Energy consumption optimization and automatic charging control system for intelligent pet robot

Abstract

The invention relates to the technical field of robot control, in particular to an energy consumption optimization and automatic charging control system for an intelligent pet robot. The system comprises a robot data acquisition module, an energy consumption state evaluation module, an energy efficiency optimization module, an automatic charging scheduling module and a battery health management module which are in communication connection in sequence. The system is driven by multi-dimensional sensing data, energy consumption characteristics are extracted in combination with potential energy field mapping and sparse modeling, and an energy consumption influence map is constructed; joint optimization of the path and the servo strategy is realized by using a quantum behavior particle swarm algorithm; scheduling a charging path through a graph attention network; and the battery health state is dynamically predicted based on an incremental learning algorithm, and the operation efficiency, the cruising ability and the system intelligence of the robot are improved. According to the method, a closed-loop control process is constructed from energy consumption evaluation to energy efficiency optimization, automatic charging and battery health management, good system integration and intelligent response capability are achieved, and the cruising ability and operation stability of the pet robot in a complex application scene are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of robot control, and particularly to an energy consumption optimization and automatic charging control system for an intelligent pet robot. Background Art

[0002] With the continuous development of artificial intelligence, the Internet of Things, and robot control technology, intelligent pet robots have gradually entered family life and played roles such as companionship, interaction, and comfort. In order to achieve natural behavior responses and continuous user companionship services, intelligent pet robots usually have functions such as multi-modal perception, path planning, voice interaction, and emotion recognition. These functions are accompanied by high-frequency sensor data acquisition, complex decision-making calculations, and continuous servo motion control during operation, resulting in a relatively high overall energy consumption level of the device.

[0003] Currently, in the field of intelligent robots, certain progress has been made in the optimization of modules such as path planning, task scheduling, and behavior control. However, there is a lack of in-depth modeling and quantitative evaluation of the relationship between multi-dimensional operation data and energy consumption behavior. Some systems attempt to introduce energy consumption estimation methods based on static models, but such methods are difficult to adapt to the dynamic change characteristics of multi-scenarios and multi-behaviors. Moreover, in terms of charging management, a preset power threshold trigger mechanism is generally adopted, which cannot achieve intelligent scheduling and active control according to task priorities, battery health status, and environmental factors, easily leading to low energy utilization efficiency and abnormal attenuation of the battery cells. Summary of the Invention

[0004] To solve the above problems, the present invention provides an energy consumption optimization and automatic charging control system for an intelligent pet robot, which solves the problems of how to accurately model and evaluate the energy consumption behavior of an intelligent pet robot in a dynamic operation environment of multi-scenarios and multi-behaviors, and combines the task urgency and battery health status to achieve adaptive energy efficiency optimization and intelligent charging scheduling, thereby significantly improving the endurance and operation stability of the pet robot in complex application scenarios.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] An energy consumption optimization and automatic charging control system for an intelligent pet robot, comprising a robot data acquisition module, an energy consumption status evaluation module, an energy efficiency optimization module, an automatic charging scheduling module, and a battery health control module that are communicatively connected in sequence;

[0007] The robot data acquisition module is used to collect multi-dimensional operation data during the operation of the pet robot in real time;

[0008] The energy consumption status assessment module is used to perform structured aggregation on the multi-dimensional operation data based on the potential energy field mapping algorithm and the feature sparse modeling mechanism, extract the space-action correlation features of the energy consumption behavior, and generate an energy consumption influencing factor map;

[0009] The energy efficiency optimization module is used to search for the optimal energy consumption solution between the task execution path and the servo control strategy based on the energy consumption influencing factor map, and generate a low-energy consumption execution instruction set by using a quantum behavior particle swarm algorithm;

[0010] The automatic charging scheduling module is used to construct a heterogeneous graph structure with robots and charging piles as nodes based on the current power level of the robot, task urgency, spatial location information and energy consumption evaluation results, and use the graph attention network algorithm to adaptively model the state weight of the charging path. Combined with the task execution time window constraint scheduling mechanism, it dynamically generates a charging priority queue and an optimal path plan, and controls the pet robot to autonomously navigate to the charging station;

[0011] The battery health control module is used to perform online modeling based on battery charge and discharge cycle data, internal resistance change rate and temperature fluctuation information, using an incremental learning algorithm to evaluate the battery attenuation trend, generate health control instructions, and adjust the maximum charge and discharge current and temperature control parameters.

[0012] Furthermore, the multi-dimensional operation data includes the pet robot's three-dimensional motion inertial navigation data, joint servo current waveform, visual information, sound response, posture information, interactive reaction, current and voltage data, and environmental temperature and humidity gradient distribution.

[0013] Furthermore, the operation process of the energy consumption status assessment module includes the following steps:

[0014] Constructing a robot behavior state tensor for energy consumption modeling, aligning the multi-dimensional operation data in time and space to form an original feature tensor structure with time, spatial position and behavior category as index dimensions;

[0015] Based on the potential energy field mapping algorithm, the original feature tensor is mapped to the energy response field, the energy consumption density function of the behavior path is constructed, and the high energy consumption concentration area and energy consumption mutation point are identified through gradient analysis;

[0016] For the identified high-energy consumption areas and action segments, a feature sparse modeling mechanism is used to construct an energy consumption attribution mapping matrix, in which a channel compression algorithm based on the minimum entropy increase criterion is used to mine the behavioral feature subsets and their activation timing patterns that are most sensitive to the unit energy consumption growth.

[0017] Combining the task identification and spatial location, a cross-temporal and spatial energy consumption behavior influence map is constructed, based on which the influence centrality and intervention priority of each characteristic factor are calculated, and an energy consumption influence factor map is output.

[0018] Furthermore, the formula of the potential energy field mapping algorithm is as follows:

[0019] ;

[0020] where represents the energy consumption potential density at the spatial position (x, y) and time t where the robot is located; N represents the number of historical behavior samples; represents the current spatial coordinates; represents the spatial coordinates of the i-th historical data; and are the timestamps of the current and the i-th data respectively; represents the energy consumption behavior activation intensity corresponding to the i-th behavior data; represents the environmental disturbance factor; represents the behavior type modulation factor, which is used to distinguish the influence intensity of different actions on energy consumption; represents the spatial scale control parameter, which determines the smoothing range of potential energy diffusion in the spatial dimension; represents the time diffusion scale, which reflects the time decay rate of the response of historical behavior to current energy consumption; represents the angle between the direction vector of the i-th behavior and the current energy consumption gradient direction; represents the direction coupling adjustment coefficient.

