Vehicle type robot behavior control method and device, equipment and medium

Abstract

The application relates to the technical field of vehicle robot control, solves the problem that a vehicle robot cannot provide sufficient emotional value for a user through action and expression interaction in the prior art, and provides a behavior control method of a vehicle robot, characterized in that the method is applied to a vehicle robot and comprises the following steps: obtaining interaction information, wherein the interaction information at least includes one of a weight change amount and voice information sent by a user; analyzing the interaction information to obtain an interaction behavior category; performing mapping processing on the interaction behavior category according to a preset mapping rule to obtain a target action to be executed and a target expression to be displayed; determining corresponding control instructions according to the target action and the target expression; and driving the vehicle robot to execute the target action and display the target expression according to the control instructions, so that behavior control is realized.

Description

Technical Field

[0001] This application relates to the field of vehicle robot control technology, and in particular to a behavior control method, device, equipment and medium for a vehicle robot. Background Technology

[0002] With the development of intelligent control technology and service robots, mobile robots with voice interaction capabilities are gradually being applied in scenarios such as child companionship, entertainment, and smart transportation. Existing vehicle-mounted robots can typically acquire user voice commands through microphones and perform corresponding walking, turning, or simple actions based on the voice content, while displaying fixed facial expressions on the screen to enhance the interactive experience.

[0003] In terms of interaction, existing technologies often treat voice input as a single command source, making it difficult to further distinguish the different interactive intentions corresponding to voice information, such as action commands, question-based voices, and emotion-related chat content. This results in a lack of targeted selection of actions and expressions. Existing patent CN120015062A (A Method for Expressing Emotional Behavior in Humanoid Robots, 2025.05.16) focuses on performing emotion recognition on user voice, outputting emotion categories such as happiness, sadness, anger, and surprise, and then generating and fusing emotional action codes and emotional voice codes accordingly. However, it does not address technical solutions for triggering interactive behaviors based on emotions. Furthermore, the facial expressions displayed on the screens of existing vehicle-mounted robots are often fixed, lacking detailed processing for different load states and scenarios, leading to monotonous and uninteresting interactive performance. Additionally, existing vehicle-mounted robots may still perform actions with the original amplitude or rhythm when carrying people or goods, resulting in insufficient comfort and safety. For different types of voice input, such as action commands, questions, or emotional chat, existing solutions usually only process them uniformly, making it difficult to select matching actions and expressions based on the differences in voice semantics. This results in monotonous interaction, low anthropomorphism, and an inability to provide emotional value to users.

[0004] Therefore, how to classify interactive behaviors based on a comprehensive consideration of voice interaction behavior and load changes, and determine the corresponding combination of actions and expressions according to preset rules to provide emotional value to users, still needs further improvement. Summary of the Invention

[0005] In view of this, embodiments of this application provide a behavior control method, device, equipment, and medium for vehicle-type robots, in order to solve the problem that in the prior art, vehicle-type robots cannot provide sufficient emotional value to users through action execution and facial expression interaction.

[0006] In a first aspect, embodiments of this application provide a behavior control method for a vehicle-type robot, the method comprising: Acquire interactive information, wherein the interactive information includes at least one of the change in load and voice information emitted by the user; The interaction information is parsed to obtain the interaction behavior category; According to the preset mapping rules, the interactive behavior categories are mapped to obtain the target action to be executed and the target expression to be displayed. Based on the target action and target expression, determine the corresponding control command; According to the control instructions, the vehicle robot is driven to perform target actions and display target expressions, thereby achieving behavior control.

[0007] Preferably, parsing the interaction information to obtain the interaction behavior category includes: When the interaction information includes the change in load and the change in load is greater than a preset first weight threshold, the interaction behavior category is determined to be a human-carrying and cargo-carrying interaction behavior. When the interaction information includes voice information emitted by the user, the voice information is semantically parsed to obtain the voice interaction behavior category; Based on the categories of human-carrying and cargo-carrying interaction behaviors and voice interaction behaviors, the interaction behavior categories are obtained.

[0008] Preferably, when the interaction behavior category is determined to be a person-carrying or object-carrying interaction behavior, the step of mapping the interaction behavior category according to a preset mapping rule to obtain the target action to be executed and the target expression to be displayed includes: Based on the load change, the preset second weight threshold and the preset third weight threshold, the load state is classified to obtain the load state category, wherein the preset second weight threshold is less than the preset third weight threshold. Based on the load state category and preset load rules, candidate actions and candidate expressions are obtained; Based on preset safety constraints, the candidate actions are filtered to obtain the target action to be executed; According to the preset priority rules, select the target emoticon to be displayed from the candidate emoticons.

