Mobile navigation system for intelligent service robot

Abstract

The invention relates to the field of navigation systems, and particularly discloses a mobile navigation system for an intelligent service robot, which comprises a data acquisition and preprocessing subsystem, an environment analysis and weight setting subsystem, a data fusion and decision subsystem, an optimization and guarantee subsystem and an external interaction and support subsystem. The data acquisition and preprocessing subsystem comprises a sensor module and a calibration and preprocessing module; the environment analysis and weight setting subsystem comprises an environment perception and evaluation module; a sensor performance classification module; an initial weight setting module; a weight dynamic adjustment module; according to the mobile navigation system for the intelligent service robot, the multi-sensor fusion technology is adopted, the weight can be dynamically adjusted according to real-time environment information through a dynamic trust weight distribution mechanism based on environment complexity, higher adaptability and robustness are achieved, and the precision and reliability of robot navigation are improved.

Description

Technical Field

[0001] The present invention relates to the field of navigation systems, and in particular to a mobile navigation system for an intelligent service robot. Background Art

[0002] The mobile navigation system of the intelligent service robot is an advanced system that integrates multiple technologies and algorithms, and is designed to achieve autonomous movement and precise positioning of the robot in complex environments. The mobile navigation system of the intelligent service robot uses sensors (such as cameras, lidar, ultrasonic sensors, inertial measurement units, etc.) and advanced algorithms (such as SLAM, visual navigation, path planning, etc.) to perceive, understand and make decisions about the robot's surrounding environment, thereby guiding the robot to complete tasks safely and efficiently.

[0003] Since each sensor has its own advantages and disadvantages, some existing intelligent service robots' mobile navigation systems often use multi-sensor fusion technology. The core of multi-sensor fusion technology lies in the trust weight distribution mechanism. The traditional trust weight distribution mechanism often adopts a fixed weight distribution method and cannot adapt to environmental changes. For example, a robot navigates in indoor and outdoor environments. It uses visual sensors as one of the main navigation means. In indoor environments, the lighting conditions are relatively stable, and the visual sensors can accurately identify obstacles and paths. However, when the robot moves to an outdoor environment, the lighting conditions may change suddenly (such as entering a shadow area, encountering strong light, etc.), which will cause the performance of the visual sensor to drop sharply or even fail to work properly. In the traditional fixed trust weight distribution mechanism, since the weight of the visual sensor is fixed, the robot cannot reduce its dependence on the visual sensor in time, which may lead to navigation errors or stagnation.

[0004] Therefore, it is necessary to propose a mobile navigation system for intelligent service robots to solve the above problems. Summary of the invention

[0005] The main purpose of the present invention is to provide a mobile navigation system for an intelligent service robot, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A mobile navigation system for an intelligent service robot, comprising a data acquisition and preprocessing subsystem, an environment analysis and weight setting subsystem, a data fusion and decision subsystem, an optimization and guarantee subsystem, and an external interaction and support subsystem;

[0008] The data acquisition and preprocessing subsystem includes a sensor module and a calibration and preprocessing module. The sensor module is used to collect environmental data around the robot, and the sensor module includes but is not limited to a laser radar, a camera, and an ultrasonic sensor. The calibration and preprocessing module is used to regularly calibrate the sensor and filter, denoise, and normalize the data collected by the sensor.

[0009] The environmental analysis and weight setting subsystem includes:

[0010] The environment perception and assessment module is used to analyze the number of obstacles in the environment, the degree of terrain change, the change in lighting conditions, the density of people or vehicles, and calculate the complexity score. The complexity formula is: C = w 1 ×N obs +w 2 ×D terrain +w 3 ×V light +w 4 ×D flow , where N obs is the number of obstacles, D terrain is the degree of terrain change, V light is the illumination change condition, D flow is the density of pedestrian or vehicle flow, w 1 、w 2 、w 3 、w 4 is the corresponding weight;

[0011] A sensor performance classification module is used to classify sensors into different performance categories based on their historical test data under different environmental conditions;

[0012] An initial weight setting module is used to set the initial weight for each sensor;

[0013] The weight dynamic adjustment module dynamically adjusts the trust weight of the sensor in real time based on the environmental complexity score and sensor performance classification. The sensor weight formula is:

[0014] W S =W init +ΔW, Where W init is the initial weight of the sensor, ΔW is the weight adjustment value, α and β are the adjustment coefficients, C threshold is the environmental complexity threshold, F s is the performance factor of the sensor under the current environmental conditions;

[0015] Performance evaluation and weight assignment module, used to assign trust weights to each sensor based on its historical performance, accuracy, and reliability;

[0016] The data fusion and decision-making subsystem includes:

[0017] A fusion algorithm module is used to fuse the processed data through an algorithm;

[0018] The navigation decision module is used to make navigation decisions based on the fusion results, taking into account map information, target setting and environmental conditions.