[0021] Furthermore, the energy consumption attribution mapping matrix is specifically a sparse behavior attribution matrix constructed by compressing feature channels based on the minimum entropy increase criterion. The rows represent the key behavior feature dimensions sensitive to energy consumption, the columns represent the divided task behavior time periods, and the matrix elements are used to depict the attribution intensity of each feature to the unit energy consumption increment. It is generated by performing channel compression, time series division, and attribution modeling on multi-dimensional operation data.

[0022] Further, the operation process of the energy efficiency optimization module includes the following steps:

[0023] Based on the energy consumption impact factor map, extract the corresponding task execution order, spatial movement path, and servo control instructions, and construct a joint optimization variable set;

[0024] Initialize the optimization population and map each parameter combination of the joint optimization variable set to a search individual;

[0025] Use the quantum-behaved particle swarm algorithm to perform joint optimization of the path and servo strategy. Calculate the particle fitness function in each generation, and iteratively update the particle state according to the historical optimal solution and the current search direction;

[0026] In each iteration, the search individuals are comprehensively evaluated according to multi-dimensional indicators such as task execution time limit, path continuity, behavior control smoothness, and unit energy consumption level, and an optimal energy consumption control scheme is generated;

[0027] Map the optimal energy consumption control scheme to a low-energy consumption control instruction set, which includes a target task sequence, a servo joint angle scheduling curve, speed constraint parameters, and a state feedback frequency.

[0028] Furthermore, the formula of the quantum behavior particle swarm algorithm is as follows:

[0029] ;

[0030] Where, represents the optimal control parameter solution of the pet robot in the current task cycle; represents the historical optimal solution position of the particle in the k-th generation; represents the dynamic step size coefficient, which is jointly set according to the task urgency, its own power level, and the energy consumption evaluation value; represents the global optimal solution position in the current iteration, which is obtained by combining the historical behavior map and the energy consumption index evaluation; represents the non-linear convergence adjustment factor; represents the random perturbation variable.

[0031] Further, the operation process of the automatic charging scheduling module includes the following steps:

[0032] Collect the current battery percentage, remaining task time, spatial coordinates of the robot, and the reachability information between it and the surrounding charging piles, and construct a heterogeneous directed graph with the robot nodes and charging pile nodes as vertices. The edge weights of the heterogeneous directed graph are initialized and modeled by combining the spatial distance, energy consumption estimate value, and path risk coefficient;

[0033] Based on the graph attention network, train and model the heterogeneous directed graph, and use the attention mechanism to perform weighted learning on the power level, task urgency, and path cost between nodes respectively, adaptively adjust the preference degree of each robot for candidate charging pile nodes, and output the path decision attention matrix;

[0034] Combined with the path decision attention matrix, construct a path optimization function based on the remaining task time and navigation energy consumption, dynamically generate a set of charging candidate paths, calculate the scheduling priority scores of each path, and select the optimal charging scheduling path with the minimum path cost and meeting the time window constraint;

[0035] Generate a navigation control instruction set according to the optimal charging scheduling path to drive the pet robot to autonomously navigate to a specified charging pile for charging tasks; the navigation control instruction set includes a sequence of path points, steering angle control, obstacle avoidance trigger conditions, and a remaining battery level threshold trigger mechanism.

[0036] Furthermore, the formula of the path preference function is as follows:

[0037] ;

[0038] Wherein, represents the scheduling priority score of the candidate charging path r; represents the energy consumption per unit distance of the path r; represents the Euclidean path length between the current spatial coordinates of the robot and the candidate charging pile; represents the maximum navigation distance that the remaining battery level of the current robot can support; represents the estimated risk cost coefficient of the path r; represents the ratio of the path time consumption to the remaining time window of the task; represents the preference weight of the pet robot for each key node in the path; and represents the cost factor weight coefficient.

[0039] Further, the battery health control module specifically has a hierarchical health warning mechanism. According to the battery life prediction result and the health threshold range, the battery state is divided into four levels: normal, mild degradation, critical deterioration, and severe failure.

[0040] The beneficial effects of the present invention are as follows:

[0041] The present invention collects multi-dimensional operation data including motion inertial navigation, servo current, attitude, environment, etc. through a robot data acquisition module, and performs aggregation processing based on a potential field mapping and feature sparse modeling mechanism, which can accurately capture the correlation features between energy consumption behaviors and spatial actions, and improve the timeliness and accuracy of energy consumption analysis. The energy efficiency optimization module, based on the energy consumption factor map, uses the quantum behavior particle swarm optimization algorithm to jointly optimize the execution path and control strategy, effectively searches for the globally optimal low-energy consumption solution, avoids the traditional method from falling into the local optimum, and improves the running efficiency of the robot and the power utilization rate. By constructing a heterogeneous graph structure with the robot and the charging pile as nodes, and introducing a graph attention network model to dynamically model the charging path state, the charging task can be flexibly scheduled according to the power state, task urgency, and location information, improving the intelligence and optimality of the charging path selection, and realizing truly automatic navigation charging. The battery health control module uses an incremental learning algorithm to perform online modeling and trend prediction on the battery state, and can dynamically adjust the health control strategy according to the charging and discharging characteristics and thermal characteristics, improving the system's response ability to battery anomalies, extending the battery life, and ensuring the safe operation of the device. Each module works collaboratively through communication connections, constructing a closed-loop control process from data acquisition, energy consumption assessment to energy efficiency optimization, automatic charging, and battery health management, with good system integration and intelligent response capabilities, significantly improving the endurance and running stability of the pet robot in complex application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 FIG. is a schematic diagram of the modules of an energy consumption optimization and automatic charging control system for an intelligent pet robot according to the present invention.

[0043] Figure 2 FIG. is a schematic flow chart of the operation process of the energy efficiency optimization module provided by an embodiment of the present invention.

[0044] Figure 3 FIG. is a schematic flow chart of the operation process of the automatic charging scheduling module provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] Please refer to Figures 1-3 as shown, the present invention relates to an energy consumption optimization and automatic charging control system for an intelligent pet robot.