[0009] Preferably, the step of classifying the load state according to the load change, a preset second weight threshold, and a preset third weight threshold to obtain load state categories includes: When the change in load is less than the preset second weight threshold, the load state category is determined to be a light load state. When the change in load is greater than or equal to the preset second weight threshold and less than the preset third weight threshold, the load state category is determined to be medium load state. When the change in load exceeds the preset third weight threshold, the load state is determined to be a high load state.

[0010] Preferably, when the interaction information includes voice information emitted by the user, semantic parsing is performed on the voice information to obtain a voice interaction behavior category, including: The voice information sent by the user is preprocessed to obtain the target audio data; Speech endpoint detection is performed on the target audio data, and valid speech segments are determined based on the detection results; The effective speech segment is input into a pre-trained audio recognition model to obtain the semantic recognition result; Based on the semantic recognition results, the speech information is classified according to intent to obtain the speech interaction behavior category.

[0011] Preferably, if the interactive behavior category includes a voice interactive behavior category, the step of mapping the interactive behavior category according to a preset mapping rule to obtain the target action to be executed and the target expression to be displayed includes: Obtain the priority of preset voice interaction behavior categories, wherein the voice interaction behavior categories include action commands, question and answer, and emotion-related chat; When the voice interaction behavior category includes action commands and / or question and answer and / or emotion-related chat, the voice interaction behavior category is mapped according to the priority, the preset candidate expression library and the preset candidate action library to obtain the target action to be executed and the target expression to be displayed.

[0012] Preferably, determining the corresponding control command based on the target action and target expression includes: Based on the target action and the preset action parameter template, a set of action control parameters is obtained; Based on the target expression and the preset expression parameter template, an expression control parameter set is obtained; According to the preset instruction format, the action control parameter set and the expression control parameter set are encapsulated and processed to generate the control instruction.

[0013] Secondly, embodiments of this application provide a behavior control device for a vehicle-type robot, the device comprising: An information acquisition module is used to acquire interactive information, wherein the interactive information includes at least one of the weight change amount and voice information issued by the user; The information parsing module is used to parse the interaction information to obtain the interaction behavior category; The behavior mapping module is used to map the interactive behavior categories according to preset mapping rules to obtain the target action to be executed and the target expression to be displayed. The control command module is used to determine the corresponding control command based on the target action and the target expression; The execution module is used to drive the vehicle robot to perform target actions and display target expressions according to the control instructions, thereby achieving behavior control.

[0014] Thirdly, embodiments of this application provide an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method of the first aspect described above.

[0015] Fourthly, embodiments of this application provide a storage medium storing computer program instructions that, when executed by a processor, implement the method of the first aspect described above.

[0016] In summary, the beneficial effects of this application are as follows: The present application provides a method, apparatus, device, and medium for controlling the behavior of a vehicle-mounted robot. The method includes: acquiring interactive information, wherein the interactive information includes at least one of a change in load and voice information emitted by a user; parsing the interactive information to obtain an interactive behavior category; mapping the interactive behavior category according to a preset mapping rule to obtain a target action to be executed and a target expression to be displayed; determining a corresponding control command based on the target action and target expression; and driving the vehicle-mounted robot to execute the target action and display the target expression according to the control command, thereby achieving behavior control. This invention acquires interactive information such as a change in load and user voice information, parses the interactive information to obtain interactive behavior categories, maps different interactive behavior categories to corresponding target actions and target expressions according to preset mapping rules, and then generates a unified control command to drive the vehicle-mounted robot to execute the target action and display the target expression, thereby achieving coordinated control of action and expression. The method provided by this invention can balance safety constraints and interaction consistency in different scenarios such as passenger and cargo transportation and voice interaction. It enables vehicle robots to adaptively adjust their movements and facial expressions according to different interaction situations, thereby improving the degree of anthropomorphism and interactive fun, and enhancing the overall interactive experience of users in scenarios such as companionship and entertainment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of this application.

[0018] Figure 1This is a schematic diagram of the overall flow of the behavior control method for the vehicle robot according to an embodiment of this application; Figure 2 This is a schematic diagram of the behavior control device for the vehicle robot according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0019] The features and exemplary embodiments of various aspects of this application will now be described in detail. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only configured to explain this application and are not configured to limit this application. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples of this application.

