Human-computer interaction methods, devices, electronic devices, and follower robots
By detecting the following status in real time and actively executing interactive data, the interaction problem of the following robot in abnormal situations in complex environments is solved, improving user experience and efficiency.
Patent Information
- Application Number
- CN202611321662.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
Existing follow-up robots are prone to getting lost or trapped in complex home environments, causing users to frequently turn back to check or search for the robot, thus reducing the user experience.
By monitoring the follow status in real time, identifying preset abnormal states, and actively executing interactive data such as voice, light, and actions, the system alerts users to abnormal situations and guides them to adjust their behavior or find the robot. This includes subdividing different abnormal states and using a countdown mechanism to optimize the interaction.
It improves the convenience of interaction between the robot and the user, enhances the user experience, reduces the user's psychological burden, and increases the efficiency and success rate of the following process.
Smart Images

Figure CN123103647A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of artificial intelligence technology, and in particular to a human-computer interaction method, device, electronic device, and follower robot. Background Technology
[0002] With the development of artificial intelligence technology, intelligent interactive devices such as children's companion robots and pet robots are gradually becoming more widespread. To provide a more natural interactive experience, follow-me robots with autonomous walking and following capabilities have emerged on the market. These follow-me robots typically perceive their environment through sensors such as vision and lidar, and can actively follow the user's movements like a pet, thereby enhancing the interactivity and fun between the robot and the user.
[0003] However, existing following robots are prone to getting lost or trapped in complex home environments, such as when encountering obstacles, when the user suddenly changes direction, or when moving too fast. When such situations occur, existing following robots typically simply stop moving or spin in place, waiting for the user to intervene. This forces users to frequently turn around to check the robot's status, or even to go back and search for the lost robot, significantly reducing the user experience and limiting the application scenarios for the following function. Summary of the Invention
[0004] This invention provides a human-computer interaction method, device, electronic device, and follower robot to address the deficiencies in related technologies.
[0005] This invention provides a human-computer interaction method, comprising: In follow mode, the follow status is detected in real time; When the following state belongs to a preset abnormal state, the first interaction data corresponding to the preset abnormal state is determined; the preset abnormal state includes at least one of the following states: user distance is less than preset distance and user loss state. Execute the first interactive data; the first interactive data includes an abnormal status prompt.
[0006] According to a human-computer interaction method provided by the present invention, the user loss state includes a first loss state and a second loss state; The duration of the first tracking loss state is less than or equal to the first preset duration; The duration of the second tracking error is greater than the first preset duration and less than or equal to the second preset duration; The first tracking loss state and the second tracking loss state correspond to different first interaction data.
[0007] According to a human-computer interaction method provided by the present invention, the real-time detection of the following state further includes: If the following state contains multiple instances of the user being lost, then the second interaction data is executed; the second interaction data includes prompts indicating that the user has disappeared multiple times.
[0008] According to a human-computer interaction method provided by the present invention, the execution of the first interaction data and the execution of the second interaction data further include: Start the countdown for the third preset duration; If all the following states within the third preset time period belong to the same preset abnormal state, then the third interactive data is executed; the third interactive data includes an exit prompt.
[0009] According to a human-computer interaction method provided by the present invention, after starting a countdown of a third preset duration, the method further includes: If the following state changes within the third preset time period, the countdown is canceled at the moment of the state change, and the changed following state is re-evaluated to determine whether it belongs to the preset abnormal state.
[0010] According to a human-computer interaction method provided by the present invention, the execution of the first interaction data, the execution of the second interaction data, and the execution of the third interaction data are all followed by: Receive user feedback data; Respond to the user feedback data.
[0011] According to a human-computer interaction method provided by the present invention, the preset abnormal state further includes an obstacle encounter state, which includes a first obstacle state, a second obstacle state, and a third obstacle state. The obstacle sizes in the first obstacle state, the second obstacle state, and the third obstacle state increase from small to large; The first obstacle state, the second obstacle state, and the third obstacle state each correspond to different first interaction data.
[0012] The present invention also provides a human-computer interaction device, comprising: The detection module is used to detect the following status in real time during follow mode; The determination module is used to determine the first interaction data corresponding to the preset abnormal state when the following state belongs to the preset abnormal state; the preset abnormal state includes at least one of the following states: user distance is less than preset distance and user loss state. An interaction module is used to execute the first interaction data; the first interaction data includes an abnormal status prompt.
[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the human-computer interaction method as described above.
[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the human-computer interaction method as described above.
[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the human-computer interaction method as described above.
[0016] The present invention also provides a follower robot, comprising: a robot body; The robot body is used to execute any of the human-computer interaction methods described above.