[0046] Embodiment

[0047] An energy consumption optimization and automatic charging control system for an intelligent pet robot includes a robot data acquisition module, an energy consumption status assessment module, an energy efficiency optimization module, an automatic charging scheduling module, and a battery health control module that are sequentially communicatively connected;

[0048] The robot data acquisition module is used to collect multi-dimensional operation data during the operation of the pet robot in real time; the multi-dimensional operation data includes three-dimensional motion inertial navigation data, joint servo current waveforms, visual information, sound responses, pose information, interaction responses, current and voltage data, and environmental temperature and humidity gradient distributions of the pet robot.

[0049] It should be noted that the hardware composition and distribution layout are as follows:

[0050] Inertial Measurement Unit (IMU): Integrated with a three-axis accelerometer and a three-axis gyroscope, it is arranged on the main body or the center of mass of the robot to collect three-dimensional acceleration and angular velocity, with a resolution of 0.01g / ° / s, and the sampling frequency can be set to 100 - 1000Hz;

[0051] Current / Voltage Sensor: Connected to the robot servo control board and the battery management system, it collects the current waveforms of each joint servo motor and the power supply voltage fluctuations to evaluate the execution status and power consumption;

[0052] Visual Sensor System: Equipped with a wide-angle camera (RGB-D depth camera is optional), it is used to collect image data and depth information within the field of view of the robot, and supports tasks such as target recognition and user recognition;

[0053] Microphone Array: Usually a MEMS microphone matrix (such as a 4-microphone circular arrangement), it is used to capture sound response information, such as voice commands and environmental sound source localization;

[0054] Joint Angle and Pose Encoder: Embedded in the servo drive unit, it records the angle changes of each moving joint in real time and performs pose fusion calculation in combination with IMU data;

[0055] Tactile Sensor and Interaction Unit: Includes pressure-sensitive diaphragms or capacitive touch modules distributed on the case, head, and back, which are used to detect interactive behaviors such as user stroking and tapping;

[0056] Environmental Sensor: Arranged with a temperature and humidity integrated module (such as DHT22) and an infrared thermal sensor array module, it constructs a temperature and humidity distribution map of the surrounding space to form environmental gradient information.

[0057] To reduce data redundancy and power consumption, this module also has a dynamic sampling and event-driven mechanism: when the robot is in a stationary or standby state, non-critical sensors such as the vision and audio modules enter the low-power mode and are only awakened when triggered by external stimuli; the main controller regularly evaluates data relevance and energy consumption, and uses an information gain threshold to control the sampling rate of each sensor; it supports an edge adaptive compression coding (such as T-DPCM + Huffman) mechanism to locally compress continuous sampling data before transmission to reduce communication bandwidth consumption.

[0058] The energy consumption status assessment module is used to perform structured aggregation on the multi-dimensional operation data based on the potential energy field mapping algorithm and the feature sparse modeling mechanism, extract the space-action correlation features of the energy consumption behavior, and generate an energy consumption influencing factor map;

[0059] The operation process of the energy consumption status assessment module includes the following steps:

[0060] Constructing a robot behavior state tensor for energy consumption modeling, aligning the multi-dimensional operation data in time and space to form an original feature tensor structure with time, spatial position and behavior category as index dimensions;

[0061] Based on the potential energy field mapping algorithm, the original feature tensor is mapped to the energy response field, the energy consumption density function of the behavior path is constructed, and the high energy consumption concentration area and energy consumption mutation point are identified through gradient analysis;

[0062] It should be noted that the behavior state tensor is input into the pre-built potential field mapping module, and the system constructs a three-dimensional energy consumption field model based on the behavior activation intensity, movement posture change amplitude and power consumption index.

[0063] In specific implementation, the system estimates the energy consumption density of continuous actions based on the robot's path trajectory and execution instruction sequence, forming a spatial distribution map. Each spatial node of the spatial distribution map is marked with the unit behavior energy consumption value in the corresponding time period, forming an intuitive and visual energy consumption "heat map".

[0064] During this process:

[0065] Perform motion-energy response analysis on each segment of the trajectory; mark the spatial segments where posture changes frequently or the servo current continues to fluctuate abnormally; automatically detect areas where the energy gradient changes sharply and mark them as "mutation points"; identify high-energy consumption patterns that recur in specific postures, movements or interactive behaviors, such as frequent waving, continuous standing, and sharp turns.

[0066] The output of this stage includes: high-energy consumption hot spots, energy consumption mutation behavior segments and their corresponding location coordinates and time windows, providing preliminary screening basis for feature screening and attribution analysis.

[0067] For the identified high-energy consumption areas and action segments, a feature sparse modeling mechanism is used to construct an energy consumption attribution mapping matrix, in which a channel compression algorithm based on the minimum entropy increase criterion is used to mine the behavioral feature subsets and their activation timing patterns that are most sensitive to the unit energy consumption growth.

[0068] Specifically, this stage adopts the channel compression and attribution modeling mechanism, including the following steps:

[0069] In the identified high - energy - consumption paragraphs, statistically analyze the temporal distribution of all original features (such as servo current, acceleration, angular velocity, temperature, visual frame change rate, etc.); conduct a correlation analysis on the change trend of each feature and the change of energy consumption per unit time; sort the feature channels according to the importance distribution of the features, and retain the top several significantly influential channels.

[0070] Divide a complete task behavior into multiple stages according to logic or time periods, such as "start - up stage", "interaction stage", "turning stage", "return stage", etc.; bind each stage to system logs or user commands to form traceable behavior - time - period labels.

[0071] During the key behavior time periods, model the compressed features to construct a sparse behavior attribution matrix; each matrix element represents the driving degree of a "specific behavior feature" on the energy - consumption fluctuation in a "specific task stage"; after normalization, this matrix can be used for horizontal comparison of multiple task scenarios to identify general - type and scenario - specific energy - consumption factors.

[0072] Combined with the activation timing of sensor data, identify the change rhythm of feature values before and after the energy - consumption peak, and analyze whether there are patterns such as "pre - activation", "co - activation", or "lag - activation"; further analyze whether some high - energy - consumption points are triggered by multiple features in cooperation or a single factor, so as to precisely control the optimization object.

[0073] The final output is: a set of key feature dimensions, their action windows in task behaviors, and the corresponding energy - consumption attribution weights, providing factor sources for atlas construction.

[0074] Combine the task identifier and spatial location to construct a cross - time - space energy - consumption behavior impact graph, calculate the influence - degree centrality and intervention priority of each feature factor based on the energy - consumption behavior impact graph, and output the energy - consumption impact factor atlas.