[0020] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0021] Example 1 Please see Figure 1 This application provides a behavior control method for a vehicle-type robot, which is applied to a vehicle-type robot. The vehicle-type robot includes a chassis drive mechanism, a lifting and swinging actuator, a load detection module, a voice acquisition module, and an expression display screen. The method includes: S1: Obtain interactive information, wherein the interactive information includes at least one of the weight change amount and voice information issued by the user; Specifically, interactive information refers to the input information used by the vehicle-mounted robot to determine user intentions or load status. Examples include increased load due to a child sitting in the vehicle, or voice commands from the user to dance. The load change refers to the difference between the current load and the robot's unloaded load, which can be obtained through seat pressure sensors, weighing sensors, or motor current feedback. Voice information can be represented by audio data collected by a microphone. Integrating load change and voice information into the interactive information system allows for subsequent parsing and mapping that covers both passenger / cargo transport and voice interaction scenarios, improving the comprehensiveness and real-time nature of interaction triggers.

[0022] S2: Parse the interaction information to obtain the interaction behavior category; Specifically, interactive behavior categories are abstract labels for the current interactive context, used to drive subsequent actions and expression selections. Examples include human-carrying / object-carrying interactions, action commands under voice interaction, and question-based or emotion-based chat. We can first distinguish between human-carrying / object-carrying interactions and voice interactions, then further refine this by applying threshold triggers to weight changes, performing semantic analysis and intent classification on voice information, and finally summarizing the analysis results into a category code that can be used for indexing rules. Subsequent actions and expressions are selected based on the category, resulting in stronger interaction consistency. Transforming raw sensor data and voice data into decision-making classification results reduces the complexity of subsequent control logic.

[0023] S3: According to the preset mapping rules, the interactive behavior category is mapped to obtain the target action to be executed and the target expression to be displayed; Specifically, the preset mapping rules are a set of correspondences between interactive behavior categories and actions and expressions. These can be expressed using lookup tables, rule trees, or state machine conditional branches. For example, a voice command for an action might be mapped to a swaying motion accompanied by a smiling expression, while carrying people or goods under high load might be mapped to a slight swaying motion accompanied by a panicked or prompting expression. Interactive behavior categories are used as index conditions to retrieve candidate actions and expressions. Then, load status, safety constraints, or priority strategies are combined to select the target action and expression. This transforms the intent into an executable output target, resulting in more stable matching of actions and expressions.

[0024] S4: Determine the corresponding control command based on the target action and target expression; Specifically, control commands are sets of data or signals that the robot's actuators and display modules can directly execute. Target actions can be further parameterized as action type, amplitude, frequency, and duration; target expressions can be parameterized as expression number, frame sequence, and display duration. First, the target action is converted into target motion parameters for the motors or servos. Then, the target expression is converted into display refresh commands. Finally, these are encapsulated into unified control commands in a preset format, along with verification and timestamps for synchronized execution. This process converts decision outputs into low-level executable commands, reducing asynchrony issues caused by multiple modules driving their own actions.

[0025] S5: According to the control instructions, drive the vehicle robot to perform target actions and display target expressions to achieve behavior control.

[0026] Specifically, behavior control refers to the robot's integrated expression on its chassis, lifting and swaying mechanism, and display screen that is consistent with the interactive context. For example, it might perform a left-right swaying motion to simulate dancing while simultaneously displaying a smiling expression, or perform small movements and display suggestive facial expressions when carrying a heavy load of passengers. The controller parses the action field in the control command and outputs PWM, CAN, or serial port control signals to drive the motor and actuator to complete the target action, while simultaneously displaying the target expression on the facial expression display screen, providing human-like feedback through the physical system and the human-machine interface. The synchronized execution of actions and expressions brings a stronger sense of immersion in the interaction, and during passenger and cargo transportation, it constrains the target actions, improving comfort and safety.

[0027] In one embodiment, S2 includes: S21: When the interaction information includes the change in load and the change in load is greater than a preset first weight threshold, the interaction behavior category is determined to be a human-carrying and cargo-carrying interaction behavior. Specifically, a preset first weight threshold is used as a criterion to distinguish between ordinary disturbances and human / object-carrying interactive behaviors. For example, if a child getting into the car causes a significant increase in the load, it can be judged as a human / object-carrying interactive behavior. By comparing the load change with the first weight threshold, the carrying status is identified, preventing the robot from outputting excessive movements in the ordinary interaction mode when carrying people, thus reducing risks. At the same time, during human / object-carrying interactions, facial expressions are displayed based on the specific load change, enhancing the fun.