[0017] The human-computer interaction method, device, electronic device, and following robot provided by this invention monitor the status in real time in following mode and actively execute first interactive data for preset abnormal states. This enables the following robot to proactively initiate first interactive data containing abnormal state prompts to the user during the following process, based on abnormal situations such as the user getting too close or losing the user. This prompts the user to assist the following robot in getting out of trouble, guides the user to adjust their walking speed in time or turn around to look for the robot, and thus assists the robot in getting out of the following predicament through human-computer collaboration. This can solve the problem of poor user convenience caused by the lack of feedback when following abnormalities occurs in the prior art, significantly improve the convenience of interaction between the following robot and the user and the user experience, and eliminate the need for the user to frequently turn around to observe the following robot's following status, effectively reducing the psychological burden on the user during the following process. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in this invention or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is one of the flowcharts of the human-computer interaction method provided by the present invention.
[0020] Figure 2 This is the second flowchart of the human-computer interaction method provided by the present invention.
[0021] Figure 3 This is a schematic diagram of the human-computer interaction device provided by the present invention.
[0022] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation
[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0024] Figure 1 This is a flowchart illustrating a human-computer interaction method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: S1, in follow mode, detects the follow status in real time; S2, when the following state belongs to a preset abnormal state, determine the first interaction data corresponding to the preset abnormal state; the preset abnormal state includes at least one of the following states: user distance is less than a preset distance and user loss state. S3, execute the first interactive data; the first interactive data includes an abnormal status prompt.
[0025] Specifically, the human-computer interaction method provided in this embodiment of the invention is executed by a human-computer interaction device, which can be configured in a follower robot with a following function. The follower robot includes, but is not limited to, a child companion robot, a pet robot, an electronic pet, an intelligent service robot, or any intelligent terminal that integrates motion control, environmental perception, and human-computer interaction modules, etc., without specific limitations here.
[0026] First, step S1 is executed, where the robot, in follow mode, monitors its following status in real time. Follow mode refers to a working mode where the robot actively tracks the user's movement trajectory and automatically accompanies them based on user instructions. This follow mode defines the relative positional following relationship between the robot and the user, enabling the robot to follow autonomously without requiring the user to manually move the robot during movement.
[0027] The following state refers to the real-time performance of the robot following the user during operation in follow mode. It can be characterized by user position, user distance, and visual tracking stability. The detection of the following state can be achieved through the collaborative use of multiple sensors carried by the robot. For example, visual sensors can be used to acquire scene images; human skeleton recognition or posture detection algorithms can be used for human identification and tracking; scene images can be used to determine the presence of obstacles ahead; a positioning system can continuously acquire the user's spatial position information; LiDAR or depth cameras can be used to measure the user's distance; and an Inertial Measurement Unit (IMU) can be used to monitor the stability of the robot's motion. By fusing and analyzing this multi-source sensor data, the robot can build a comprehensive understanding of its current following state.
[0028] After the user instructs the robot to enter follow mode, the robot can first prompt the user to remain stationary for a first specified duration. The robot then continuously monitors the scene image and identifies the user within it, treating them as the target to follow. After the first specified duration, a visual notification can be displayed to indicate successful target recognition. Subsequently, a second specified duration can be counted down before the robot begins tracking and locating the user. Both the first and second specified durations can be set as needed; their values can be the same or different. For example, both can be set to 3 seconds. The first specified duration provides the robot with sufficient scene imagery and data to determine the target, while the second specified duration provides the user with preparation time, ensuring they enter a safe following area and providing a stable sampling period for the visual algorithm to lock onto target features.
[0029] Then, step S2 is executed. If the following state is a preset abnormal state, the first interaction data corresponding to the preset abnormal state can be determined. The following state can include a normal following state and an abnormal following state. The normal following state refers to the state in which the robot's following task is progressing smoothly. The abnormal following state refers to the state in which the robot encounters an abnormality in the process of performing the following task. This can include the state of encountering an obstacle, the state of the user distance being less than a preset distance, and the state of losing the user.
[0030] The "user distance less than preset distance" state refers to a situation where the actual spatial distance between the following robot and the user is lower than a safety threshold. This may cause the following robot to be unable to effectively avoid obstacles or to collide with the user. The preset distance is the safety threshold, which can be set as needed, such as 10cm, 15cm, etc. The "user lost" state refers to a situation where the following robot temporarily or completely loses its location and tracking of the user due to reasons such as the user turning too fast, entering behind an obstruction, or changes in ambient light.
[0031] Preset abnormal states are scenarios that require user intervention or control, thereby avoiding failures or safety hazards caused by the robot's simple mechanical following. Preset abnormal states can be at least one of the abnormal following states mentioned above, such as the state where the user distance is less than a preset distance or the state where the user is lost. In addition, it can also include the state of encountering obstacles, etc.
[0032] The robot can be pre-configured with an interaction strategy library, which stores the first interaction data matching various preset abnormal states. When it is determined that the following state belongs to a preset abnormal state, the first interaction data matching the preset abnormal state is determined from the interaction strategy library.