[0075] It should be noted that the energy - consumption behavior impact graph is a directed weighted graph structure, and the energy - consumption behavior impact graph includes several nodes and edges;

[0076] The nodes represent the key behavior units decomposed from the robot in the task process, and each node contains the corresponding behavior type, execution time period, spatial location, and unit energy - consumption index;

[0077] The edges represent the coupling associations between behavior units in terms of spatial paths or energy - consumption features, used to describe the cooperation relationship or adjacent scheduling relationship between two behaviors during execution. Further, edge weights are set on the edges to quantify the strength of this coupling association. The edge weights are comprehensively modeled based on the co - occurrence frequency of energy - consumption feature factors and the spatial - path overlap degree, reflecting the propagation effect of feature factors among multiple behavior units and the energy - consumption superposition trend.

[0078] Specifically, based on the complete task process of the robot, it is decomposed into multiple task behavior units, such as "starting to move", "environmental perception", "task response", "path correction", etc.; each behavior unit serves as a graph node, attached with its timestamp, position coordinates, main control instructions, and energy consumption labels.

[0079] Based on the "multi-behavior - single feature" or "single-behavior - multi-feature" relationships found in the sparse behavior attribution matrix, edges are established between behavior nodes; the weights of the edges are calculated based on the sharing degree and energy consumption contribution degree of these features among various behaviors; at the same time, spatial path coupling is considered, that is, although some behaviors belong to different task units, their execution paths physically coincide, and path influence labels are attached to the edges.

[0080] Using graph theory betweenness centrality or propagation centrality algorithms, calculate the structural influence of each feature factor (or behavior node) in the entire energy consumption behavior influence graph; this analysis not only considers the energy consumption contribution of the node itself, but also considers its triggering effect on the energy consumption chain propagated to other nodes.

[0081] Fuse and evaluate the influence with information such as execution frequency and position concentration to generate a feature intervention priority list; features or behavior segments with high priority will be marked for key constraints, substitution, or adjustment by the subsequent energy efficiency optimization module;

[0082] The finally generated energy consumption influence factor map is a multi-dimensional graph structure, which not only visually displays the complex relationship between behavior and energy consumption, but also provides a basis for task path reconstruction, strategy switching, and charging scheduling.

[0083] The energy consumption attribution mapping matrix is specifically a sparse behavior attribution matrix constructed by compressing feature channels based on the minimum entropy increase criterion. The rows represent the key behavior feature dimensions sensitive to energy consumption, the columns represent the divided task behavior time periods, and the matrix elements are used to depict the attribution intensity of each feature to the unit energy consumption increment, which is generated by channel compression, time series division, and attribution modeling of multi-dimensional operation data.

[0084] Furthermore, the formula of the potential field mapping algorithm is as follows:

[0085] ;

[0086] Among them, represents the energy consumption potential density at the spatial position (x, y) and time t where the robot is located; N represents the number of historical behavior samples, which is extracted from the three-dimensional motion inertial navigation data, joint servo current waveforms, and voltage and current data collected by the robot data acquisition module; represents the current spatial coordinates; represents the spatial coordinates of the i-th historical data; and are the timestamps of the current and the i-th data respectively, which are time tags from the robot operation data; represents the activation intensity of the energy consumption behavior corresponding to the i-th behavior data, which is comprehensively normalized by the instantaneous amplitude of the joint servo current, voltage fluctuation and attitude stability, and is used to quantify the direct contribution of this behavior to the overall energy consumption; represents the environmental disturbance factor, which combines the change rate of temperature and humidity gradient and the change amplitude of the sound response signal to reflect the driving disturbance of the local environment on the energy consumption behavior; represents the behavior type modulation factor, which is used to distinguish the influence intensity of different actions on the energy consumption; represents the spatial scale control parameter, which determines the smoothing range of the potential energy diffusion in the spatial dimension and is set according to the activity radius range of each component of the robot; represents the time diffusion scale, which reflects the time decay rate of the historical behavior on the current energy consumption response; represents the angle between the direction vector of the i-th behavior and the current energy consumption gradient direction, which is calculated by combining the pose change trend and the energy consumption density gradient direction; represents the direction coupling adjustment coefficient, which is used to regulate the coupling intensity of the robot behavior direction and the energy consumption field gradient on the final energy consumption density response.

[0087] The energy efficiency optimization module is used to search for the optimal energy consumption solution between the task execution path and the servo control strategy based on the energy consumption influence factor map by using the quantum-behaved particle swarm algorithm, and generate a low-energy consumption execution instruction set;

[0088] Among them, the operation process of the energy efficiency optimization module includes the following steps:

[0089] Based on the energy consumption influence factor map, extract the corresponding task execution sequence, spatial movement path and servo control instructions, and construct a joint optimization variable set;

[0090] Specifically, the system first calls the energy consumption influence factor map generated by the energy consumption status evaluation module, and extracts the relevant task units, spatial path nodes and key control features in the current task scenario.

[0091] Specifically include:

[0092] Task execution sequence: such as "move to the interaction point → interact with the user → return to the base station", etc., extract the logical sequence and execution window of each behavior;

[0093] Spatial movement path: identify the set of spatial trajectories that the robot may choose, including the distances of different path segments, environmental constraints and historical energy consumption labels;

[0094] Servo control parameters: the set of servo parameters for each task stage, such as joint rotation angle, execution speed, response inertia, etc.

[0095] The system combines the information in the above three dimensions into a set of jointly optimized variable sets. Each set of variables contains a complete configuration combination of a task path - action control, which is used as the basic unit of the solution space for the subsequent search algorithm.

[0096] Initialize the optimization population and map each set of parameter combinations in the jointly optimized variable set to a search individual;

[0097] Use the quantum - behaved particle swarm algorithm for the joint optimization of the path and servo strategy. Calculate the particle fitness function in each generation, and iteratively update the particle state according to the historical optimal solution and the current search direction;

[0098] It should be noted that this step mainly consists of the following sub - processes:

[0099] 1. Particle representation structure: Each particle is a candidate solution, and its state vector contains three parts: a set of encodings of the task execution order (for example, representing the task arrangement with an integer sequence); an index for path selection (selecting a certain path from the alternative path library); a set of servo parameter combinations composed of continuous variables (such as angular velocity limit, joint rotation amplitude, acceleration threshold, etc.); Each particle represents a complete "execution plan".