[0028] S22: When the interaction information includes voice information issued by the user, perform semantic analysis on the voice information to obtain the voice interaction behavior category; Specifically, voice interaction behavior categories are the subdivisions of the semantic intent of speech, such as action commands, questions, and emotion-related chat. After preprocessing the audio, including noise reduction, endpoint detection, and feature extraction, the audio is input into a recognition model to obtain semantic recognition results. Subsequently, intent classification is performed, outputting multiple possible categories, thus elevating the speech from a sound signal to an interactive intent, allowing for targeted selection of actions and expressions. Through category subdivision, the robot can choose different combinations of expressions and actions for dance commands, questioning voices, or emotional chat, resulting in a more human-like performance.

[0029] S23: Based on the categories of manned and cargo-carrying interaction behaviors and voice interaction behaviors, obtain the interaction behavior categories.

[0030] Specifically, the interaction behavior categories are obtained by combining the categories of human-carrying and cargo-carrying interaction behaviors with voice interaction behaviors.

[0031] In one embodiment, when the interaction behavior is determined to be a human-carrying or object-carrying interaction behavior, step S3 includes: S31: Based on the load change, the preset second weight threshold and the preset third weight threshold, the load state is classified to obtain the load state category, wherein the preset second weight threshold is less than the preset third weight threshold. Specifically, load status categories reflect the level of load-bearing capacity. Preset second and third weight thresholds serve as criteria for different handling scenarios. Subsequent adjustments to the action amplitude and refined expression selection are made according to the load level. The change in load is compared with the two thresholds to determine the range it falls into and outputs light, medium, and high load status categories. The interaction between people and objects is further refined, displaying different expressions under different load states to enhance interactive expressiveness.

[0032] S32: Based on the load state category and preset load rules, candidate actions and candidate expressions are obtained; Specifically, the preset load rules are the filtering criteria from load state to candidate set. For example, light load allows for more swaying and can display relaxed expressions such as smiling or humming; medium load is limited to small forward movements, swaying, or slow ascent and descent, and displays expressions such as sighing, crying, or dizziness; high load displays warning or angry expressions. Based on the load state category and preset load rules, candidate action sets and candidate expression sets that meet the load rules are extracted from the action library and expression library to narrow down the selection range and provide a basis for subsequent safety screening and priority selection.

[0033] S33: Based on preset safety constraints, the candidate actions are filtered to obtain the target action to be executed; Specifically, safety constraints are used to limit the risks of movement while carrying passengers or cargo, such as limiting the maximum swaying amplitude, limiting the speed of ascent and descent, limiting instantaneous acceleration, or prohibiting certain combinations of movements. The motion parameters of candidate movements are verified, and movements that do not meet the constraints are eliminated. If multiple constraints are met, the target movement is selected based on comfort or energy consumption strategies. The aim is to ensure both safety and passenger comfort while maintaining interactivity, avoiding excessive swaying and discomfort caused by large movements.

[0034] S34: Select the target expression to be displayed from the candidate expressions according to the preset priority rules.

[0035] Specifically, priority rules are used to determine the most appropriate expression from multiple candidate expressions. For example, under high load, prompting expressions take precedence over angry expressions, or under medium load, panicked or crying expressions are used to indicate caution. Priority weights are assigned to each candidate expression, and the target expression is selected sequentially according to priority, based on the current load state category or continuous interaction state. This improves the scene adaptability of expression expression and avoids inconsistencies between expressions and the load state. Through priority selection, expression output is more stable, users can more easily understand the robot's state and intentions, and the interactive fun is more controllable.

[0036] In one embodiment, S31 includes: S311: When the change in load is less than the preset second weight threshold, the load state category is determined to be a light load state. Specifically, the change in load is compared with a second weight threshold. If the change is less than the threshold, a light load state category is output and the corresponding load rule branch is entered. This allows for greater motion freedom in light load scenarios, avoiding overly conservative restrictions on interaction, while still allowing for a wider range of swinging or probing movements, resulting in more flexible interaction.

[0037] S312: When the change in load is greater than or equal to the preset second weight threshold and less than the preset third weight threshold, the load state category is determined to be medium load state. Specifically, a medium load state corresponds to a child sitting in a semi-sitting or lightly seated position, or placing a relatively heavy object on the child but not reaching a high load level. In implementation, the condition is determined by simultaneously meeting a range greater than the second threshold and less than the third threshold, outputting the medium load state category and triggering the medium load rule. Through medium load grading, comfort and safety are improved while retaining some interactive effects.