[0033] Here, the first interactive data refers to the interactive content used by the robot to communicate with and guide users in solving problems. This includes abnormal status prompts, clearly conveying to the user what the specific problem is. The interaction of the first interactive data can be achieved through at least one of the following: voice, light, action, and on-screen emoticons. For example, an orange breathing light indicates an obstacle, a blue flashing light indicates losing track, and a solid red light indicates the user is too close. Another example is that when the voice prompt says "Here, here," the robot turns its head towards the user and nods; when the voice prompt says "I'm here," it raises a robotic arm or spins in place. If the robot has a display screen, it can display corresponding cartoon emoticons, which can include expressions such as distress, anxiety, and happiness.
[0034] When the user is less than the preset distance, the abnormal status message could be something like, "You're too close, please move away," "You're too close, move away quickly," or "I'm going to bump into you, get out of the way!" and it could stop moving forward at the same time. When the user is lost, the abnormal status message could be, "I can't find you, please return to my line of sight." When encountering an obstacle, the abnormal status message could be, "I can't get through! Help me!" "Something's blocking my way, I can't get through!" or "We're going to bump into each other! Lead me around it!" In this way, the following robot no longer silently executes incorrect actions, but actively prompts the user with the problem.
[0035] Finally, in step S3, the robot invokes the hardware actuator to execute the first interactive data. For example, the text-based first interactive data can be converted into an audio signal and played through the speech synthesis module. The content could be: "You're walking too fast, wait for me!" Alternatively, a confused or sad animated expression can be displayed on the screen to indicate to the user that the robot cannot follow properly. Simultaneously, flashing light effects can be used to provide a more intuitive notification to the user.
[0036] The human-computer interaction method provided in this invention monitors the status in real time in follow mode and actively executes first interaction data for preset abnormal states. This enables the following robot to proactively initiate first interaction data containing abnormal state prompts to the user based on abnormal situations such as the user getting too close or losing the user. This prompts the user to assist the following robot in getting out of trouble, guiding the user to adjust their walking speed or turn back to find the robot. In this way, the robot is assisted in getting out of following difficulties through human-computer collaboration. This solves the problem of poor user convenience caused by the lack of feedback when following abnormalities occurs in the prior art. It significantly improves the convenience of interaction and user experience between the following robot and the user, eliminating the need for the user to frequently turn back to observe the following robot's status and effectively reducing the psychological burden on the user during the following process.
[0037] Based on the above embodiments, the lost user status includes a first lost status and a second lost status; The duration of the first tracking loss state is less than or equal to the first preset duration; The duration of the second tracking error is greater than the first preset duration and less than or equal to the second preset duration; The first tracking loss state and the second tracking loss state correspond to different first interaction data.
[0038] Specifically, to further enhance the follow-up robot's ability to handle scenarios where users are lost, the state of the lost user can be more finely categorized. The lost user state can be divided into a first lost state and a second lost state based on the duration of the loss. The core basis for this classification is the duration of the loss, that is, the length of time elapsed from when the follow-up robot first lost the user, i.e., when it confirmed that the user could not be located, to the current moment. By introducing different time thresholds, the follow-up robot can adopt differentiated interaction strategies for different degrees of loss, thereby achieving a more intelligent and human-like response.
[0039] The first tracking loss state corresponds to a short tracking loss period, specifically, a tracking loss period less than or equal to a first preset duration. The first preset duration is a pre-defined time threshold, the value of which can be flexibly configured according to the actual application scenario and user experience requirements, for example, it can be set to 1 second, 2 seconds, or 5 seconds. When the tracking loss period is less than or equal to the first preset duration, it can be assumed that the user has likely just left the robot's field of vision or sensor range, but has not yet gone far away. For example, if the user turns around to get something, the robot still has a high probability of finding the user again within a short time. At this point, the robot determines that it is currently in the first tracking loss state.
[0040] The second tracking loss state corresponds to a situation where the tracking loss duration is relatively long but still within an acceptable range. Specifically, it means the tracking loss duration is longer than the first preset duration but less than or equal to the second preset duration. The second preset duration is also a preset time threshold, and its value must be greater than the first preset duration; for example, it can be set to 10 seconds or 15 seconds. When the tracking loss duration is longer than the first preset duration but less than or equal to the second preset duration, it indicates that the user has been away for some time. The tracking robot may find it difficult to immediately resume tracking using its regular search strategies, requiring more explicit guidance or response from the user. At this point, the tracking robot determines that it is currently in the second tracking loss state. It should be noted that if the tracking loss duration exceeds the second preset duration, it may trigger higher-level emergency handling logic, such as stopping tracking, remaining in place, or automatically returning to the charging station.