[0100] 2. Particle behavior update mechanism: In each round of iteration, the particle state is updated by sampling the quantum position update model, rather than simple velocity - position evolution: Use the current global optimal particle (gbest) to guide the probability distribution of all particles; Introduce a "center - guiding mechanism" so that each particle not only approaches its individual historical optimum but also approaches the group mean, reflecting the aggregation of swarm intelligence; In the update of path selection and task order variables, use probability perturbation and re - sorting strategies to introduce jumpiness and maintain population diversity.

[0101] 3. Coupling optimization mechanism: This module executes a multi - variable joint optimization problem, that is, simultaneously adjusting the discrete task sequence, path selection index, and continuous servo parameters: The change of the task order directly affects the time - window scheduling and servo load rhythm; The path selection changes the overall spatial energy - consumption structure;

[0102] The adjustment of control parameters affects the single - behavior energy consumption and system stability; There is a high degree of coupling among the three, and the joint evolution of the solution space is realized through a unified particle state encoding structure.

[0103] In each round of iteration, comprehensively evaluate the search individuals according to multi - dimensional indicators such as task execution time limit, path continuity, behavior control smoothness, and unit energy - consumption level, and generate an optimal energy - consumption control scheme;

[0104] It should be noted that after one round of iteration is completed, the system needs to evaluate the solutions of all particles, select the best-performing candidate solution, and guide the search direction for the next round. This evaluation is no longer a single-index optimization, but a weighted comprehensive evaluation based on multi-dimensional performance indicators to ensure the practicality and stability of the final solution in actual operation.

[0105] The main evaluation indicators are as follows:

[0106] 1. Task completion index: Check whether the solution represented by the particle can complete all tasks within the set time window, and whether there are any omissions or repeated call logic errors. If there are logical loopholes, the solution will be judged to have a low fitness.

[0107] 2. Path rationality index: Conduct continuity, closure, and reachability tests on the movement path, including avoiding backward paths, non-physically feasible paths, or paths that exceed the obstacle range.

[0108] 3. Servo control smoothness index: Analyze the change trend of each segment of the motion control curve to detect whether there are situations such as overly frequent servo actions, too rapid acceleration and deceleration switching, or discontinuous actions. These can lead to energy consumption waste and execution instability.

[0109] 4. Unit energy consumption index: Based on the weight labels in the energy consumption impact factor map, combined with the simulated execution path and the servo control model, calculate the estimated energy consumption value of each particle per unit time and per unit distance. This value is one of the most important target parameters of the optimization algorithm.

[0110] 5. System response redundancy: Analyze whether sufficient buffer time and servo tolerance space are reserved for each task node under the particle solution. For example, when performing certain high-dynamic tasks (such as rapid turning), whether the servo is allowed to warm up or inertial release to avoid extreme loads.

[0111] After synthesizing the above multi-dimensional indicators, the system generates the final fitness score for each particle through normalization and weighted fusion. The scoring result not only determines the current optimal particle but also is used for the quantum behavior iteration guidance of the next generation of particles.

[0112] Map the optimal energy consumption control solution to a low-energy consumption control instruction set, which includes the target task sequence, servo joint angle scheduling curve, speed constraint parameters, and state feedback frequency.

[0113] It should be noted that after all iterations are completed (or early convergence meets the termination conditions), the system will automatically extract the solution of the global optimal particle and structurally transform it into a low-energy consumption control instruction set, which is sent to the robot motion controller and task executor.

[0114] This control instruction set mainly includes the following four parts:

[0115] 1. Task sequence planning instructions: including task numbers, execution order, and start trigger conditions; each task is accompanied by an execution time window (earliest - latest start time); support for nested interruption recovery and task jump instructions to improve fault tolerance.

[0116] 2. Path navigation and attitude transition instructions: clearly specify the node numbers and spatial displacement targets of the robot's movement path; include attitude connection modes (smooth turning / step adjustment); configure specific path weights for different terrain types or interaction areas.

[0117] 3. Servo joint scheduling parameter set: provide the angle change curve of each servo joint within the task segment; specify the start speed, peak speed, and end deceleration control of each segment of movement; especially set the power constraint of the servo motor, response time limit, and safety braking strategy.

[0118] 4. State feedback and monitoring frequency setting: specify the data acquisition frequency of key nodes (such as sampling 10 times per second); enable a dynamic feedback adjustment mechanism. When an overload or abnormal time consumption occurs in a certain task segment, the instruction can temporarily adjust subsequent parameters to achieve online adaptability.

[0119] The finally generated control instruction set will be uploaded to the robot's embedded controller as an execution template and can be iteratively used in subsequent tasks, cooperating with the system learning module to achieve long - term energy consumption optimization.

[0120] Furthermore, the formula of the quantum - behaved particle swarm algorithm is as follows:

[0121] ;

[0122] Where, represents the optimal control parameter solution of the pet robot within the current task cycle; represents the historical optimal solution position of the particle in the k - th generation; represents the dynamic step - size coefficient, which is jointly set according to the task urgency, its own battery level, and energy consumption evaluation value; represents the global optimal solution position in the current iteration, obtained by combining the historical behavior map and energy consumption index evaluation; represents the non - linear convergence adjustment factor, which reflects the expected range of path continuity and servo control smoothness, and is calculated from the servo speed constraint and path smoothness index in the joint optimization variable set; represents the random perturbation variable, in the interval (0,1), which is dynamically generated in combination with the quantum search mechanism in the energy - efficiency optimization module and the global search diversity requirement, and is used to simulate the perturbation information entropy of the search space.

[0123] The automatic charging scheduling module is used to construct a heterogeneous graph structure with the robot and the charging pile as nodes based on the current battery level of the robot, the urgency of the task, the spatial location information, and the energy consumption evaluation result. The graph attention network algorithm is used to adaptively model the state weights of the charging path. Combining with the task execution time window constraint scheduling mechanism, a charging priority queue and an optimal path plan are dynamically generated, and the pet robot is controlled to autonomously navigate to the charging station;

[0124] Among them, the operation process of the automatic charging scheduling module includes the following steps:

[0125] Collect the current battery percentage, remaining task time, spatial coordinates of the robot, and the reachability information between the robot and the surrounding charging piles, and construct a heterogeneous directed graph with the robot node and the charging pile node as vertices. The edge weights of the heterogeneous directed graph are initialized and modeled by combining the spatial distance, the estimated energy consumption value, and the path risk coefficient;

[0126] Specifically, the system first collects the core state data related to the scheduling decision in the scheduling server or the local edge computing unit, including: the remaining battery percentage of the current pet robot; the remaining time window of the task being executed or to be executed; the spatial coordinates of the current location (two-dimensional or three-dimensional map coordinates); the spatial locations, occupancy status, and queuing situations of all nearby charging piles; the path connectivity, distance, obstacle information, and risk assessment (such as narrow channels, electromagnetic interference areas, etc.) between the current location of the robot and each charging pile.