[0019] Preferably, the weight allocation formula in the performance evaluation and weight allocation module is: W = k 1 P+k 2 R+k 3 S, where k 1 , k 2 , k 3 It is a coefficient set according to importance, P is the accuracy of the sensor, R is the reliability of the sensor, and S is the historical performance score.

[0020] Preferably, when the sensor performance classification module classifies the performance categories of the sensor, the performance categories include but are not limited to complex illumination type, dense obstacle type, and stable terrain type.

[0021] Preferably, the initial weight setting module sets an initial weight for each sensor based on the general performance and accuracy of the sensor.

[0022] Preferably, the environmental data collected by the sensor module includes but is not limited to distance, image, and obstacle information.

[0023] Preferably, the calibration formula used by the calibration and preprocessing module to calibrate the sensor is: T=aM+b, where M is the measured value of the sensor, the true value is T, and a and b are coefficients determined through multiple calibration experiments;

[0024] The filtering is median filtering: for a series of sensor data x 1 ,x 2 ,…x n , the result y after median filtering i is the data value at the median position after sorting;

[0025] Denoising is mean denoising: for the original data x i , the denoised value Where m is the window size;

[0026] Normalization uses linear normalization: its formula is where x i is the original data value, x min and x max are the minimum and maximum values ​​of the data respectively.

[0027] Preferably, the algorithm used by the fusion algorithm module to fuse data is Kalman filtering, particle filtering, or a fusion method in deep learning.

[0028] Preferably, the optimization and guarantee subsystem includes:

[0029] The learning and optimization module is used to record the robot's navigation process and results, as well as sensor data, and use reinforcement learning algorithms to optimize weight adjustment strategies and navigation strategies;

[0030] The uncertainty processing module is used to continuously monitor sensor data and temporarily adjust the weights of related sensors when continuous anomalies occur;

[0031] A map building and updating module that creates and maintains a map of the robot’s working environment based on prior knowledge of sensor data;

[0032] The target setting module is used to specify the robot's target location or task;

[0033] The motion control module is used to convert navigation decisions into specific motor control instructions to control the robot's speed, steering and movement.

[0034] Preferably, the external interaction and support subsystem includes:

[0035] Communication module, used for data transmission and interaction between the robot and external devices or systems;

[0036] Safety monitoring module for real-time detection of possible dangerous situations;

[0037] The power management module is used to monitor the power status of the robot.

[0038] A navigation method for a mobile navigation system of an intelligent service robot comprises the following steps:

[0039] S1: The sensor module collects environmental data and transmits it to the calibration and preprocessing module;

[0040] S2: The calibrated and preprocessed data enters the environment perception and evaluation module, the sensor performance classification module, and the initial weight setting module respectively;

[0041] S3: The environment perception and assessment module calculates the environment complexity score, the sensor performance classification module determines the sensor performance category, and the initial weight setting module sets the initial weight for the sensor. All this information is input into the weight dynamic adjustment module.

[0042] S4: The weight dynamic adjustment module adjusts the sensor weight according to the input information, and then passes the result to the performance evaluation and weight allocation module;

[0043] S5: After performance evaluation and weight allocation, the sensor data is input into the fusion algorithm module for fusion processing;

[0044] S6: The fusion result enters the navigation decision module. At the same time, the map construction and update module provides map information, and the target setting module determines the target, which jointly support the formulation of navigation decisions;

[0045] S7: The output of the navigation decision module is passed to the learning and optimization module, the uncertainty processing module and the motion control module;

[0046] S8: The learning and optimization module optimizes the system strategy based on historical data, the uncertainty processing module handles abnormal sensor data, and the motion control module controls the actual movement of the robot;

[0047] S9: The communication module is responsible for data interaction with the outside world during the entire process, while the power management module and security monitoring module ensure the stable and safe operation of the system.