[0041] When the robot detects a user has lost track of its follower, it starts an internal timer to continuously record the duration of the loss. Based on the timer's reading, the robot dynamically determines whether the current user loss state is a first or second loss state. For example, suppose the first preset duration is 1 second and the second preset duration is 10 seconds. When the robot begins to lose track of a user, the timer starts counting from 0. When the timer reading is between 0 and 1 second, the robot determines it to be a first loss state and executes the corresponding first interaction data. The first interaction data corresponding to the first loss state can include encouraging waiting prompts, such as a cheerful tone or voice message like "Wait for me, I'm coming right away!", "Slow down, I can't catch up!", or "You're walking too fast! Wait for me~", to encourage the user to pause or slow down. If the user does not return or re-identify within 1 second, the timer reading continues to increase. When it exceeds 1 second but is less than 10 seconds, the robot determines it to be a second loss state and executes the first interaction data corresponding to the second loss state. The first interactive data corresponding to the second lost state can include help-seeking and urging prompts, such as issuing more urgent voice prompts like "Where are you? Where did you go?", "Where are you? Come to me quickly!", "I can't find you, come find me quickly!", etc., to encourage users to actively search for and follow the robot.
[0042] In this embodiment of the invention, by subdividing the user's state of being lost, the monotony and ineffectiveness of a single loss-of-follow notification can be avoided. This allows the following robot to exhibit a progressive response, ranging from gentle prompting to urgent assistance, when faced with loss situations of varying severity. This human-like interaction design is more in line with human social habits and can effectively guide users to take appropriate actions. Moreover, this hierarchical mechanism based on the duration of being lost makes the following robot's interaction strategy more precise and efficient. It avoids issuing overly strong warnings when the user has only briefly left, and also avoids using an overly mild tone after the user has actually gone far away, thus achieving a good balance between reducing user interference and increasing the success rate of problem resolution.
[0043] Based on the above embodiments, the real-time detection of the following state further includes: If the following state contains multiple instances of the user being lost, then the second interaction data is executed; the second interaction data includes prompts indicating that the user has disappeared multiple times.
[0044] Specifically, to address the complex scenario of repeated user loss, a mechanism for monitoring the frequency of user loss can be further introduced. During real-time monitoring of the following status, the robot not only determines whether it is currently in a user loss state but also statistically records the frequency of such occurrences. Here, multiple user loss states refer to the robot detecting a user loss state more than once within a complete following task cycle. When the robot loses a user multiple times consecutively or intermittently, it indicates that the current following environment may be complex, or the user's behavior pattern exceeds the robot's normal adaptability. At this point, a second interactive data is triggered. This second interactive data differs from the first; its core content is a prompt indicating multiple user disappearances. This prompt aims to convey a higher-level signal to the user that the robot has repeatedly encountered difficulties and requires more active cooperation or a change in the user's current behavior pattern. For example, the prompt could be something like, "Hey, where did you go again?" or "Oh dear, I can't find you again! Can you stop and wait for me?", or even a humorous complaint like, "I was all ready! Where did you go again?". The tone and content of the notification when a user disappears multiple times will be clearer and more emphasized than the notification when the user is lost for the first time, in order to make the user pay enough attention.
[0045] In practical implementation, the follower robot can maintain a counter, which increments each time a user is detected to have lost track of another user. When the counter value exceeds one, the follower robot stops executing the first interaction data for that single instance of losing track and instead executes the second interaction data. Furthermore, to prevent repeated triggering due to the same reason within a short period, the follower robot can reset the counter after executing the second interaction data, or use a sliding window method to count only the number of times a user has lost track of another user within a recent time period.
[0046] In this embodiment of the invention, by introducing a higher-level abnormal state of repeated loss of following, the problem of users becoming annoyed or ignoring the robot due to repeated loss of following in complex environments and sending the same prompts can be solved. The robot can interact in a more obvious and urgent way, effectively attracting the user's attention. Moreover, this frequency-based differentiated response mechanism makes the robot's interactive behavior more intelligent and human-like. It can remember its recent mistakes and adjust its communication strategy accordingly, thereby guiding the user to take more effective measures to help the robot resume normal following, ultimately improving the efficiency and success rate of human-computer interaction.
[0047] Based on the above embodiments, the execution of the first interactive data and the execution of the second interactive data further include: Start the countdown for the third preset duration; If all the following states within the third preset time period belong to the same preset abnormal state, then the third interactive data is executed; the third interactive data includes an exit prompt.
[0048] Specifically, to further enhance the follower robot's ability to cope with persistent abnormal states, countdown mechanisms can be introduced during both the execution of the first and second interactive data. When the follower robot begins executing the first or second interactive data, it will not wait indefinitely for user response, but will simultaneously start a third preset countdown.