[0038] S313: When the change in load is greater than the preset third weight threshold, the load state category is determined to be a high load state.

[0039] Specifically, high-load conditions typically correspond to stable passenger or heavy cargo carrying, requiring motion control that prioritizes safety and smoothness. In implementation, a high-load condition category is output once the comparison result exceeds a third threshold. Subsequent candidate action generation and safety screening then impose stricter parameter ranges to prevent actions such as large-amplitude swaying or rapid ascent / descent. Simultaneously, warnings and angry expressions are displayed. High-load determination makes the control strategy more robust, reduces swaying impact, and improves passenger safety and experience.

[0040] In one embodiment, S22 includes: S221: Preprocess the voice information sent by the user to obtain the target audio data; Specifically, preprocessing improves the robustness of speech recognition. The target audio data is a normalized audio frame sequence suitable for subsequent endpoint detection and model input. Preprocessing can be performed through automatic gain control, bandpass filtering, echo cancellation, silence compression, and resampling to suppress the impact of environmental noise and device differences on recognition, making subsequent semantic parsing more stable. After preprocessing improves the signal-to-noise ratio, action commands and question speech are less likely to be misinterpreted, thus improving the accuracy of interaction.

[0041] S222: Perform speech endpoint detection on the target audio data, and determine the valid speech segments based on the detection results; Specifically, speech endpoint detection is used to locate the start and end of speech, avoiding inputting silent or noisy segments into the recognition model. A valid speech segment is a continuous audio interval containing the user's valid vocal content. In implementation, this can be achieved by using an energy threshold plus a zero-crossing rate, or by using a VAD-based probabilistic method to output the start and end points, and then cropping the valid speech segment from the target audio data. This reduces invalid computation and misidentification, allowing semantic recognition to focus on the actual speech content. Accurate endpoint detection significantly reduces the probability of mistaking ambient sound for commands, improving interaction reliability.

[0042] S223: Input the effective speech segment into the pre-trained audio recognition model to obtain the semantic recognition result; Specifically, the pre-trained audio recognition model can be an end-to-end speech recognition or speech semantic model. The semantic recognition result can be a text sequence, a keyword set, or a semantic vector representation, such as recognizing semantic information like "dancing," "what's wrong," and "don't be afraid." Feature extraction is performed on effective speech segments, such as Mel-spectrum or MFCC, and then fed into the model for inference to obtain the output. Semantic information that can be used for intent classification is extracted from the speech signal, providing a basis for the selection of actions and expressions in subsequent steps.

[0043] S224: Based on the semantic recognition results, classify the speech information according to intent to obtain the speech interaction behavior category.

[0044] Specifically, intent classification is used to convert semantic recognition results into category labels such as action commands, questions, and emotional chat. Voice interaction behavior categories are a set of categories that can be used for mapping rules. In implementation, keyword rules with confidence thresholds can be used, or a lightweight classifier can be used to classify semantic vectors and output primary and secondary categories to cover ambiguous statements. Transforming complex semantics into specific voice interaction behavior categories facilitates stable execution of rule mapping, resulting in different action and expression schemes under different semantic types, thus enhancing anthropomorphism and engagement.

[0045] In one embodiment, if the interaction behavior category includes a voice interaction behavior category, step S3 includes: S35: Obtain the priority of preset voice interaction behavior categories, wherein the voice interaction behavior categories include action commands, question and answer, and emotion-related chat; Specifically, the voice interaction behaviors include action commands, question and answer, and emotion-related chat. When these three voice interaction behaviors overlap, they are executed in order of priority. For example, the preset priority is to execute action commands first, then answer questions and answer, and finally start emotion-related chat.

[0046] S36: When the voice interaction behavior category includes action commands and / or question and answer and / or emotion-related chat, the voice interaction behavior category is mapped according to the priority, the preset candidate expression library and the preset candidate action library to obtain the target action to be executed and the target expression to be displayed.

