People searching method and robot

Abstract

The embodiment of the invention provides a person searching method and a robot, which are used for solving the problems that people cannot be accurately searched and the person searching accuracy is relatively low when people are searched in the prior art. According to the embodiment of the invention, the robot searches people according to the received sound, so that the problem that face recognition cannot be carried out due to limited image acquisition angles is avoided, and the user of the target identifier can be accurately determined according to whether the voiceprint feature existing in the sound is matched with the target voiceprint feature corresponding to the user of the target identifier, so that the user experience is improved. And according to the setting positions of the plurality of sound acquisition devices and the time of respectively acquiring the sound matched with the target voiceprint feature, the distance and the orientation between a sound maker, namely the user of the target identifier, and the robot can be accurately determined, so that the accuracy of searching people can be improved. The scheme provided by the invention has real-time performance, reliability and trustworthiness.

Description

Technical Field

[0001] The present application relates to the technical fields of image processing, speech recognition and intelligent service robots, and in particular to a method and robot for finding people. Background Art

[0002] With the development of science and technology and the progress of human society, intelligent technology has penetrated into people's daily lives, bringing great convenience to people's lives. Intelligent robots are a major product that has emerged in the development of intelligent technology. At present, 25 countries in the world are involved in the research of service robots, and more than 40 service robots of 7 types have entered experimental and semi-commercial applications. Service robots are divided into household service robots and commercial service robots. The more common service robots on the market are mainly outdoor cleaning robots, sweeping robots, welcoming robots, shopping guide robots, smart home control robots, nurse assistants, etc. These robots have brought many conveniences to users in different fields. Among them, mobile household service robots are the mainstream in the current market. Intelligent following is the main application demand in the research field of mobile household service robots, which better realizes the intelligence of human-computer interaction and completes the designated tasks issued by users.

[0003] At present, the existing robot search methods can be generally divided into robot fixed-point image collection search and robot active search, that is, fixed-point search and active search. The difference between fixed-point search and active search lies in whether the robot moves. Among them, the fixed-point search method includes: the robot is fixed at a certain position, and the collection device installed on the robot or other collection devices collect images of passing people, and compare them with the target face through the face recognition algorithm, so as to obtain the location information of the target user corresponding to the target face. The active search method includes: the robot patrols and moves along a pre-set route, such as the living room, balcony, kitchen, study, bedroom, and living room. When the robot moves, the collection device set on the robot collects images during the movement, and compares them with the target face through the face recognition algorithm, so as to obtain the location information of the target user corresponding to the target face.

[0004] However, due to the limited collection angle of the collection device, the methods in the related technology can only cover a limited range. In addition, the accuracy of the current face recognition algorithm has certain requirements on the angle of the face. The use of fixed collection equipment will reduce the accuracy of the face recognition algorithm, and the adverse factors such as light and occlusion at fixed points need to be actively eliminated by humans. Summary of the invention

[0005] The embodiments of the present application provide a method and a robot for finding people, which are used to solve the problem that the prior art cannot find people accurately and has a low accuracy rate in finding people.

[0006] In a first aspect, an embodiment of the present application provides a method for finding a person, the method comprising:

[0007] If a search command is received, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature corresponding to the user with the target identifier;

[0008] If sounds collected by multiple sound collection devices are received, it is determined whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sounds, and if so, a first distance and a first position of the sender of the sound matching the target voiceprint feature relative to the robot are determined according to the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected;

[0009] The robot is controlled to move based on the first distance and the first orientation.

[0010] In a second aspect, an embodiment of the present application further provides a robot, including a controller, a plurality of sound collection devices, and a mobile chassis;

[0011] The multiple sound collecting devices are used to collect sounds and send the collected sounds to the controller;

[0012] The mobile chassis is used to carry the robot and move under the control of the controller;

[0013] The controller is used for, if receiving a search command, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature saved corresponding to the user with the target identifier; if receiving sounds collected by multiple sound collection devices, determining whether there is a sound in the received sound whose voiceprint feature matches the target voiceprint feature, and if so, determining a first distance and a first orientation of the sender of the sound matching the target voiceprint feature relative to the robot based on the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature were respectively collected; and controlling the robot to move based on the first distance and the first orientation.

[0014] In the embodiment of the present application, if the robot receives a search command, it obtains the target identification of the user carried in the search command, and obtains the target voiceprint features corresponding to the user of the target identification; if the robot receives sounds collected by multiple sound collection devices, it determines whether there is a sound whose voiceprint features match the target voiceprint features in the received sound, and if so, determines the first distance and first orientation of the sender of the sound matching the target voiceprint features relative to the robot according to the setting positions of the multiple sound collection devices and the time when the sounds matching the target voiceprint features are collected respectively; and controls the robot to move based on the first distance and the first orientation. Since the robot searches for people according to the received sound in the embodiment of the present application, the problem of being unable to perform face recognition due to the limited angle of image collection is avoided, and the user of the target identification can be accurately determined according to whether the voiceprint features in the sound match the target voiceprint features corresponding to the user of the target identification, and the distance and orientation between the sender of the sound, i.e., the user of the target identification, and the robot can be accurately determined according to the setting positions of the multiple sound collection devices and the time when the sounds matching the target voiceprint features are collected respectively, thereby improving the accuracy of searching for people. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A schematic diagram of a person-searching process provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of a process of searching for a person based on sound provided in an embodiment of the present application;

[0018] Figure 3 A schematic diagram of a robot path provided in an embodiment of the present application;

[0019] Figure 4 A schematic diagram of a robot structure provided in an embodiment of the present application;

[0020] Figure 5 A detailed schematic diagram of a person-searching process provided in an embodiment of the present application;

[0021] Figure 6 A schematic diagram of a process for identifying a user with a target identifier provided in an embodiment of the present application;

[0022] Figure 7 A schematic diagram of a person-searching process provided in an embodiment of the present application;

[0023] Figure 8 A schematic diagram of a process of controlling a robot to move provided in an embodiment of the present application;

[0024] Fig. 9 A schematic diagram of another robot structure provided in an embodiment of the present application;

[0025] Fig.10 A schematic diagram of the structure of a person-searching device provided in an embodiment of the present application;

[0026] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be described clearly and completely in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of this application.