[0049] The third preset duration is a pre-set time threshold, the value of which can be flexibly configured according to the actual application scenario and the user's expected response speed, such as 30 seconds, 40 seconds, etc. This countdown represents the reasonable time limit within which the following robot is willing to wait for the user to react to the current preset abnormal state.
[0050] During the countdown, the robot continuously monitors changes in the following status. If, within the third preset time period, the following status remains within the same preset abnormal state—that is, the type of abnormal state detected by the robot does not change (e.g., it remains in a state of losing the user or remaining in a state where the user distance is less than a preset distance)—then the robot determines that the user has failed to effectively resolve the abnormal issue. At this point, the robot will no longer simply repeat the first interaction data but will execute the third interaction data. The content of the third interaction data is an exit prompt, the core purpose of which is to clearly convey to the user that, due to the prolonged unresolved issue, the robot will soon exit the current following mode or take other conservative actions.
[0051] The exit prompt can vary depending on the type of preset abnormal state. For example, if the following state is always in the "encountering an obstacle" state within the third preset time period, the exit prompt could be "I can't get through, I'll go play somewhere else first"; if the following state is always in the "the user is less than the preset distance" state within the third preset time period, the exit prompt could be "Aren't you coming? I'll go play by myself then"; if the following state is always in the "lost user" state within the third preset time period, the exit prompt could be "Nobody's here, I'll play by myself for a while", "La la la, I'll play by myself for a while!", "I haven't been able to find you for too long, I need to rest first", "If you don't pay attention to me, I'll go home by myself", or "I'm going to explore first! Call me when you find me!" etc., thus prompting the user to take action at the last minute.
[0052] It is important to note that the emphasis of "same type of preset abnormal state" is on the uniformity of the preset abnormal state type, rather than the subtle changes in the state. For example, changing from the first lost-follow state to the second lost-follow state, although the degree of lost-following is more severe, still belongs to the broad category of lost-follower user states, and therefore is still considered a "same type of preset abnormal state".
[0053] In practice, the countdown can begin simultaneously with the execution of the first or second interactive data. For example, when the robot detects that it has lost track of the user and issues a "Wait for me" prompt, a 30-second countdown immediately begins. If the robot fails to recapture the user within 30 seconds, meaning it remains in the "lost user" state, then at the moment the countdown reaches zero, the robot executes the third interactive data, plays an exit prompt, and then stops following or enters standby mode. Conversely, if a fundamental change in the preset abnormal state occurs during the countdown, such as from "lost user" to "user distance less than preset distance," the countdown can be reset or terminated, as the new abnormal state may require different processing logic.
[0054] During the countdown, the robot can repeat the prompt every 5 seconds and attempt to move a short distance in the direction where the user was last lost. If the user is not found by the end of the countdown, it will execute a third interactive message, such as "I can't find you, I'm going home now!", and automatically navigate to the charging dock or the safe location where it last interacted with the user.
[0055] In this embodiment of the invention, when the user remains unresponsive, the robot will not endlessly repeat the same interactive content. Instead, if the following state remains in the same preset abnormal state for a third preset time period, it will execute third interactive data including an exit prompt. This avoids the robot getting stuck in an infinite loop of prompts, thus saving energy and computing resources. Furthermore, the countdown mechanism provides the user with a clear time expectation, enhancing the predictability of the interaction and giving the user a final chance to recover. By executing the exit prompt, the robot can gracefully end the currently failed following task, rather than silently stopping or getting stuck in a deadlock state, improving the user experience and the sense of control over the robot's behavior.
[0056] Based on the above embodiments, a countdown of a third preset duration is initiated, followed by: If the following state changes within the third preset time period, the countdown is canceled at the moment of the state change, and the changed following state is re-evaluated to determine whether it belongs to the preset abnormal state.
[0057] Specifically, to further enhance the follower robot's adaptability to dynamic environments, the operation rules of the countdown mechanism can be optimized. After the countdown of the third preset duration begins, the follower robot does not mechanically wait for the countdown to end, but continuously monitors changes in the following state. Here, a change in the following state means that the current following state detected by the follower robot has undergone a substantial change compared to the following state at the start of the countdown. This change can be the elimination of preset abnormal states, such as the follower robot returning from a user-lost state to a normal following state; it can also be a change in the type of preset abnormal state, such as changing from a user-lost state to a user-distance-less-preset-distance state; or it can be a change in the same type of preset abnormal state, such as changing from a first-lost-state to a second-lost-state.
[0058] When the robot detects a change in the following status during the countdown, it cancels the currently running third preset duration countdown. This is because the original countdown was set based on the following status before the change. When the following status has changed, the original countdown is no longer applicable, and continuing to count down might lead to unreasonable decisions. For example, if a 30-second countdown was started due to the user being lost, but the user suddenly appears in the robot's field of vision at the 5th second, the user loss status is resolved, and the robot should immediately stop the countdown, perhaps playing a cheerful message like "Haha, found you! Let's keep going!", and then resume normal following mode.