[0047] Specifically, an action command is a voice category in which the user explicitly requests the robot to perform a certain action. Candidate actions are a predefined set of actions, such as swaying left and right, a tilting / lifting motion, or a combination of dance movements. When the voice interaction behavior category includes an action command, the corresponding action is selected as the target action from the preset candidate actions. Then, according to preset action-expression association rules, a target expression matching the target action is determined from the preset candidate expressions. When the voice interaction behavior category includes a question-and-answer session, a target expression matching "thinking" is determined from the preset candidate expressions. Then, according to preset thinking expression-action association rules, a target action matching "thinking" is determined. When the voice interaction behavior category is an emotion-related chat, a target expression matching the emotion is determined from the preset candidate expressions. Then, according to preset emotion-action association rules, a target action matching the emotion is determined. The action-expression association rules are the correspondence between actions and expressions; for example, dancing actions match a smiling eyebrow, dizziness actions match a dizzy expression, and low-amplitude tilting motions match a flustered or prompting expression. Using the target action's action type as an index, a matching expression is selected from a set of candidate expressions to ensure semantic consistency between the action and the expression. This avoids a disjointed experience caused by an action resembling dancing while the expression is crying. If multiple matches exist, selection is based on priority or a random strategy, making the interaction more flexible and engaging. Action-expression association rules make the expression more natural, allowing users to better understand the robot's intentions. When the voice interaction behavior category includes question-and-answer, a target expression matching the thinking state is determined from a preset set of candidate expressions. Question-type voices are those where the user asks a question or inquires; thinking expressions can include panic, dizziness, or other expressions used to express a state of thought. After outputting the question category identifier, a thinking expression is directly selected from the candidate expressions, or different intensities of thinking expression, such as contemplation, doubt, or panic, are selected based on the question's difficulty and confidence level. This provides perceptible feedback to the question-and-answer interaction, enhancing the continuity of the interaction.

[0048] For example, if a user asks, "Move forward 10 centimeters and tell me what the weather is like today," including action commands and questions, the robot will first prioritize moving forward 10 centimeters and matching the relevant facial expression. Then, it will determine the target facial expression that matches the user's thought and execute the target action that matches the thought, such as a slight sway. After thinking, it will then provide an answer. If a user asks, "Move forward 10 centimeters and tell me what the weather is like today; I'm in a great mood," including action commands, questions, and conversations related to emotions, the robot will first prioritize moving forward 10 centimeters and matching the relevant facial expression. Then, it will determine the target facial expression that matches the user's thought and execute the target action that matches the thought, such as a slight sway. After thinking, it will provide an answer. Finally, based on the user's great mood, it will display a happy facial expression and match the corresponding action, such as a large sway. The introduction of different thinking expressions enhances the anthropomorphism of the robot and makes the user feel that the robot is responsive.

[0049] S38: Based on the preset rules for associating thinking expressions with actions, determine the target action that matches the thinking; Specifically, the rules for associating thinking expressions with actions are used to select more appropriate actions based on the thinking state, such as slight swaying from side to side, minor rises and falls, or slow movements after a short pause. Using the selected thinking expression as an index, the target action is chosen from the set of thinking actions, allowing question-based interactions to not only have facial expressions but also coordinated body movements, forming a more complete behavioral expression. Through rule association, actions and expressions are presented synchronously, enhancing the consistency and fun of the interaction.

[0050] S39: When the voice interaction behavior category is chat related to emotions, determine the target expression that matches the emotion from the preset candidate expressions; Specifically, emotion-related chats involve voice messages expressing emotions such as comfort, complaint, and happiness. Emotion-matching emoticons can be selected from expressions like smiling, crying, anger, and panic. Based on the emotion tags or tendencies output according to the intended category, the corresponding emoticon is selected from the candidate set. Furthermore, recent interaction history is used to smooth out changes and avoid frequent emoticon jumps. The emotional elements in the voice are translated into intuitive screen feedback, making the chat interaction more immersive. Emotion-matching emoticons significantly enhance the interactive experience in companionship scenarios, improve user experience, and provide emotional value to users.

[0051] S310: Based on the preset emotion-action association rules, determine the target action that matches the emotion.

[0052] Specifically, the emotion-action association rules are used to extend emotional expressions into corresponding action styles. For example, happiness is paired with swaying left and right to enhance excitement, panic with small, rapid swaying, and crying with slow, probing movements. Using emotional expressions or emotion categories as indices, candidate action sets are selected, and target action parameters are output according to the parameter set. Furthermore, safety constraints are added to prevent overly aggressive actions. This allows emotional chat expression to go beyond screen emoticons, enhancing the perceptibility of emotional expression through actions. The consistent linkage between actions and emotions brings a more human-like and engaging interactive experience.