[0028] In order to accurately search for a person, the embodiments of the present application provide a method and a robot for searching for a person.

[0029] The method for searching for people includes: if a search command is received, then the target identifier of the user carried in the search command is obtained, and the target voiceprint feature corresponding to the user with the target identifier is obtained; if sounds collected by multiple sound collection devices are received, then it is determined whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sound, and if so, the first distance and first orientation of the sender of the sound matching the target voiceprint feature relative to the robot are determined according to the setting positions of the multiple sound collection devices and the time when the sounds matching the target voiceprint feature are respectively collected; and the robot is controlled to move based on the first distance and the first orientation. Thus, searching for people is accurately achieved.

[0030] Figure 1 A schematic diagram of a person-searching process provided in an embodiment of the present application, the process includes the following steps:

[0031] S101: If a person-searching instruction is received, a target identifier of a user carried in the person-searching instruction is obtained, and a target voiceprint feature corresponding to the user with the target identifier is obtained.

[0032] The person-searching method provided in the embodiment of the present application is applied to a robot.

[0033] In order to accurately realize the search for people, the robot can first receive the search for people instruction, and then start to search for people after receiving the search for people instruction. When the operator needs to search for people, he can issue the search for people instruction through voice or the display screen of the robot. Specifically, the operator can select a preset page on the display screen of the robot, and select the target identifier of the user to be found on the preset page. After selecting the target identifier, the operator can click a preset button, such as the "Search" button, and the robot can receive the search for people instruction, wherein the user's target identifier can be the user's name, the user's photo, etc.; or the operator can make a voice, and the voice contains the user's target identifier, such as "Find ***", wherein "***" is the user's target identifier. After receiving the voice, the robot determines that it has received the search for people instruction.

[0034] If the robot receives a search command, it obtains the target identifier of the user carried in the search command. If the operator issues the search command through voice, the robot can identify whether there is any pre-saved identifier of any user in the voice corresponding to the search command. If so, the identifier of the existing user is determined as the target identifier of the user carried in the search command. If the operator issues the search command through the robot's display screen, the search command carries the target identifier of the user.

[0035] In order to accurately find people, the robot can identify them through voiceprint features. Therefore, the robot locally stores voiceprint features for users with different IDs. After obtaining the target ID of the user, the robot can obtain the target voiceprint features stored for the user with the target ID, so as to facilitate the subsequent identification of the user based on the target voiceprint features, that is, to start the search operation. If the robot does not save the target voiceprint features for the user with the target ID, it means that the user has not registered the voiceprint features, and the search cannot be performed at this time. The robot can output a preset message for prompting registration, informing the operator that the user is not registered and needs to register first.

[0036] It should be noted that the user of the target identification needs to be registered in advance, which is specifically performed by the registration module of the robot. After the registration is completed, the registration module saves the voiceprint features of the user of the target identification during registration.

[0037] S102: If sounds collected by multiple sound collection devices are received, determine whether there is a sound whose voiceprint characteristics match the target voiceprint characteristics in the received sounds. If so, determine the first distance and first orientation of the person who makes the sound matching the target voiceprint characteristics relative to the robot based on the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint characteristics are respectively collected.

[0038] In order to accurately search for people, the robot can receive sounds collected by multiple sound collection devices. The process of collecting sounds by multiple collection devices can also be called sound pickup. If sounds collected by multiple sound collection devices are received, the voiceprint features present in the sounds are identified. Specifically, the voiceprint features of the sounds are identified through a voiceprint recognition algorithm, and it is determined whether there is a sound in the sound whose voiceprint features match the target voiceprint features. If there is a sound that matches the target voiceprint features, it is determined that the user currently to be found has made the sound, and the speaker of the sound is the user of the target identification.

[0039] In order to determine the distance and orientation of the target identified user relative to the robot, the robot can determine the emitter of the sound matching the target voiceprint feature, that is, the first distance and first orientation of the user relative to the robot, based on the setting positions of multiple sound collection devices and the time when the sound matching the target voiceprint feature is collected respectively. Among them, the sound collection device can be a microphone. In order to accurately determine the first distance and first orientation of the emitter of the sound relative to the robot, the designer of the robot can arrange multiple sound collection devices according to a certain rule to form an array. Specifically, the robot can use sound source localization technology to process the setting positions of multiple collection devices of the machine and the time when the sound is collected respectively, and then determine the distance and orientation of the emitter of the sound relative to the robot. Among them, how the robot specifically uses sound source localization technology to determine the distance and orientation of the emitter of the sound relative to the robot is an existing technology and will not be repeated here.

[0040] It should be noted that the person who makes the sound whose existing voiceprint feature matches the target voiceprint feature is the target identified user, so the first distance and first position of the target identified user relative to the robot can be accurately determined.

[0041] Among them, the voice processing module of the robot can process the sound in the environment of the robot in real time, and determine the distance and direction of the user relative to the robot based on the received sound for target identification. The voice processing module runs in a separate thread.

[0042] S103: Control the robot to move based on the first distance and the first orientation.

[0043] After determining the first distance and first position of the person who makes the sound whose voiceprint features match the target voiceprint features relative to the robot, the robot can control the robot to move based on the first distance and the first position, so that the robot moves to the side of the user with the target identification. Specifically, the robot can control its chassis to turn to the first position and move the first distance.

[0044] It should be noted that after determining the first distance and the first orientation, the robot can call the navigation function through the interface to realize its own movement and automatically avoid obstacles during the movement.

[0045] In the embodiments of the present application, the voiceprint recognition technology is equivalent to solving the technical difficulty of failing to find a person when searching based on images alone. Specifically, when the shooting angle and lighting conditions of the acquisition device are not good, resulting in large fluctuations in the accuracy of user identification based on image target identification, the correct target user can still be searched through voiceprint recognition and sound source localization technology.