[0127] Based on this, the system constructs a heterogeneous directed graph structure, and the node types of the heterogeneous directed graph structure include: robot nodes, charging pile nodes;

[0128] The edge represents the path that the robot can reach a certain charging pile, and the edge weight initialization includes: the spatial path distance; the estimated energy consumption of the current path (combined with the slope, turning frequency, etc.); the path safety or stability level (i.e., the path risk coefficient, such as aisle congestion, interference area, etc.).

[0129] Based on the graph attention network, the heterogeneous directed graph is trained and modeled. The attention mechanism is used to perform weighted learning on the battery level, task urgency, and path cost between nodes respectively, and adaptively adjust the preference degree of each robot for the candidate charging pile nodes, and output the path decision attention matrix;

[0130] Specifically, after establishing the heterogeneous graph structure, the system calls the graph attention network model (GAT) deployed locally or on the server side to perform preference learning and state modeling on the charging path.

[0131] This process includes the following:

[0132] 1. Node Feature Encoding: Encode the current battery level and task urgency (remaining time and remaining actions of the task) for robot nodes; encode whether a charging pile node is occupied, the path length from the robot's position, and the current waiting queue length; all node features are converted into a unified vector format through an embedding mechanism and used as the input to the graph network.

[0133] 2. Attention Mechanism Training: The graph attention network takes each robot as the center and models the edge weight attention for all candidate charging pile nodes connected to it; during the training process, the weighted learning path cost, battery urgency, and task execution pressure are used to dynamically adjust the preference weights of the robot for different charging piles; the training outputs a path decision attention matrix, indicating the willingness or priority of the robot to choose different charging paths.

[0134] Combined with the path decision attention matrix, construct a path optimization function based on the remaining task time and navigation energy consumption, dynamically generate a set of candidate charging paths, calculate the scheduling priority scores of each path, and select the optimal charging scheduling path with the minimum path cost and meeting the time window constraint;

[0135] Specifically, after obtaining the path decision attention matrix, the system further performs refined path optimization, constructs a comprehensive scoring function by combining task deadlines, path energy consumption, and behavior continuity, etc., and selects the final charging path.

[0136] This step includes the following:

[0137] 1. Construct a set of candidate charging paths: Screen several charging pile nodes with higher attention scores; for each target node, call the map module to plan the actual path of the robot to this node, including the waypoint sequence and estimated time consumption; screen out currently unavailable charging piles (such as occupied or temporarily blocked paths);

[0138] 2. Calculate the scheduling priority score for each path, mainly including: the remaining task time minus the navigation time required, that is, the task urgency - navigation time margin; the power matching degree between the cumulative path energy consumption and the current battery level of the robot; the path preference score output by the attention network; the impact of the obstacle avoidance complexity and turning frequency of the current path on the navigation execution stability;

[0139] 3. Screen the optimal path plan: Standardize all scoring results; eliminate paths that do not meet the time window limit or exceed the energy consumption limit; select the path with the highest comprehensive score as the optimal scheduling path for this round.

[0140] Generate a navigation control instruction set according to the optimal charging scheduling path, and drive the pet robot to autonomously navigate to the specified charging pile for the charging task; the navigation control instruction set includes a waypoint sequence, steering angle control, obstacle avoidance trigger conditions, and a remaining battery threshold trigger mechanism.

[0141] Specifically, the instruction set includes the following:

[0142] 1. Waypoint sequence: Clearly define the consecutive path nodes that the robot needs to pass through during movement, and mark the spatial coordinates and expected arrival time of each node.

[0143] 2. Steering angle control information: Include the steering angle, rotation direction, and angle change rate required by the robot at each node to ensure smooth and energy-efficient movement.

[0144] 3. Obstacle avoidance trigger conditions: Configure local obstacle avoidance trigger rules for each path segment. For example, switch to the local path planning mode or pause mode when the distance from an obstacle is less than a certain threshold.

[0145] 4. Remaining battery threshold control mechanism: Set the "emergency return threshold". If the system detects that the battery level is lower than the set threshold during the journey to the charging pile, immediately switch to the backup charging path or the nearest charging point; configure a status re-evaluation mechanism to re-evaluate the path when the route is interrupted or the task time window changes.

[0146] Furthermore, the formula of the path preference function is as follows:

[0147]

[0148] Where, represents the scheduling priority score of the candidate charging path r, which is used for sorting and screening in the path set. The smaller the value, the higher the priority; represents the energy consumption per unit distance of the path r, which is calculated based on the current servo behavior control parameters of the task and the map terrain information; represents the Euclidean path length between the current spatial coordinates of the robot and the candidate charging pile; represents the maximum navigation distance that the remaining battery of the current robot can support, which is inversely deduced based on the remaining battery capacity and the unit navigation energy consumption evaluated in the robot battery health control module; represents the estimated risk cost coefficient of the path r; represents the ratio of the path time consumption to the remaining time window of the task, which is used to describe whether the current path can complete the charging task within the task time limit; represents the preference weight of the pet robot for each key node in the path; and represent the cost factor weight coefficient, which dynamically adjusts the relative importance of the two dimensions of "energy consumption - distance" and "risk - credibility", and is usually allocated by the policy layer according to the importance of the current task.

[0149] The battery health control module is used to perform online modeling using an incremental learning algorithm based on battery charge and discharge cycle data, internal resistance change rate, and temperature fluctuation information, evaluate the battery degradation trend, and generate health control instructions to adjust the maximum charge and discharge current and temperature control parameters; the battery health control module specifically has a hierarchical health warning mechanism, and divides the battery state into four levels: normal, mild degradation, critical deterioration, and severe failure according to the battery life prediction result and the health threshold range.