[0048] Compared with the prior art, the present invention provides a mobile navigation system for an intelligent service robot, which has the following beneficial effects:

[0049] The mobile navigation system for the intelligent service robot includes a variety of sensors through the set sensor module, adopts multi-sensor fusion technology and cooperates with the set environment analysis and weight setting subsystem to realize dynamic adjustment of weights. The dynamic trust weight allocation mechanism based on environmental complexity can dynamically adjust the weight according to real-time environmental information, and has stronger adaptability and robustness. Moreover, by comprehensively considering various environmental factors and sensor performance characteristics, the mechanism can more accurately evaluate the environmental complexity and sensor performance, so as to obtain more accurate trust weight allocation results, which helps to improve the accuracy and reliability of robot navigation. Moreover, the mechanism supports flexible combinations and weight adjustment methods of multiple sensors and multiple environmental factors, and can be customized and optimized according to actual needs, which helps to meet navigation needs in different application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the structure of the present invention;

[0051] Figure 2 It is a schematic diagram of the structure of the optimization and guarantee subsystem of the present invention;

[0052] Figure 3 is a schematic diagram of the structure of the external interaction and support subsystem of the present invention;

[0053] Figure 4 It is a specific system flow chart of the present invention;

[0054] Figure 5It is a flow chart of the navigation method of the present invention. DETAILED DESCRIPTION

[0055] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0056] like Figure 1-Figure 5 As shown, a mobile navigation system for an intelligent service robot includes a data acquisition and preprocessing subsystem, an environment analysis and weight setting subsystem, a data fusion and decision subsystem, an optimization and guarantee subsystem, and an external interaction and support subsystem;

[0057] The data acquisition and preprocessing subsystem includes a sensor module and a calibration and preprocessing module. The sensor module is used to collect environmental data around the robot in all directions, and the environmental data includes but is not limited to information such as distance, image, obstacle, etc., and the sensor module includes but is not limited to sensors such as laser radar, camera, and ultrasonic wave;

[0058] Furthermore, the calibration and preprocessing module is used to regularly calibrate the sensor to ensure the accuracy of the measurement, as well as filter, denoise and normalize the data collected by the sensor to improve the data quality:

[0059] Specifically, the calibration formula used by the calibration and preprocessing module to calibrate the sensor is: T = aM + b, where M is the measured value of the sensor, the true value is T, and a and b are coefficients determined through multiple calibration experiments;

[0060] The filtering is median filtering. For a series of sensor data x 1 ,x 2 ,…x n , the result y after median filtering i is the data value at the median position after sorting;

[0061] Denoising is mean denoising. For the original data x i , the denoised value Where m is the window size;

[0062] Normalization is linear normalization, and its formula is where x i is the original data value, x min and x max are the minimum and maximum values ​​of the data respectively.

[0063] Furthermore, the environmental analysis and weight setting subsystem includes:

[0064] The environment perception and assessment module is used to analyze the number of obstacles in the environment, the degree of terrain change, the change in lighting conditions, the density of people or vehicles, and calculate the complexity score. The complexity formula is: C = w 1 ×N obs +w 2 ×D terrain +w 3 ×V light +w 4 ×D flow , where N obs is the number of obstacles, D terrain is the degree of terrain change, V light is the illumination change condition, D flow is the density of pedestrian or vehicle flow, w 1 、w 2 、w 3 、w 4 is the corresponding weight;

[0065] Furthermore, the sensor performance classification module is used to classify the sensor into different performance categories according to the historical test data of the sensor under different environmental conditions, and the performance categories during classification include but are not limited to complex illumination type, dense obstacle type, and stable terrain type. The specific classification method is as follows:

[0066] 1. Collect historical test data of sensors under various environmental conditions, including environmental parameters such as light intensity, obstacle density, terrain stability, and the corresponding measurement data of the sensors.

[0067] 2. Extract features from the collected data that can characterize the performance of the sensor, such as measurement accuracy under different lighting conditions, detection ability for dense obstacles, stability in stable terrain, etc.

[0068] 3. Clarify the specific definitions and characteristic ranges of performance categories such as complex lighting type, dense obstacle type, and stable terrain type.

[0069] 4. You can use classification algorithms in machine learning, such as decision trees, support vector machines (SVM), K-nearest neighbors (KNN), etc.

[0070] 5. Train the selected classification algorithm using the extracted features and the defined classification criteria.

[0071] 6. Use the test data set to evaluate the trained model and calculate indicators such as accuracy, recall, and F1 value to determine the performance of the model.

[0072] 7. Adjust and optimize the model based on the evaluation results, such as adjusting algorithm parameters, increasing training data, improving feature extraction, etc.

[0073] An initial weight setting module is also provided, which sets the initial weight for each sensor based on the general performance and accuracy of the sensor. The general performance and accuracy usually refer to a series of characteristics and measurement accuracy exhibited by the sensor under typical working conditions, including the following aspects:

[0074] Stability: The stability of the sensor output, that is, the consistency and repeatability of the results when multiple measurements are taken under the same conditions.