[0053] In one embodiment, S4 includes: S41: Obtain the action control parameter set based on the target action and the preset action parameter template; Specifically, a preset motion parameter template refers to a pre-established data set of motion types, amplitudes, frequencies, durations, and acceleration limits, such as the amplitude and frequency of left-right swaying, and the stroke and speed of rising and falling. Parameters corresponding to the target motion are selected from the preset motion parameter template, default parameters are read and modified based on load conditions or safety constraints, and the final parameter set is output. This transforms abstract motions into calculable and executable control inputs, ensuring that different motions are issued with uniform parameters. Parameterization makes motion control more stable and facilitates reducing amplitude or speed in manned scenarios, achieving a balance between safety and performance.

[0054] S42: Obtain the expression control parameter set based on the target expression and the preset expression parameter template; Specifically, a preset expression parameter template refers to a pre-established set of parameters used to drive the display of a target expression. Common parameters include expression number, frame sequence index, refresh rate, display duration, or transition method, such as the animation frame sequence and duration of a smiling expression. Based on the index of the target expression in the expression library, the corresponding resource path or frame data is read, and then a parameter set executable by the display module is generated. Furthermore, transition parameters for expression switching are added to avoid flickering and improve the stability of expression display.

[0055] S43: According to the preset instruction format, the action control parameter set and the expression control parameter set are encapsulated and processed to generate the control instruction.

[0056] Specifically, the preset command format is an encapsulation protocol for action and expression parameters, which can include a command header, length, action field, expression field, verification field, and timestamp, used for reliable transmission between the control module and the execution module. The action control parameter set is written to the action field, and the expression control parameter set is written to the expression field. They are concatenated in a fixed order, and a checksum is calculated to form a single control command for the execution mechanism and display module to parse. Unifying multi-source control information into a single transmittable and parseable command ensures that action execution and expression display are consistent with the expected results, improving system stability and scalability.

[0057] Example 2 Please see Figure 2 This application provides a behavior control device for a vehicle-type robot, the device comprising: An information acquisition module is used to acquire interactive information, wherein the interactive information includes at least one of the weight change amount and voice information issued by the user; The information parsing module is used to parse the interaction information to obtain the interaction behavior category; The behavior mapping module is used to map the interactive behavior categories according to preset mapping rules to obtain the target action to be executed and the target expression to be displayed. The control command module is used to determine the corresponding control command based on the target action and the target expression; The execution module is used to drive the vehicle robot to perform target actions and display target expressions according to the control instructions, thereby achieving behavior control. It should be noted that each module and unit in the vehicle robot behavior control device of this embodiment corresponds one-to-one with each step in the vehicle robot behavior control method of the aforementioned embodiments. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned vehicle robot behavior control method, and will not be repeated here.

[0058] Example 3 In addition, the behavior control method for the vehicle robot in this application embodiment can be implemented by an electronic device. Figure 3 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0059] Electronic devices may include processors and memory storing computer program instructions.

[0060] Specifically, the processor may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0061] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0062] Computer-readable media include both permanent and non-permanent, removable and non-removable media, which can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated communication signals and carrier waves.

[0063] The processor reads and executes computer program instructions stored in the memory to implement any of the vehicle robot behavior control methods in the above embodiments.

[0064] In one example, the electronic device may also include a communication interface and a bus. For example, Figure 3 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0065] The communication interface is mainly used to enable communication between modules, devices, units and / or equipment in the embodiments of this application.

[0066] A bus, including hardware, software, or both, couples components of an electronic device together. For example, and not limitingly, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, a bus may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0067] Example 4 Furthermore, in conjunction with the behavior control method for the vehicle-type robot in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the behavior control methods for the vehicle-type robot in the above embodiments.

[0068] In summary, the vehicle robot behavior control method, device, equipment, and medium provided in this application acquire interactive information such as load changes and user voice information, parse the interactive information to obtain interactive behavior categories, and map different interactive behavior categories to corresponding target actions and target expressions according to preset mapping rules. Then, a unified control command is generated to drive the vehicle robot to execute the target actions and display the target expressions, thereby achieving coordinated control of actions and expressions. The method provided by this invention can balance safety constraints and interaction consistency in different scenarios such as passenger and cargo transportation and voice interaction, enabling the vehicle robot to adaptively adjust its action forms and facial expressions according to different interactive situations, improving the degree of anthropomorphism and interactive fun, and enhancing the overall interactive experience for users in scenarios such as companionship and entertainment.