[0150] It should be noted that establishing a battery multivariable degradation prediction model based on the incremental learning algorithm specifically includes:

[0151] Using the incremental extreme learning machine (IELM) to input the recent multi-dimensional time series characteristics of the battery, dynamically adjust the weights of the hidden layer, and continuously fine-tune the fitting accuracy of the model for the battery SOH;

[0152] Fusing a recursive linear regression model (such as RLS) to track the change trajectory of the internal resistance and derive the maximum degradation rate within a certain future period;

[0153] Introducing a Bayesian confidence interval prediction mechanism to generate a confidence interval for the prediction result, evaluate the prediction reliability, and feedback it to the scheduling module for task adaptation adjustment;

[0154] Special event identification: Combining the behavior of the battery temperature rising and the internal resistance rising non-linearly, identifying thermal runaway or early fault hidden dangers through an autoencoder anomaly detection network.

[0155] Specifically, after obtaining the predicted health status result, this sub-module generates the following control parameter instructions according to the standardized battery behavior model:

[0156] Limit the maximum charging rate (C-rate): When the battery is in the mild or moderate degradation stage, reduce the C-rate from 1.0C to 0.5C to slow down the formation of lithium dendrites;

[0157] Adjust the charging termination voltage threshold: Dynamically set the maximum charging voltage according to the temperature state (such as adjusting from 4.2V to 4.1V);

[0158] Control the discharge cut-off threshold: Raise the minimum working voltage (such as from 3.0V to 3.2V) to avoid over-discharge;

[0159] Trigger the active heat dissipation mechanism: If the temperature exceeds the limit but has not entered the fault state, control the PWM fan or Peltier component to cool down quickly;

[0160] Dynamically adjust the task scheduling priority: If the battery is approaching the health critical threshold, transfer the task to other robot nodes to improve the system robustness.

[0161] It should be noted that the battery health control module is equipped with a hierarchical health warning mechanism, which is used to classify the battery according to the predicted battery life parameters and the current health status, and trigger corresponding response strategies. Specifically:

[0162] When the system evaluates that the health status of the battery is good, that is, the state of health (SOH) value is greater than or equal to 90%, and the internal resistance change and temperature rise trend are within the normal fluctuation range, the system determines that the battery is in a normal state, without special intervention, and allows the robot to continue running according to the current task execution plan;

[0163] If the system monitors that the battery SOH is between 70% and 90%, or there is a slight increase in internal resistance and abnormal temperature fluctuation, but it does not affect the task execution duration and current output stability, the system classifies the battery as a mild degradation state. In this state, the system will automatically reduce the charging rate and prompt the user to perform regular maintenance to delay performance degradation;

[0164] When the battery's SOH drops to between 50% and 70%, or the model prediction shows that its available life is less than one typical task cycle, the system determines that the battery is in a critical degradation state, and will immediately trigger the operation limit mechanism, lower the upper limit power of the servo drive, and preferentially transfer the remaining tasks to other robot nodes;

[0165] If the system detects that the battery's SOH is already below 50%, or discovers non-linear failure signs such as a sharp temperature rise, voltage drop, or internal resistance mutation, it is considered that the battery has entered a serious failure state. The system will abort all current task processes, enforce power-off protection, and notify the user to replace the battery as soon as possible through voice broadcast or mobile application to ensure safe use.

[0166] The above classification of health levels and dynamic responses can form an autonomous closed-loop mechanism, enabling the intelligent pet robot to have the capabilities of battery safety prediction, risk avoidance, and adaptive control during long-term operation.

[0167] In summary, the present invention integrates multiple types of sensors such as IMU, vision, audio, current and voltage, combines the potential energy field mapping and sparse modeling mechanism, not only realizes the spatial-behavior aggregation evaluation of energy consumption, but also visualizes the generation of the energy consumption impact factor map, providing a structured support for subsequent optimization. By adopting the minimum entropy increase criterion and sparse channel compression strategy, high-sensitive behavior features and their activation patterns are extracted, the attribution sources of abnormal energy consumption in different task stages are clarified, and the interpretability analysis of high-energy consumption behaviors and the priority ranking of interventions are realized.

[0168] The present invention introduces the quantum behavior update model and the group center guidance mechanism in the joint optimization of task paths and servo strategies, realizes the coupled search across task sequences, spatial trajectories and servo parameters, and greatly improves the global optimization capability and search stability of low-energy instruction set generation. The heterogeneous graph scheduling framework established based on the GAT network can integrate power, task urgency and path cost information in real time, realize the linkage optimization of charging priority sorting and path selection, and improve the accuracy and robustness of scheduling response.

[0169] The present invention uses incremental extreme learning and recursive regression models to track the battery SOH in real time, and combines Bayesian prediction and anomaly detection mechanisms to achieve multi-level health status recognition and adaptive adjustment of corresponding charging and discharging parameters, effectively extending battery life and ensuring operational safety. By deeply coupling the energy consumption map with the scheduling algorithm, the system can adjust the path, servo response and charging timing in real time according to task changes, forming a closed-loop optimization control mechanism with energy consumption as the core, significantly reducing system operating power consumption, and improving endurance and task execution stability.

[0170] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. An intelligent pet robot energy consumption optimization and automatic charging control system, characterized in that, It includes a robot data acquisition module, an energy consumption status evaluation module, an energy efficiency optimization module, an automatic charging scheduling module and a battery health control module which are sequentially connected in communication; The robot data acquisition module is used to collect multi-dimensional operation data of the pet robot during operation in real time; The energy consumption status assessment module is used to perform structured aggregation on the multi-dimensional operation data based on the potential energy field mapping algorithm and the feature sparse modeling mechanism, extract the space-action correlation features of the energy consumption behavior, and generate an energy consumption influencing factor map; The energy efficiency optimization module is used to search for the optimal energy consumption solution between the task execution path and the servo control strategy based on the energy consumption influencing factor map, and generate a low-energy consumption execution instruction set by using a quantum behavior particle swarm algorithm; The automatic charging scheduling module is used to construct a heterogeneous graph structure with robots and charging piles as nodes based on the current power level of the robot, task urgency, spatial location information and energy consumption evaluation results, and use the graph attention network algorithm to adaptively model the state weight of the charging path. Combined with the task execution time window constraint scheduling mechanism, it dynamically generates a charging priority queue and an optimal path plan, and controls the pet robot to autonomously navigate to the charging station; The battery health control module is used to perform online modeling based on battery charge and discharge cycle data, internal resistance change rate and temperature fluctuation information, using an incremental learning algorithm to evaluate the battery attenuation trend, generate health control instructions, and adjust the maximum charge and discharge current and temperature control parameters.