[0046] Figure 2 A schematic diagram of a process of searching for a person based on sound provided in an embodiment of the present application, the process includes the following steps:

[0047] S201: Receive a person-searching instruction.

[0048] S202: Obtain the target identifier of the user carried in the person-searching instruction.

[0049] S203: Obtain the target voiceprint feature saved corresponding to the target identified user.

[0050] S204: Receive sounds collected by multiple sound collection devices.

[0051] S205: Determine whether there is a voiceprint feature matching the target voiceprint feature in the received sound, if so, execute S206, if not, execute S204.

[0052] S206: Determine a first distance and a first orientation of a person who makes a sound matching the target voiceprint feature relative to the robot based on the locations of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are collected.

[0053] S207: Control the robot to move based on the first distance and the first orientation.

[0054] Since in the embodiment of the present application, the robot searches for people based on the received sound, thus avoiding the problem of limited image acquisition angles that cannot perform face recognition, and according to whether the voiceprint features in the sound match the target voiceprint features corresponding to the target identified user, the target identified user can be accurately determined, and according to the setting positions of multiple sound collection devices and the time when the sounds matching the target voiceprint features are collected respectively, the distance and direction between the sender of the sound, i.e. the target identified user and the robot can be accurately determined, thereby improving the accuracy of searching for people. The scheme involved in the present application is real-time and reliable, and meets the trustworthy characteristics.

[0055] In order to improve the accuracy of searching for people, based on the above embodiment, in the embodiment of the present application, after obtaining the target identifier of the user carried in the searching instruction, the method further includes:

[0056] According to the order of pre-saved preset points, after controlling the robot to move to the corresponding preset points in sequence, determine the user's position based on the first image captured by the acquisition device installed on the robot, and control the robot to move to the position. According to the second image containing the user captured by the acquisition device, determine whether the user is the user of the target identification. If so, determine that the search process is over, otherwise, control the robot to move to the next preset point of the preset point.

[0057] In order to improve the accuracy of searching for people, after obtaining the target identifier carried in the search instruction, the robot can move to the side of each user in the space where the robot is located to determine whether the user is the user of the target identifier.

[0058] Specifically, since the space where the robot is located may contain multiple subspaces, that is, multiple rooms, and it is impossible to identify every user in a certain subspace, the robot can control the robot to move to the corresponding preset points in sequence according to the order of the preset points saved in advance, wherein the order of each preset point can be the preset point of the kitchen, the preset point of the living room, the preset point of the elderly room, and the preset point of the bedroom; or it can be other custom orders. When the robot moves to a preset point, it can be considered to enter the room corresponding to the preset point. After the robot moves to the corresponding preset point, according to the first image captured by the acquisition device installed on the robot, the position of each user in the first image in the actual scene is determined. Specifically, the robot can locally save a pre-trained position recognition model, and the robot can input the first image into the position recognition model, and the position recognition model outputs the distance and orientation of each user in the first image relative to the acquisition device that captures the first image. The robot can determine the position of each user according to the acquired distance and orientation of each user relative to the acquisition device that captures the first image. Among them, the acquisition device for acquiring the first image is installed on the robot, so the distance and orientation of each user relative to the acquisition device are the distance and orientation of each user relative to the robot. Specifically, the robot can move to the corresponding user's side in sequence according to the distance of each user relative to the acquisition device. After moving to the side of a certain user, according to the second image containing the user acquired by the acquisition device, it is determined whether the user in the second image is the user of the target identification. If the user in the second image is the user of the target identification, it is determined that the search process is over. If the user in the second image is not the user of the target identification, the robot is controlled to move to the next preset point of the preset point.

[0059] Among them, in order to identify whether the user in the second image is the target identified user, the robot can use a face recognition algorithm to identify the facial features in the second image, and determine whether the facial features match the facial features of the target identified user. If they match, it means that the user in the second image is the target identified user, otherwise, it means that the user in the second image is not the target identified user.

[0060] Among them, the order of each preset point can be called the room order, and the room order can be automatically generated according to the room number in the point cloud map. The point cloud map is a Simultaneous Localization and Mapping (SLAM) map. In order to determine each preset point, the operator can turn on the robot's mapping function mode and push the robot to move around the home to cover the entire house as much as possible. In the mapping process, space identification points can also be set. For example, when the robot is pushed to the kitchen, it is manually or voice-set that this is the kitchen. The robot will record the current location and the corresponding kitchen, where the location recorded by the robot is the preset point corresponding to the kitchen. Specifically, the operation can be performed on the robot's mapping model.

[0061] In the embodiment of the present application, the robot specifically detects the user in the current environment through a human body detection algorithm, calculates the user's location, and cooperates with the path planning module to formulate a search path corresponding to each preset point.

[0062] Figure 3 A schematic diagram of a robot path provided in an embodiment of the present application.

[0063] Figure 3 The order of the preset points is the preset point corresponding to room 1, the preset point corresponding to room 2, and the preset point corresponding to room 3. If the first image collected at the preset point corresponding to room 1 contains three users, the robot is controlled to move to the three users in turn. If the three users are not the target users, the robot is controlled to move to the preset point corresponding to room 2. If the first image collected at the preset point corresponding to room 2 contains one user, the robot is controlled to move to the user. If the user is not the target user, the robot is controlled to move to the preset point corresponding to room 3. If the first image collected at the preset point corresponding to room 2 contains two users, the robot is controlled to move to the two users.

[0064] The method provided in the embodiment of the present application is generally used in a closed environment, such as a home environment, or other closed environments. The specific search operation can be performed by the main control module of the robot, wherein the main control module is based on the Android operating system, so the search function can be embedded by developing an Android application (Application, APP), and installed in the main control module of the robot to realize the whole house search function. For the robot main control module installed with other types of operating systems, such as the Robot Operating System (Robot Operating System, ROS) and the Linus Operating System (Linux Operating System, Linux), the application layer program under the corresponding system can also be written to implement it, and there is no restriction on the type of operating system used by the robot main control module.