[0075] Response Time: The time it takes for a sensor to react to a change in its environment and output a valid measurement.

[0076] Resolution: The smallest change a sensor can detect.

[0077] Working Range: The range of maximum and minimum values ​​that a sensor can effectively measure.

[0078] Anti-interference ability: The ability of the sensor to maintain normal operation and accurate measurement in the presence of other interference factors (such as electromagnetic interference, temperature changes, etc.).

[0079] The steps to set the initial weights can be as follows:

[0080] 1. Establish an evaluation index system: determine specific indicators for measuring the general performance and accuracy of the sensor, and assign a certain weight to each indicator. For example, stability has a weight of 30%, response time has a weight of 20%, and accuracy has a weight of 50%.

[0081] 2. Test and evaluate each sensor: Test the sensor under standard environmental conditions to obtain specific performance data on each evaluation indicator. For example, the stability evaluation score of sensor A is 80 points (full score 100), the response time is 10 milliseconds (the goal is within 5 milliseconds), and the accuracy evaluation score is 90 points.

[0082] 3. Calculate the comprehensive score: Calculate the comprehensive score of each sensor according to the predetermined indicator weights. Assume that the comprehensive score of sensor A above = 80 × 0.3 + 10 / 5 × 0.2 × 100 + 90 × 0.5 = 86 points.

[0083] 4. Normalized score: Normalize the comprehensive scores of all sensors to be between 0 and 1. Assuming that the comprehensive scores of all sensors are between 60 and 90, normalize them using the formula (sensor score - 60) / (90 - 60).

[0084] 5. Determine the initial weight: Determine the initial weight of each sensor based on the normalized score. Sensors with higher normalized scores receive higher initial weights, and vice versa.

[0085] Based on the above, this system is equipped with a weight dynamic adjustment module, which dynamically adjusts the trust weight of the sensor in real time according to the environmental complexity score and sensor performance classification. The sensor weight formula is:

[0086] W S =W init +ΔW, Where W init is the initial weight of the sensor, ΔW is the weight adjustment value, α and β are the adjustment coefficients, C threshold is the environmental complexity threshold, F s is the performance factor of the sensor under the current environmental conditions;

[0087] A performance evaluation and weight allocation module is also provided to allocate a trust weight to each sensor based on the sensor's historical performance, accuracy, and reliability. The weight allocation formula is: W = k 1 P+k 2 R+k 3 S, where k 1 , k 2 , k 3 It is a coefficient set according to importance, P is the accuracy of the sensor, R is the reliability of the sensor, and S is the historical performance score;

[0088] Furthermore, the data fusion and decision-making subsystem includes:

[0089] The fusion algorithm module is used to fuse the processed data through an algorithm. The algorithm used by the fusion algorithm module to fuse data is a Kalman filter, a particle filter, or a fusion method in deep learning. Here, a conventional algorithm is used, and the specific formula and process are not described in detail.

[0090] The navigation decision module is used to make navigation decisions based on the fusion results, taking into account map information, target setting and environmental conditions, such as planning the optimal path and selecting the appropriate action strategy.

[0091] Furthermore, the optimization and guarantee subsystem includes:

[0092] The learning and optimization module is used to record the robot's navigation process and results, as well as sensor data, and use reinforcement learning algorithms to optimize weight adjustment strategies and navigation strategies to improve system performance and adaptability;

[0093] The uncertainty processing module is used to continuously monitor sensor data. When continuous anomalies occur, the weights of related sensors are temporarily adjusted to reduce their impact on navigation decisions. The specific processing methods are as follows:

[0094] 1. Obtain sensor data in real time, including sensor measurement values, timestamps and other information.

[0095] 2. Set rules for abnormal judgment, such as multiple consecutive measurement values ​​exceeding the reasonable range, abnormally large rate of change of measurement values, etc.

[0096] Analyze the acquired data to determine whether there are continuous abnormalities.

[0097] 3. When continuous abnormalities in sensor data are detected, the weight adjustment mechanism is triggered.

[0098] 4. Determine the magnitude and method of weight adjustment, which can be calculated based on the severity and duration of the anomaly, for example, reducing the weight in a linear or nonlinear manner.

[0099] 5. Apply the calculated new weights to the relevant sensors.

[0100] 6. Observe the impact of weight adjustment on navigation decisions and evaluate whether the purpose of reducing the impact of abnormal sensors has been achieved.

[0101] 7. Continuously monitor the data of abnormal sensors, and gradually restore their weights to normal levels after the abnormal situation disappears and stabilizes for a period of time.