2. The energy consumption optimization and automatic charging control system of an intelligent pet robot according to claim 1, characterized in that, The multi-dimensional operation data includes the pet robot's three-dimensional motion inertial navigation data, joint servo current waveform, visual information, sound response, posture information, interactive reaction, current and voltage data, and environmental temperature and humidity gradient distribution.

3. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 1, characterized in that, The operation process of the energy consumption status assessment module includes the following steps: Constructing a robot behavior state tensor for energy consumption modeling, aligning the multi-dimensional operation data in time and space to form an original feature tensor structure with time, spatial position and behavior category as index dimensions; Based on the potential energy field mapping algorithm, the original feature tensor is mapped to the energy response field, the energy consumption density function of the behavior path is constructed, and the high energy consumption concentration area and energy consumption mutation point are identified through gradient analysis; For the identified high-energy consumption areas and action segments, a feature sparse modeling mechanism is used to construct an energy consumption attribution mapping matrix, in which a channel compression algorithm based on the minimum entropy increase criterion is used to mine the behavioral feature subsets and their activation timing patterns that are most sensitive to the unit energy consumption growth. Combining the task identification and spatial location, a cross-temporal and spatial energy consumption behavior influence map is constructed, based on which the influence centrality and intervention priority of each characteristic factor are calculated, and an energy consumption influence factor map is output.

4. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 3, characterized in that, The formula of the potential energy field mapping algorithm is as follows: ； Among them, represents the energy consumption potential density at the spatial position (x, y) and time t where the robot is located; N represents the number of historical behavior samples; represents the current spatial coordinate; represents the spatial coordinate of the i-th historical data; and are the timestamps of the current and the i-th data respectively; represents the activation intensity of the energy consumption behavior corresponding to the i-th behavior data; represents the environmental disturbance factor; represents the behavior type modulation factor, which is used to distinguish the influence intensity of different actions on energy consumption; represents the spatial scale control parameter, which determines the smoothing range of potential energy diffusion in the spatial dimension; represents the time diffusion scale, which reflects the time decay rate of the response of historical behavior to the current energy consumption; represents the angle between the direction vector of the i-th behavior and the current energy consumption gradient direction; represents the direction coupling adjustment coefficient.

5. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 3, characterized in that, The energy consumption attribution mapping matrix is ​​specifically a sparse behavior attribution matrix constructed based on the minimum entropy increase criterion to compress the feature channel. The rows represent the key behavior feature dimensions that are sensitive to energy consumption, and the columns represent the divided task behavior time periods. The matrix elements are used to characterize the attribution intensity of each feature to the unit energy consumption increment. It is generated by channel compression, time series division and attribution modeling of multidimensional operation data.

6. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 1, characterized in that, The operation process of the energy efficiency optimization module includes the following steps: Based on the energy consumption impact factor map, extract the corresponding task execution order, spatial movement path, and servo control instructions to construct a joint optimization variable set; Initialize the optimization population and map each parameter combination of the joint optimization variable set to a search individual; Use the quantum-behaved particle swarm optimization algorithm to jointly optimize the path and servo strategy. Calculate the particle fitness function in each generation and iteratively update the particle state based on the historical optimal solution and the current search direction; In each round of iteration, comprehensively evaluate the search individuals according to multi-dimensional indicators such as task execution time limit, path continuity, smoothness of behavior control, and unit energy consumption level to generate an optimal energy consumption control scheme; Map the optimal energy consumption control scheme to a low-energy consumption control instruction set, which includes the target task sequence, servo joint angle scheduling curve, speed constraint parameters, and state feedback frequency.

7. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 6, characterized in that, The formula of the quantum-behaved particle swarm optimization algorithm is as follows: ； Among them, represents the optimal control parameter solution of the pet robot within the current task cycle; represents the historical optimal solution position of the particle in the k-th generation; represents the dynamic step size coefficient, which is jointly set according to the task urgency, its own power level, and the energy consumption evaluation value; represents the global optimal solution position in the current iteration, which is obtained by combining the historical behavior map and the energy consumption index evaluation; represents the non-linear convergence adjustment factor; represents the random perturbation variable.

8. An intelligent pet robot energy consumption optimization and automatic charging control system according to claim 1, characterized in that, The operation process of the automatic charging scheduling module includes the following steps: Collect the current battery percentage, remaining task time, spatial coordinates of the robot, and the reachability information between the robot and surrounding charging piles. Construct a heterogeneous directed graph with the robot node and the charging pile node as vertices. Initialize and model the edge weights of the heterogeneous directed graph by combining spatial distance, energy consumption estimation value, and path risk coefficient; Based on the graph attention network, train and model the heterogeneous directed graph. Use the attention mechanism to perform weighted learning on the battery level, task urgency, and path cost between nodes respectively, adaptively adjust the preference degree of each robot for candidate charging pile nodes, and output the path decision attention matrix; Combine the path decision attention matrix to construct a path selection function based on the remaining task time and navigation energy consumption, dynamically generate a set of charging candidate paths, calculate the scheduling priority scores of each path, and select the optimal charging scheduling path with the minimum path cost and meeting the time window constraint; According to the optimal charging scheduling path, generate a navigation control instruction set to drive the pet robot to autonomously navigate to the specified charging pile for charging tasks; the navigation control instruction set includes a sequence of path points, steering angle control, obstacle avoidance trigger conditions, and a remaining battery threshold trigger mechanism.

9. The intelligent pet robot energy consumption optimization and automatic charging control system according to claim 8, characterized in that The formula of the path selection function is as follows: ； Among them, represents the scheduling priority score of the candidate charging path r; represents the energy consumption per unit distance of the path r; represents the Euclidean path length between the current spatial coordinates of the robot and the candidate charging pile; represents the maximum navigation distance that the remaining battery power of the current robot can support; represents the estimated risk cost coefficient of the path r; represents the ratio of the path time consumption to the remaining time window of the task; represents the preference weight of the pet robot for each key node in the path; and represents the cost factor weight coefficient.

10. The intelligent pet robot energy consumption optimization and automatic charging control system according to claim 1, characterized in that The battery health control module specifically has a hierarchical health warning mechanism. According to the battery life prediction result and the health threshold range, the battery state is divided into four levels: normal, mild degradation, critical deterioration, and severe failure.

Images