[0065] By adopting the method provided in the embodiment of the present application, the robot can autonomously reach a position closer to the user, thereby solving the recognition difficulty caused by the long distance between the robot and the user in the related technology.

[0066] Figure 4 A schematic diagram of a robot structure provided in an embodiment of the present application.

[0067] Depend on Figure 4 It can be seen that the robot includes a display screen, an algorithm board, a camera module, a sound pickup module, a torso frame and a mobile chassis, wherein the algorithm board is used for searching for people, the camera module is the collection device described in the embodiment of the present application, and the sound pickup module is the multiple sound collection devices described in the embodiment of the present application.

[0068] Specifically, the robot's path planning module is responsible for planning the robot's walking path during the search process. In this application, the robot's search path is divided into two parts: the sequence of preset points for the robot to walk and the path planning when it moves to each user's location after reaching a preset point.

[0069] When searching for people, the robot can search for the user with the target ID in a certain environment where the map has been completed. During the movement of the robot, infrared sensors and other sensors can detect whether there are obstacles in front of it, and then perform path planning. The method of the embodiment of the present application enables the robot to search for the user with the target ID like a human, especially in the scenario where the user needs to be notified of a message in person, which can shorten the distance between humans and machines, enable the robot to better integrate into the family, and improve the user experience.

[0070] In order to accurately determine the location of the user, based on the above embodiments, in the embodiment of the present application, the determining the location of the user according to the first image captured by the acquisition device installed on the robot includes:

[0071] According to a first image captured by a capture device installed on the robot, an area where a user is located in the first image is identified; and a depth image captured by a depth capture device installed on the robot is acquired, and according to depth data of an area corresponding to the area in the depth image, a second distance and a second orientation of the user relative to the robot are determined;

[0072] The position of the user in the point cloud map is determined according to the current position of the robot in the point cloud map, the second distance and the second orientation.

[0073] In order to determine the location of the user, after acquiring the first image captured by the acquisition device installed on the robot, the robot can identify the area where the user is located in the first image. Specifically, the robot can pre-store a user recognition model locally, and the robot can input the first image into the user recognition model to obtain the area where each user is located in the first image output by the user recognition model. This process is equivalent to detection through a human body detection algorithm. The area where the user is located can be a rectangular frame that completely contains the user. In order to accurately determine the location of the user, the robot is also equipped with a depth acquisition device. The robot can obtain the depth image acquired by the depth acquisition device when the acquisition device acquires the first image, wherein the depth acquisition device and the acquisition device can acquire the same area in the environment, and the robot can determine the depth data of the area corresponding to the area in the depth image for each user in the first image, and determine the second distance and second orientation of the user in the area relative to the robot's depth acquisition device according to the depth data of the area corresponding to the area in the depth image, wherein the depth acquisition device is installed on the robot, so the second distance and the second orientation are also the distance and orientation of the user in the area relative to the robot, specifically, the robot can determine the distance and orientation of the user in the area relative to the robot according to the minimum depth data in the area corresponding to the area in the depth image. Specifically, how to determine the distance and orientation of the object corresponding to the depth data relative to the depth acquisition device according to the depth data is a prior art and will not be described in detail here.

[0074] After determining the second distance and the second orientation of the user relative to the robot, the robot can determine the user's position in the point cloud map according to its current position in the point cloud map, the second distance and the second orientation. After obtaining the position of each user in the point cloud map, the robot can control itself to go to the corresponding position by calling the navigation function through the interface.

[0075] Specifically, the robot's path planning module controls the robot chassis to move to a preset point, detects the number of users in the currently captured image, where the number of users can be considered to be the number of users in the room corresponding to the preset point, and calculates the user's position in the point cloud map in combination with the depth image captured by the depth acquisition device, and moves to the side of each user in turn.

[0076] It should be noted that after receiving the search command, the voice processing module responsible for searching through sound and the visual processing module responsible for identifying images start working at the same time. The two operate independently. When either side finds the user with the target identifier, the search is considered successful.

[0077] In an embodiment of the present application, when the robot moves to the location of each user in turn, it can automatically plan the best path through navigation obstacle avoidance technology and path planning algorithm, so that when the robot reaches the user's location, it can identify the user. Compared with the carpet search method of fixed routes, this method greatly improves the accuracy and efficiency of finding people.

[0078] Figure 5 A detailed schematic diagram of a person-searching process provided in an embodiment of the present application, the process includes the following steps:

[0079] in, Figure 5 The process of searching for people based on images is described in .

[0080] S501: Receive a person-searching instruction.

[0081] S502: Obtain the target identifier of the user carried in the person-searching instruction.

[0082] S503: According to the sequence of preset points saved in advance, control the robot to move to the next preset point.

[0083] S504: Determine the position of each user according to the first image captured by the capturing device installed on the robot.

[0084] S505: Move to the location of the unidentified user, and obtain a second image captured by a capture device and containing the user at the location.

[0085] S506: Determine whether the user included in the second image is the user of the target identification, if yes, execute S507, if not, execute S508.

[0086] S507: Determine that the person search process is finished.

[0087] S508: Determine whether there is an unrecognized user among each user in the first image, if so, execute S505, if not, execute S509.

[0088] S509: Determine whether there is a next preset point of the preset point in the sequence of preset points, if yes, execute S503, if no, execute S510.

[0089] S510: Determine that the person search process fails.

[0090] In order to accurately search for a person, based on the above embodiments, in the embodiment of the present application, after obtaining the target identifier of the user carried in the search instruction, the method further includes:

[0091] Obtaining target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of the target user's clothing, hairstyle, body shape, and other non-facial features;

[0092] The determining, based on the second image containing the user acquired by the acquisition device, whether the user is the user of the target identification comprises:

[0093] Whether the user is the user of the target identification is determined according to whether the second image containing the user collected by the collection device contains the target facial features or the target static features.