[0069] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0070] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0072] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0073] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0074] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0075] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A behavior control method for a vehicle-type robot, characterized in that, Applied to vehicle-mounted robots, the method includes: Acquire interactive information, wherein the interactive information includes at least one of the change in load and voice information emitted by the user; The interaction information is parsed to obtain the interaction behavior category; According to the preset mapping rules, the interactive behavior categories are mapped to obtain the target action to be executed and the target expression to be displayed. Based on the target action and target expression, determine the corresponding control command; According to the control instructions, the vehicle robot is driven to perform target actions and display target expressions, thereby achieving behavior control.

2. The method according to claim 1, characterized in that, The step of parsing the interaction information to obtain the interaction behavior category includes: When the interaction information includes the change in load and the change in load is greater than a preset first weight threshold, the interaction behavior category is determined to be a human-carrying and cargo-carrying interaction behavior. When the interaction information includes voice information emitted by the user, the voice information is semantically parsed to obtain the voice interaction behavior category; Based on the categories of human-carrying and cargo-carrying interaction behaviors and voice interaction behaviors, the interaction behavior categories are obtained.

3. The method according to claim 2, characterized in that, When the interaction behavior category is determined to be a human-carrying or object-carrying interaction behavior, the step of mapping the interaction behavior category according to a preset mapping rule to obtain the target action to be executed and the target expression to be displayed includes: Based on the load change, the preset second weight threshold and the preset third weight threshold, the load state is classified to obtain the load state category, wherein the preset second weight threshold is less than the preset third weight threshold. Based on the load state category and preset load rules, candidate actions and candidate expressions are obtained; Based on preset safety constraints, the candidate actions are filtered to obtain the target action to be executed; According to the preset priority rules, select the target emoticon to be displayed from the candidate emoticons.

4. The method according to claim 3, characterized in that, The load state is classified according to the load change, a preset second weight threshold, and a preset third weight threshold to obtain load state categories, including: When the change in load is less than the preset second weight threshold, the load state category is determined to be a light load state. When the change in load is greater than or equal to the preset second weight threshold and less than the preset third weight threshold, the load state category is determined to be medium load state. When the change in load exceeds the preset third weight threshold, the load state is determined to be a high load state.

5. The method according to claim 2, characterized in that, When the interaction information includes voice information emitted by the user, the voice information is semantically parsed to obtain the voice interaction behavior category, including: The voice information sent by the user is preprocessed to obtain the target audio data; Speech endpoint detection is performed on the target audio data, and valid speech segments are determined based on the detection results; The effective speech segment is input into a pre-trained audio recognition model to obtain the semantic recognition result; Based on the semantic recognition results, the speech information is classified according to intent to obtain the speech interaction behavior category.

6. The method according to claim 2, characterized in that, If the interactive behavior category includes a voice interactive behavior category, the step of mapping the interactive behavior category according to a preset mapping rule to obtain the target action to be executed and the target expression to be displayed includes: Obtain the priority of preset voice interaction behavior categories, wherein the voice interaction behavior categories include action commands, question and answer, and emotion-related chat; When the voice interaction behavior category includes action commands and / or question and answer and / or emotion-related chat, the voice interaction behavior category is mapped according to the priority, the preset candidate expression library and the preset candidate action library to obtain the target action to be executed and the target expression to be displayed.

7. The method according to claim 1, characterized in that, The step of determining the corresponding control command based on the target action and target expression includes: Based on the target action and the preset action parameter template, a set of action control parameters is obtained; Based on the target expression and the preset expression parameter template, an expression control parameter set is obtained; According to the preset instruction format, the action control parameter set and the expression control parameter set are encapsulated and processed to generate the control instruction.

8. A behavior control device for a vehicle-type robot, characterized in that, The device includes: An information acquisition module is used to acquire interactive information, wherein the interactive information includes at least one of the weight change amount and voice information issued by the user; The information parsing module is used to parse the interaction information to obtain the interaction behavior category; The behavior mapping module is used to map the interactive behavior categories according to preset mapping rules to obtain the target action to be executed and the target expression to be displayed. The control command module is used to determine the corresponding control command based on the target action and the target expression; The execution module is used to drive the vehicle robot to perform target actions and display target expressions according to the control instructions, thereby achieving behavior control.

9. A robot, characterized in that, include: At least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, The method as described in any one of claims 1-7 is implemented when the computer program instructions are executed by the processor.

Images