similarity between the questioner robot apparatus and each of the respondent robot apparatuses.

9. The information processing device according to claim 1, wherein: The weight management unit determines a weight of each of the answers based on a degree of software matching between the questioner robot apparatus and each of the respondent robot apparatuses.

10. The information processing device according to claim 1, wherein: The weight management unit determines the weight of each of the answers based on the length of the stay period of each of the respondent robot devices in the area where the object exists.

11. The information processing device according to claim 1, wherein: The final answer determination unit determines, based on the weighting of each of the answers, that an answer having high validity as an answer to the question among the plurality of answers is the final answer.

12. The information processing device according to claim 1, wherein: The questioner robot is a robot capable of manipulating the object.

13. The information processing device according to claim 12, wherein: The final answer is used to control the operation of the object by the questioner robot.

14. The information processing device according to claim 12, wherein: The information processing device is mounted on the questioner robot device.

15. The information processing apparatus according to claim 1, further comprising a database storage unit which stores the question and the final answer to the question in association with each other, wherein The database storage unit is arranged to be accessible from another robotic device.

16. The information processing device according to claim 15, wherein: The final answer stored in the database storage unit is retained or deleted based on feedback from the other robot apparatus that has referred to the final answer.

17. An information processing method, comprising: Through the arithmetic unit, With respect to answers to a question obtained from each of a plurality of answerer robot devices and meta-information of the answers, determining a weight of the answers based on the meta-information of the answers, wherein the plurality of answerer robot devices have received the question from the questioner robot device; as well as deriving a final answer to the question from the respective answers from the plurality of respondent robots based on the weighting of the answers, wherein the question is a question about an object identified by the questioner robot device, the object having a shape and being unknown to the questioner robot device, and The meta-information of the answer includes: information about the last use date, last viewing date, number of uses or frequency of use of the object by each of the respondent robot devices, or information about the distance between each of the respondent robot devices and the object.

18. A storage medium storing a program which, when executed, causes a computer to function as: a weight management unit that determines, with respect to an answer to a question obtained from each of a plurality of answerer robot apparatuses and meta-information of the answer, a weight of the answer based on the meta-information of the answer, wherein the plurality of answerer robot apparatuses have received the question from the questioner robot apparatus; and a final answer determination unit that derives a final answer to the question from the respective answers from the plurality of answerer robot apparatuses based on weights of the answers, in, The question is a question about an object identified by the questioner robotic device, the object having a shape and being unknown to the questioner robotic device, and The meta-information of the answer includes: information about the last use date, last viewing date, number of uses or frequency of use of the object by each of the respondent robot devices, or information about the distance between each of the respondent robot devices and the object.

Images