[0094] In order to find people accurately, the robot locally stores corresponding facial features and static features for users with different identifications, wherein the static features are used to represent at least one of the corresponding user's clothing, hairstyle, and other non-facial features. After obtaining the target identification of the user carried in the search instruction, the robot can obtain the facial features and static features saved for the target identified user. For the sake of distinction, the acquired facial features and static features are respectively referred to as target facial features and target static features. Among them, the target facial features and target static features are correspondingly saved by the robot when the user with the target identification is registered. If the target facial features saved for the target identified user are not obtained, the preset registration prompt information is output. If the target static features saved for the target identified user are not obtained, there is no need to prompt registration.

[0095] When identifying whether the user in the second image is the target user, the robot can identify the facial features in the second image through the face detection and recognition algorithm, and compare the facial features with the target facial features. If the facial features are consistent with the target facial features, it is determined whether the second image contains the target facial features, indicating that the user in the second image is the target user. Among them, there may be a situation where the second image only contains the user but not the user's face. At this time, in order to improve the accuracy of finding people, the robot can determine whether the second image contains the target static features. If the second image contains the target static features, it is determined that the user in the second image is the target user. If the second image does not contain the target facial features and does not contain the target static features, it means that the user in the second image is not the target user.

[0096] Among them, the correspondence between the user of the target identifier and the target voiceprint features, target facial features and target static features can be saved in the robot's local database, or saved in a server that interacts with the robot. If saved in the server, after the robot obtains the target identifier carried in the search instruction, it sends the target identifier to the server, and the server obtains the target voiceprint features, target facial features and target static features corresponding to the user of the target identifier, and the server sends the target voiceprint features, target facial features and target static features to the robot.

[0097] Figure 6 A schematic diagram of a process for identifying a user with a target identifier provided in an embodiment of the present application, the process includes the following steps:

[0098] S601: Acquire a second image including a user.

[0099] S602: Determine whether there are target facial features or target static features in the second image. If yes, execute S603; if no, execute S604.

[0100] S603: Determine that the user in the second image is the target identified user.

[0101] S604: Determine that the user in the second image is not the target identified user.

[0102] In order to improve the accuracy of searching for people, based on the above embodiments, in the embodiment of the present application, the method further includes:

[0103] If the second image contains the target facial feature but does not contain the target static feature, the static feature contained in the second image is determined as another target static feature, and the another target static feature is saved for the user of the target identifier.

[0104] Since a user's clothing, hairstyle, body shape and other non-facial features are prone to change, in order to further improve the accuracy of finding people, when determining that the user in the second image is the target identified user, the static features in the second image can be determined as another target static feature, thereby facilitating the identification of the target identified user in subsequent searching.

[0105] Specifically, if the second image contains the target facial features, it means that the user in the second image is the target identified user; if the second image contains the target facial features but does not contain the target static features, it means that the current static features of the target identified user have changed. The robot determines the static features contained in the second image as another target static feature, and saves the other target static feature for the target identified user, thereby facilitating the subsequent accurate identification of the target identified user and improving the efficiency of finding people.

[0106] In actual application, after the robot reaches a preset point, it identifies the position of each user in the first image. When controlling itself to move to the position of a certain user, the robot can first use a face recognition algorithm to identify the facial features in the second image collected at that position. If the facial features are consistent with the target facial features, it is determined that the person is found successfully. At this time, information can be broadcast. Specifically, information related to the successful search can be broadcast, or information related to the communication with the target user can be broadcast.

[0107] In the present application, the identification of the target user identified by static features can be called a re-identification algorithm. This algorithm is equivalent to relying on the overall posture of the user to identify whether it is the target user. Since the identification of the target user identified by static features does not require the collection of images containing faces, it is equivalent to user queries that can cross between collection devices, which can improve the accuracy of finding people. It should be noted that if the target static features are not saved for the target user, in the process of finding people, after the target user is found, the static features in the image containing the target user are saved. Specifically, the target user and the static features can be saved in the sample library. When the target user is found next time, the target user can be identified based on the static features, that is, the target user can be identified by static features other than facial features, reducing the risk of being unable to find people when the face cannot be captured, thereby improving the accuracy and efficiency of finding people.

[0108] In order to accurately search for people, based on the above embodiments, in the embodiment of the present application, the method further includes:

[0109] If the sounds collected by the multiple sound collection devices are not received, or the received sounds do not contain sounds whose voiceprint features match the target voiceprint features, a preset sound message is output to inquire whether there is a user with the target identification.

[0110] It is possible that the target user is just in a quiet state and does not make any sound. In order to accurately train, the robot can output a preset sound message to inquire whether there is a target user, thereby prompting the target user to make a sound, so as to facilitate the identification of the target user based on the voiceprint feature. Specifically, the robot can obtain a pre-saved prompt text, and use the text-to-speech (TTS) function to generate a preset sound message from the prompt text, wherein when the preset sound message is output, it can be broadcasted through the microphone installed on the robot.

[0111] Specifically, if the robot does not receive the sounds collected by the multiple sound collection devices, or there is no sound whose voiceprint features match the target voiceprint features among the received sounds, the robot outputs the preset sound information to inquire whether there is a user with the target identification in the current scene. The preset sound information may also be output when the robot fails to find the user with the target identification through both sound and image.

[0112] Figure 7 A schematic diagram of a person-searching process provided in an embodiment of the present application.

[0113] Depend on Figure 7 It can be seen that after receiving the search command, the robot searches for the person based on the image and the sound respectively. Specifically, how to search for the person based on the image and the sound has been described in the above embodiment and will not be described in detail here. If any method is used to successfully search for the person, the process ends. If none of the methods is successful, the preset sound information is output.

[0114] In order to accurately find a person, based on the above embodiments, in an embodiment of the present application, controlling the robot to move based on the first distance and the first orientation includes:

[0115] According to the current position of the robot in the point cloud map, the first distance and the first orientation, determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature; and control the robot to move based on the target position.