[0059] After canceling the countdown, the robot doesn't stop processing; instead, it immediately re-evaluates whether the changed following state falls under a preset abnormal state. This re-evaluation means the robot needs to analyze and categorize the changed following state. If the changed state is a preset abnormal state—for example, changing from "lost user" to "user distance less than preset distance"—the robot should re-execute the first interaction data corresponding to the changed state and can start a new countdown. If the changed state is not a preset abnormal state, meaning the robot has returned to normal following, then no further exception handling is needed, and the robot can continue its normal following task.
[0060] In this embodiment of the invention, the following robot can be endowed with the ability to respond quickly to unexpected situations, avoiding lag or unreasonable behavior caused by fixed countdowns. For example, when a user only leaves briefly and then returns, the following robot can immediately cancel the countdown and resume normal operation, instead of stubbornly executing the exit prompt, thereby improving the naturalness and smoothness of the interaction. Moreover, the dynamic adjustment mechanism allows the following robot to flexibly respond to the evolution of the following state. For example, when it changes from a state of losing the user to a state where the user distance is less than a preset distance, the following robot can seamlessly switch its interaction strategy instead of continuing to use the old strategy for losing the user, thereby ensuring that the most appropriate feedback is provided to the user at any time, improving the intelligence and reliability of human-computer interaction.
[0061] Based on the above embodiments, the execution of the first interactive data, the execution of the second interactive data, and the execution of the third interactive data are all further included afterward: Receive user feedback data; Respond to the user feedback data.
[0062] Specifically, in order to build a more complete two-way human-computer interaction closed loop, after the robot executes the first interaction data, the second interaction data, and the third interaction data, a mechanism for receiving and responding to user feedback can be introduced respectively.
[0063] After the follow-up robot provides an abnormal status prompt to the user via voice, image, or other means, it enters a listening or waiting state to receive user feedback data. This user feedback data refers to any form of response the user makes to the follow-up robot's initial interaction data, including but not limited to voice commands, gestures, touch sensing, facial expressions, or remote commands sent via a mobile terminal. For example, after the follow-up robot issues a "I can't find you" prompt, the user can say "I'm here," wave to the follow-up robot, or directly click the "I'm here" button on the terminal application (APP) to reassure the follow-up robot.
[0064] After receiving user feedback data, the follower robot will receive and parse the data, and respond based on the parsing results. Responding to user feedback data means that the follower robot performs corresponding actions or adjusts its own state based on the content of the user feedback data. The specific response method depends on the content of the user feedback data and the preset interaction logic.
[0065] For example, if the user feedback data is the voice command "I am here", the follow robot can try to lock the user's location by sound source localization or restarting visual recognition; if the user feedback data is that the user actively approaches the follow robot and touches its head, the follow robot can recognize it as a reassurance signal and reduce the interaction intensity or resume normal following; if the user feedback data is a stop following command sent through the terminal application, the follow robot should immediately exit the follow mode and enter standby mode.
[0066] In practice, follow-me robots can be equipped with various input devices such as microphone arrays, cameras, and tactile sensors to receive different types of user feedback data. For voice feedback, the follow-me robot can use Automatic Speech Recognition (ASR) technology to convert audio into text, and then use Natural Language Understanding (NLU) technology to extract the user's intent. For visual feedback, the follow-me robot can use computer vision technology to recognize the user's gestures, posture, or facial orientation. For tactile feedback, the follow-me robot can determine the user's intent based on the pressure value and position of the touch sensor. Through these diverse input channels, the follow-me robot can more comprehensively understand the user's responses, thereby making more accurate responses.
[0067] In this embodiment of the invention, the original one-way proactive prompting from the following robot is upgraded to a two-way human-computer dialogue, enabling users to directly respond to the robot's requests for help, thereby accelerating the resolution of abnormal situations. For example, a user's simple statement, "I'm here," allows the following robot to quickly relocate, which is far more efficient than the robot blindly searching on its own. Secondly, this closed-loop interaction enhances the user's sense of participation and control. The user is no longer a passive observer receiving information but a collaborator actively participating in problem-solving, significantly strengthening the tacit understanding and trust between humans and machines. Finally, by receiving and responding to user feedback data, the following robot can more flexibly adapt to the personalized needs and immediate intentions of different users, making the entire following interaction process more intelligent, natural, and efficient.
[0068] Based on the above embodiments, the preset abnormal state also includes an obstacle encounter state, which includes a first obstacle state, a second obstacle state, and a third obstacle state. The obstacle sizes in the first obstacle state, the second obstacle state, and the third obstacle state increase from small to large; The first obstacle state, the second obstacle state, and the third obstacle state each correspond to different first interaction data.