[0116] After obtaining the first distance and first orientation of the person who makes the sound matching the target voiceprint feature relative to the robot, the robot can determine the target position of the person who makes the sound matching the target voiceprint feature in the point cloud map according to its current position in the point cloud map, the first distance and the first orientation. This process is to map the position of the person who makes the sound matching the target voiceprint feature to the point cloud map. The robot then controls itself to move based on the target position.

[0117] In order to improve the user experience of the target identifier, on the basis of the above embodiments, in an embodiment of the present application, controlling the robot to move based on the first distance and the first orientation includes:

[0118] Determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance;

[0119] The robot is controlled to move based on the moving distance and the first orientation.

[0120] If the robot controls itself to move to the position of the user with the target identification, it may collide with the user with the target identification. Therefore, after determining the first distance and the first position of the person who makes the sound matching the target voiceprint feature relative to the robot, the robot determines the difference between the first distance and the preset distance threshold, determines the difference as the movement distance of the robot, and controls the movement of the robot based on the movement distance and the first position. Specifically, the robot is controlled to move the movement distance toward the first position, so that the robot stops beside the user with the target identification.

[0121] Figure 8 A schematic diagram of a process of controlling a robot to move provided in an embodiment of the present application, the process comprising the following steps:

[0122] S801: Determine a difference between a first distance and a preset distance threshold.

[0123] S802: Determine the difference as the moving distance.

[0124] S803: Control the robot to move based on the moving distance and the first position.

[0125] When the robot moves based on the moving distance and the first orientation, the robot can be stopped when the distance between the robot and the user of the target identification is the preset distance threshold.

[0126] In order to accurately find a person, based on the above embodiments, in the embodiment of the present application, the method further includes:

[0127] If an obstacle is encountered during movement, the movement will stop and the preset information will be output.

[0128] Since the robot may encounter obstacles during its movement, preventing it from moving, in order to accurately find people, the robot can output preset information to explain the current situation, making it easier for the operator or other users to adjust the robot and then find people.

[0129] Fig. 9 A schematic diagram of the structure of another robot provided in an embodiment of the present application, the structure comprising:

[0130] Controller 901, multiple sound collection devices 902, mobile chassis 903;

[0131] The multiple sound collecting devices 902 are used to collect sounds and send the collected sounds to the controller 901;

[0132] The mobile chassis 903 is used to carry the robot and move under the control of the controller;

[0133] The controller 901 is used for, if receiving a search command, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature corresponding to the user with the target identifier; if receiving sounds collected by multiple sound collection devices, determining whether there is a sound in the received sound whose voiceprint feature matches the target voiceprint feature, and if so, determining a first distance and a first orientation of the sender of the sound matching the target voiceprint feature relative to the robot based on the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature were respectively collected; and controlling the robot to move based on the first distance and the first orientation.

[0134] Furthermore, the robot also includes a collection device 904;

[0135] The acquisition device 904 is used to acquire the first image and the second image, and send the first image and the second image to the controller 901;

[0136] The controller 901 is also used to control the robot to move to corresponding preset points in sequence according to the pre-saved sequence of preset points, determine the user's position according to a first image captured by a capture device installed on the robot, and control the robot to move to the position, and determine whether the user is the user of the target identification according to a second image containing the user captured by the capture device. If so, the search process is determined to be over, otherwise, the robot is controlled to move to the next preset point of the preset point.

[0137] Further, the controller 901 is specifically used to identify the area where the user is located in the first image according to the first image acquired by the acquisition device installed on the robot; and obtain the depth image acquired by the depth acquisition device installed on the robot, and determine the second distance and second orientation of the user relative to the robot according to the depth data of the area corresponding to the area in the depth image;

[0138] The position of the user in the point cloud map is determined according to the current position of the robot in the point cloud map, the second distance and the second orientation.

[0139] The controller 901 is further used to obtain the target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of the target user's clothing, hairstyle, body shape and other non-face features;

[0140] Furthermore, the controller 901 is specifically configured to determine whether the user is the user of the target identification according to whether the second image containing the user acquired by the acquisition device contains the target facial features or the target static features.

[0141] Furthermore, the controller 901 is specifically used to determine the static feature contained in the second image as another target static feature if the second image contains the target facial feature but does not contain the target static feature, and save the other target static feature for the user of the target identifier.

[0142] Furthermore, the robot also includes a sound output device 905;

[0143] The sound output device 905 is used to receive preset sound information sent by the controller and output the sound information;

[0144] The controller 901 is specifically used to output preset sound information through a sound output device to inquire whether there is a user with a target identification if no sound collected by multiple sound collection devices is received or there is no sound whose voiceprint characteristics match the target voiceprint characteristics among the received sounds.

[0145] Furthermore, the controller 901 is specifically used to determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature according to the current position of the robot in the point cloud map, the first distance and the first orientation; and control the robot to move based on the target position.

[0146] Further, the controller 901 is specifically configured to determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance;

[0147] The robot is controlled to move based on the moving distance and the first orientation.

[0148] Furthermore, the controller 901 is specifically configured to stop moving and output preset information if an obstacle is encountered during movement.

[0149] Fig.10 A schematic diagram of a person-searching device provided in an embodiment of the present application is provided, wherein the device comprises:

[0150] The receiving and obtaining module 1001 is used for obtaining the target identification of the user carried in the search instruction and obtaining the target voiceprint feature corresponding to the user with the target identification if a search instruction is received;

[0151] The processing module 1002 is configured to determine whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sound if sounds collected by multiple sound collection devices are received, and if so, determine a first distance and a first orientation of a sender of the sound matching the target voiceprint feature relative to the robot according to the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected;

[0152] The control module 1003 is used to control the robot to move based on the first distance and the first orientation.

[0153] Furthermore, the processing module 1002 is also used to determine the user's position based on a first image captured by a capture device installed on the robot, and control the robot to move to the position after controlling the robot to move to corresponding preset points in sequence according to the pre-saved sequence of preset points, and to determine whether the user is the user of the target identification based on a second image containing the user captured by the capture device. If so, the search process is determined to be over; otherwise, the robot is controlled to move to the next preset point of the preset point.