[0069] Specifically, the obstacle encounter state can be further subdivided into three stages: the first obstacle state, the second obstacle state, and the third obstacle state. The obstacle size increases from small to large in the first, second, and third obstacle states.
[0070] In the first obstacle state, it indicates that there are small obstacles such as toys or slippers ahead. The robot can navigate around them by making a slight turn. At this time, the corresponding first interaction data can include self-recovery prompts, such as a short beep or a low-brightness flashing light, and attempt to navigate around them autonomously.
[0071] The second obstacle state indicates that there is a large obstacle such as a chair or sofa in front, which the robot cannot bypass on its own. In this case, the corresponding first interaction data can include an obstacle removal request prompt, such as a voice request for help like "There's something blocking my way, can you help me move it?", while the head or body moves towards the obstacle to guide the user's attention.
[0072] The third obstacle state indicates that the robot is stuck, such as its wheels getting stuck in carpet gaps or tangled in electrical wires. In this case, the corresponding first interaction data can include safety warnings and rescue requests, such as an emergency voice message: "I'm stuck! Help me!", which can be accompanied by rapidly flashing red lights, and all drive motors can be stopped to prevent damage.
[0073] In this embodiment of the invention, by subdividing the state of encountering obstacles, the monotony and ineffectiveness of a single prompt can be avoided, enabling the following robot to exhibit different reactions when facing different obstacle situations. This anthropomorphic interaction design is more in line with human social habits and can effectively guide users to take appropriate actions.
[0074] In summary, such as Figure 2 The diagram shows the complete flow of a human-computer interaction method provided in an embodiment of the present invention. Figure 2 The system simultaneously considers three preset abnormal states: encountering obstacles, the user distance being less than the preset distance, and losing the user. Furthermore, it processes these three preset abnormal states sequentially.
[0075] After starting to follow, it enters follow mode and monitors the follow status in real time; Determine if the following state belongs to a preset abnormal state: First, determine whether the following state is in the state of encountering an obstacle; If the following state is an obstacle encounter state, the first interaction data corresponding to the obstacle encounter state is determined and executed, and a countdown of the third preset duration is started. If the following state is an obstacle encounter state within the third preset duration, the third interaction data is executed. If the following state changes within the third preset duration, the countdown is canceled at the moment of state change, and the following state after the change is re-evaluated to see if it belongs to a preset abnormal state. If the following state is not in the obstacle encounter state, then continue to determine whether the following state is in the user distance less than the preset distance state. If the following state is in the user distance less than the preset distance state, then determine and execute the first interaction data corresponding to the user distance less than the preset distance state, and start the countdown for the third preset duration. If the following state is in the user distance less than the preset distance state for the entire third preset duration, then execute the third interaction data. If the following state changes within the third preset duration, then cancel the countdown at the moment of state change, and re-determine whether the changed following state belongs to the preset abnormal state. If the following state is not in the state where the user distance is less than the preset distance, then continue to determine whether the following state is in the state of losing the user. If the following state is in the state of losing the user, then determine and execute the first interaction data corresponding to the state of losing the user, and start a countdown of the third preset duration. If the following state is in the state of losing the user within the third preset duration, then execute the third interaction data. If the following state changes within the third preset duration, then cancel the countdown at the moment of state change, and re-determine whether the changed following state belongs to the preset abnormal state. If the following status does not belong to the user loss status, then execute the following task and continue to monitor the following status in real time; If the user is lost multiple times during the following state, the second interaction data is executed and a countdown for the third preset duration is started. If the user is lost during the third preset duration, the third interaction data is executed. If the following state changes during the third preset duration, the countdown is canceled at the moment of the state change, and the changed following state is re-evaluated to determine whether it belongs to the preset abnormal state.
[0076] like Figure 3 As shown, based on the above embodiments, this embodiment of the invention provides a human-computer interaction device, including: Detection module 31 is used to detect the following status in real time in follow mode; The determining module 32 is used to determine the first interaction data corresponding to the preset abnormal state when the following state belongs to the preset abnormal state; the preset abnormal state includes at least one of the following states: user distance is less than preset distance and user loss state. The interaction module 33 is used to execute the first interaction data; the first interaction data includes an abnormal status prompt.
[0077] Based on the above embodiments, the human-computer interaction device provided in the embodiments of the present invention includes a first loss state and a second loss state in the user loss state. The duration of the first tracking loss state is less than or equal to the first preset duration; The duration of the second tracking error is greater than the first preset duration and less than or equal to the second preset duration.
[0078] Based on the above embodiments, the human-computer interaction device provided in this embodiment of the invention further includes an interaction module for: After real-time detection of the following status, if the following status shows multiple instances of losing the user, then the second interaction data is executed; the second interaction data includes prompts for multiple user disappearances.