[0154] Furthermore, the processing module 1002 is specifically used to identify the area where the user is located in the first image based on the first image captured by the acquisition device installed on the robot; and obtain the depth image captured by the depth acquisition device installed on the robot, and determine the second distance and the second orientation of the user relative to the robot based on the depth data of the area corresponding to the area in the depth image; determine the position of the user in the point cloud map based on the current position of the robot in the point cloud map, the second distance and the second orientation.

[0155] Furthermore, the processing module 1002 is further used to obtain the target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of the target user's clothing, hairstyle, body shape and other non-face features;

[0156] The processing module 1002 is specifically used to determine whether the user is the user of the target identification according to whether the second image containing the user collected by the collection device contains the target facial features or the target static features.

[0157] Furthermore, the processing module 1002 is also used to determine the static feature contained in the second image as another target static feature if the second image contains the target facial feature but does not contain the target static feature, and save the other target static feature for the user of the target identifier.

[0158] Furthermore, the processing module 1002 is also used to output preset sound information to inquire whether there is a user with the target identification if no sound collected by multiple sound collection devices is received, or there is no sound whose voiceprint characteristics match the target voiceprint characteristics among the received sounds.

[0159] Furthermore, the control module 1003 is specifically used to determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature according to the current position of the robot in the point cloud map, the first distance and the first orientation; and control the robot to move based on the target position.

[0160] Furthermore, the control module 1003 is specifically configured to determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance; and control the robot to move based on the moving distance and the first orientation.

[0161] Furthermore, the processing module 1002 is also used to stop moving and output preset information if an obstacle is encountered during the moving process.

[0162] Fig.11 The present invention provides a schematic diagram of an electronic device structure according to an embodiment of the present invention. Based on the above embodiments, the present invention further provides an electronic device, such as Fig.11 As shown, it includes: a processor 1101, a communication interface 1102, a memory 1103 and a communication bus 1104, wherein the processor 1101, the communication interface 1102, and the memory 1103 communicate with each other through the communication bus 1104;

[0163] The memory 1103 stores a computer program. When the program is executed by the processor 1101, the processor 1101 performs the following steps:

[0164] If a search command is received, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature corresponding to the user with the target identifier;

[0165] If sounds collected by multiple sound collection devices are received, it is determined whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sounds, and if so, a first distance and a first position of the sender of the sound matching the target voiceprint feature relative to the robot are determined according to the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected;

[0166] The robot is controlled to move based on the first distance and the first orientation.

[0167] Furthermore, the processor 1101 is also used to control the robot to move to corresponding preset points in sequence according to the pre-saved sequence of preset points, determine the position of the user according to a first image captured by a capture device installed on the robot, and control the robot to move to the position, and determine whether the user is the user of the target identification according to a second image containing the user captured by the capture device. If so, it is determined that the search process is ended, otherwise, the robot is controlled to move to the next preset point of the preset point.

[0168] Further, the processor 1101 is specifically configured to identify, based on a first image captured by a capture device installed on the robot, an area where a user is located in the first image; and obtain a depth image captured by a depth capture device installed on the robot, and determine, based on depth data of an area corresponding to the area in the depth image, a second distance and a second orientation of the user relative to the robot;

[0169] The position of the user in the point cloud map is determined according to the current position of the robot in the point cloud map, the second distance and the second orientation.

[0170] Furthermore, the processor 1101 is further configured to obtain target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of clothing, hairstyle, body shape and other non-facial features of the target user;

[0171] The processor 1101 is specifically configured to determine whether the user is the user of the target identification according to whether the second image containing the user acquired by the acquisition device contains the target facial features or the target static features.

[0172] Furthermore, the processor 1101 is also used to determine the static feature contained in the second image as another target static feature if the second image contains the target facial feature but does not contain the target static feature, and save the other target static feature for the user of the target identifier.

[0173] Furthermore, the processor 1101 is also used to output preset sound information to inquire whether there is a user with the target identification if no sound collected by multiple sound collection devices is received, or there is no sound whose voiceprint characteristics match the target voiceprint characteristics among the received sounds.

[0174] Furthermore, the processor 1101 is specifically used to determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature according to the current position of the robot in the point cloud map, the first distance and the first orientation; and control the robot to move based on the target position.

[0175] Further, the processor 1101 is specifically configured to determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance;

[0176] The robot is controlled to move based on the moving distance and the first orientation.

[0177] Furthermore, the processor 1101 is also used to stop moving and output preset information if an obstacle is encountered during movement.

[0178] The communication bus mentioned in the above server can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0179] The communication interface is used for communication between the above electronic device and other devices.

[0180] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0181] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor (Network Processor, NP), etc.; it can also be a digital signal processing processor (Digital Signal Processing, DSP), an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc.

[0182] On the basis of the above embodiments, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program executable by an electronic device, and when the program is run on the electronic device, the electronic device implements the following steps when executing:

[0183] The memory stores a computer program, and when the program is executed by the processor, the processor performs the following steps:

[0184] If a search command is received, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature corresponding to the user with the target identifier;

[0185] If sounds collected by multiple sound collection devices are received, it is determined whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sounds, and if so, a first distance and a first position of the sender of the sound matching the target voiceprint feature relative to the robot are determined according to the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected;

[0186] The robot is controlled to move based on the first distance and the first orientation.

[0187] In a possible implementation manner, after obtaining the target identifier of the user carried in the person-searching instruction, the method further includes:

[0188] According to the order of pre-saved preset points, after controlling the robot to move to the corresponding preset points in sequence, determine the user's position based on the first image captured by the acquisition device installed on the robot, and control the robot to move to the position. According to the second image containing the user captured by the acquisition device, determine whether the user is the user of the target identification. If so, determine that the search process is over, otherwise, control the robot to move to the next preset point of the preset point.

[0189] In a possible implementation manner, determining the location of the user according to the first image captured by a capture device installed on the robot includes:

[0190] According to a first image captured by a capture device installed on the robot, an area where a user is located in the first image is identified; and a depth image captured by a depth capture device installed on the robot is acquired, and according to depth data of an area corresponding to the area in the depth image, a second distance and a second orientation of the user relative to the robot are determined;

[0191] The position of the user in the point cloud map is determined according to the current position of the robot in the point cloud map, the second distance and the second orientation.

[0192] In a possible implementation manner, after obtaining the target identifier of the user carried in the person-searching instruction, the method further includes:

[0193] Obtaining target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of the target user's clothing, hairstyle, body shape, and other non-facial features;

[0194] The determining, based on the second image containing the user acquired by the acquisition device, whether the user is the user of the target identification comprises:

[0195] Whether the user is the user of the target identification is determined according to whether the second image containing the user collected by the collection device contains the target facial features or the target static features.

[0196] In a possible implementation, the method further includes:

[0197] If the second image contains the target facial feature but does not contain the target static feature, the static feature contained in the second image is determined as another target static feature, and the another target static feature is saved for the user of the target identifier.

[0198] In a possible implementation, the method further includes:

[0199] If the sounds collected by the multiple sound collection devices are not received, or the received sounds do not contain sounds whose voiceprint features match the target voiceprint features, a preset sound message is output to inquire whether there is a user with the target identification.

[0200] In a possible implementation, controlling the robot to move based on the first distance and the first orientation includes:

[0201] According to the current position of the robot in the point cloud map, the first distance and the first orientation, determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature; and control the robot to move based on the target position.

[0202] In a possible implementation, controlling the robot to move based on the first distance and the first orientation includes:

[0203] Determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance;

[0204] The robot is controlled to move based on the moving distance and the first orientation.

[0205] In a possible implementation, the method further includes:

[0206] If an obstacle is encountered during movement, the movement will stop and the preset information will be output.

[0207] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0208] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0209] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0210] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0211] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for finding a person, characterized in that: Applied to a robot, the method comprises: If a search command is received, obtaining a target identifier of the user carried in the search command, and obtaining a target voiceprint feature corresponding to the user with the target identifier; If sounds collected by multiple sound collection devices are received, it is determined whether there is a sound whose voiceprint feature matches the target voiceprint feature in the received sounds, and if so, a first distance and a first position of the sender of the sound matching the target voiceprint feature relative to the robot are determined according to the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected; The robot is controlled to move based on the first distance and the first orientation.

2. The method according to claim 1, characterized in that After obtaining the target identifier of the user carried in the person-searching instruction, the method further includes: According to the order of pre-saved preset points, after controlling the robot to move to the corresponding preset points in sequence, determine the user's position based on the first image captured by the acquisition device installed on the robot, and control the robot to move to the position. According to the second image containing the user captured by the acquisition device, determine whether the user is the user of the target identification. If so, determine that the search process is over, otherwise, control the robot to move to the next preset point of the preset point.

3. The method according to claim 2, characterized in that Determining the position of the user according to the first image captured by a capture device installed on the robot includes: According to a first image captured by a capture device installed on the robot, an area where a user is located in the first image is identified; and a depth image captured by a depth capture device installed on the robot is acquired, and according to depth data of an area corresponding to the area in the depth image, a second distance and a second orientation of the user relative to the robot are determined; The position of the user in the point cloud map is determined according to the current position of the robot in the point cloud map, the second distance and the second orientation.

4. The method according to claim 2, characterized in that After obtaining the target identifier of the user carried in the person-searching instruction, the method further includes: Obtaining target facial features and target static features saved for the target identified user, wherein the static features are used to represent at least one of the target user's clothing, hairstyle, body shape, and other non-facial features; The determining, based on the second image containing the user acquired by the acquisition device, whether the user is the user of the target identification comprises: Whether the user is the user of the target identification is determined according to whether the second image containing the user collected by the collection device contains the target facial features or the target static features.

5. The method according to claim 4, characterized in that The method further comprises: If the second image contains the target facial feature but does not contain the target static feature, the static feature contained in the second image is determined as another target static feature, and the another target static feature is saved for the user of the target identifier.

6. The method according to claim 1, characterized in that The method further comprises: If the sounds collected by the multiple sound collection devices are not received, or the received sounds do not contain sounds whose voiceprint features match the target voiceprint features, a preset sound message is output to inquire whether there is a user with the target identification.

7. The method according to claim 1, characterized in that The controlling the robot to move based on the first distance and the first orientation comprises: According to the current position of the robot in the point cloud map, the first distance and the first orientation, determine the target position in the point cloud map of the person who makes the sound matching the target voiceprint feature; and control the robot to move based on the target position.

8. The method according to claim 1, characterized in that: The controlling the robot to move based on the first distance and the first orientation comprises: Determine a difference between the first distance and a preset distance threshold, and determine the difference as a moving distance; The robot is controlled to move based on the moving distance and the first orientation.

9. The method according to any one of claims 1 to 8, characterized in that: The method further comprises: If an obstacle is encountered during movement, the movement will stop and the preset information will be output.

10. A robot, characterized in that: Including controller, multiple sound collection devices, and mobile chassis; The multiple sound collecting devices are used to collect sounds and send the collected sounds to the controller; The mobile chassis is used to carry the robot and move under the control of the controller; The controller is configured to, if receiving a person-searching instruction, obtain a target identifier of a user carried in the person-searching instruction, and obtain a target voiceprint feature corresponding to the user of the target identifier; If sounds collected by multiple sound collection devices are received, it is determined whether there is a sound with a voiceprint feature matching the target voiceprint feature in the received sound. If so, the first distance and first orientation of the person who makes the sound matching the target voiceprint feature relative to the robot are determined based on the setting positions of the multiple sound collection devices and the times when the sounds matching the target voiceprint feature are respectively collected; and the robot is controlled to move based on the first distance and the first orientation.

Images