[0079] Based on the above embodiments, the human-computer interaction device provided in this embodiment of the invention further includes an interaction module for: When executing the first interactive data and when executing the second interactive data, a countdown of a third preset duration is started. If all the following states within the third preset time period belong to the same preset abnormal state, then the third interactive data is executed; the third interactive data includes an exit prompt.
[0080] Based on the above embodiments, the human-computer interaction device provided in this embodiment of the invention further includes an interaction module for: After the countdown for the third preset duration is started, if the following state changes within the third preset duration, the countdown is canceled at the moment of the state change, and it is re-determined whether the changed following state belongs to the preset abnormal state.
[0081] Based on the above embodiments, the human-computer interaction device provided in this embodiment of the invention further includes an interaction module for: After executing the first interactive data, the second interactive data, and the third interactive data, user feedback data is received. Respond to the user feedback data.
[0082] Based on the above embodiments, the human-computer interaction device provided in the embodiments of the present invention further includes an obstacle encounter state in the preset abnormal state, which includes a first obstacle state, a second obstacle state, and a third obstacle state. The obstacle sizes in the first obstacle state, the second obstacle state, and the third obstacle state increase from small to large.
[0083] Specifically, the functions of each module in the human-computer interaction device provided in the embodiments of the present invention correspond one-to-one with the operation flow of each step in the above-mentioned method embodiments, and the achieved effects are also the same. For details, please refer to the above embodiments, and this will not be repeated in the embodiments of the present invention.
[0084] Based on the above embodiments, this invention also provides a follower robot, including: a robot body; The robot body is used to execute the human-computer interaction methods provided in the above embodiments.
[0085] Specifically, the robot body can be equipped with various sensors, such as vision sensors, positioning systems, LiDAR or depth cameras, and inertial measurement units. The robot body also includes hardware actuators, such as speech synthesis modules, displays, and lighting modules.
[0086] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute the human-computer interaction methods provided in the above embodiments.
[0087] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to related technologies, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0088] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the human-computer interaction methods provided in the above embodiments.
[0089] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the human-computer interaction methods provided in the above embodiments. This computer-readable storage medium can be either a non-transitory computer-readable storage medium or a transient computer-readable storage medium, and is not specifically limited herein.
[0090] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A human-machine interaction method, characterized in that, include: In follow mode, the follow status is detected in real time; When the following state belongs to a preset abnormal state, the first interaction data corresponding to the preset abnormal state is determined; The preset abnormal states include at least one of the following: the user distance is less than the preset distance and the user has been lost. Execute the first interactive data; the first interactive data includes an abnormal status prompt.
2. The human-machine interaction method of claim 1, wherein, The status of the lost user includes a first lost status and a second lost status; The duration of the first tracking loss state is less than or equal to the first preset duration; The duration of the second tracking error is greater than the first preset duration and less than or equal to the second preset duration; The first tracking loss state and the second tracking loss state correspond to different first interaction data.
3. The human-computer interaction method according to claim 1, characterized in that, The real-time detection and following state also includes: If the following state contains multiple instances of the user being lost, then the second interaction data is executed; the second interaction data includes prompts indicating that the user has disappeared multiple times.
4. The human-computer interaction method according to claim 3, characterized in that, The execution of the first interactive data and the execution of the second interactive data further include: Start the countdown for the third preset duration; If all the following states within the third preset time period belong to the same preset abnormal state, then the third interactive data is executed; the third interactive data includes an exit prompt.
5. The human-computer interaction method according to claim 4, characterized in that, The countdown for the third preset duration is then initiated, followed by: If the following state changes within the third preset time period, the countdown is canceled at the moment of the state change, and the changed following state is re-evaluated to determine whether it belongs to the preset abnormal state.
6. The human-computer interaction method according to claim 4, characterized in that, The execution of the first interactive data, the execution of the second interactive data, and the execution of the third interactive data are all followed by: Receive user feedback data; Respond to the user feedback data.
7. The human-computer interaction method according to any one of claims 1-6, characterized in that, The preset abnormal state also includes an obstacle encounter state, which includes a first obstacle state, a second obstacle state, and a third obstacle state; The obstacle sizes in the first obstacle state, the second obstacle state, and the third obstacle state increase from small to large; The first obstacle state, the second obstacle state, and the third obstacle state each correspond to different first interaction data.
8. A human-computer interaction device, characterized in that, include: The detection module is used to detect the following status in real time during follow mode; The determination module is used to determine the first interaction data corresponding to the preset abnormal state when the following state belongs to the preset abnormal state; the preset abnormal state includes at least one of the following states: user distance is less than preset distance and user loss state. An interaction module is used to execute the first interaction data; the first interaction data includes an abnormal status prompt.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the human-computer interaction method as described in any one of claims 1-7.
10. A following robot, characterized in that, include: The robot itself; The robot body is used to perform the human-computer interaction method as described in any one of claims 1-7.