[0102] 8. Record abnormal conditions of sensor data, the process and results of weight adjustment, and provide data support for subsequent system optimization and fault diagnosis.

[0103] During the implementation process, it is necessary to optimize and adjust the anomaly judgment rules, weight adjustment strategies, etc. according to the actual system requirements and sensor characteristics to ensure the effectiveness and adaptability of the uncertainty processing module.

[0104] The map building and updating module creates and maintains a map of the robot’s working environment based on prior knowledge of sensor data, and updates the map information in real time to reflect changes in the environment;

[0105] The target setting module is used to specify the robot's target position or task, such as specifying specific coordinates, a specific operating area, or performing a specific task;

[0106] The motion control module is used to convert navigation decisions into specific motor control instructions to control the robot's speed, steering and movement to achieve smooth and accurate movement.

[0107] Furthermore, the external interaction and support subsystem includes:

[0108] Communication module, used for data transmission and interaction between the robot and external devices or systems;

[0109] Safety monitoring module, which is used to detect possible dangerous situations in real time, such as collision risk, system failure, overheating, etc. Once a danger is detected, it will immediately trigger emergency braking or sound an alarm to ensure the safety of the robot and the surrounding environment;

[0110] The power management module monitors the power status of the robot, including battery power, charging status, etc. It also optimizes power usage, performs low-power warnings and energy-saving management to ensure that the system runs stably with sufficient power.

[0111] A navigation method using the above navigation system comprises the following steps:

[0112] Step 1: The sensor module collects environmental data and transmits it to the calibration and preprocessing module;

[0113] Step 2: The calibrated and preprocessed data enters the environment perception and evaluation module, the sensor performance classification module, and the initial weight setting module respectively;

[0114] Step 3: The environment perception and assessment module calculates the environment complexity score, the sensor performance classification module determines the sensor performance category, and the initial weight setting module sets the initial weight for the sensor. All this information is input into the weight dynamic adjustment module.

[0115] Step 4: The weight dynamic adjustment module adjusts the sensor weights according to the input information, and then passes the results to the performance evaluation and weight allocation module;

[0116] Step 5: After performance evaluation and weight allocation, the sensor data is input into the fusion algorithm module for fusion processing;

[0117] Step 6: The fusion result enters the navigation decision module. At the same time, the map construction and update module provides map information, and the target setting module determines the target, which jointly supports the formulation of navigation decisions;

[0118] Step 7: The output of the navigation decision module is passed to the learning and optimization module, the uncertainty processing module and the motion control module;

[0119] Step 8: The learning and optimization module optimizes the system strategy based on historical data, the uncertainty processing module handles abnormal sensor data, and the motion control module controls the actual movement of the robot;

[0120] Step 9: The communication module is responsible for data interaction with the outside world during the entire process, while the power management module and security monitoring module ensure the stable and safe operation of the system.

[0121] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A mobile navigation system for an intelligent service robot, characterized in that: It includes data collection and preprocessing subsystem, environmental analysis and weight setting subsystem, data fusion and decision-making subsystem, optimization and guarantee subsystem, and external interaction and support subsystem; The data acquisition and preprocessing subsystem includes a sensor module and a calibration and preprocessing module. The sensor module is used to collect environmental data around the robot, and the sensor module includes but is not limited to a laser radar, a camera, and an ultrasonic sensor. The calibration and preprocessing module is used to regularly calibrate the sensor and filter, denoise, and normalize the data collected by the sensor. The environmental analysis and weight setting subsystem includes: The environmental perception and assessment module is used to analyze the number of obstacles in the environment, the degree of terrain change, the change in lighting conditions, the density of people or vehicles, and calculate the complexity score. The complexity formula is: C = w1 × N obs +w2×D terrain +w3×V light +w4×D flow , where N obs is the number of obstacles, D terrain is the degree of terrain change, V light is the illumination change condition, D flow is the density of pedestrian or vehicle flow, w1, w2, w3, w4 are the corresponding weights; A sensor performance classification module is used to classify sensors into different performance categories based on their historical test data under different environmental conditions; An initial weight setting module is used to set the initial weight for each sensor; The weight dynamic adjustment module dynamically adjusts the trust weight of the sensor in real time based on the environmental complexity score and sensor performance classification. The sensor weight formula is: Where W init is the initial weight of the sensor, ΔW is the weight adjustment value, α and β are the adjustment coefficients, C threshold is the environmental complexity threshold, F s is the performance factor of the sensor under the current environmental conditions; Performance evaluation and weight assignment module, used to assign trust weights to each sensor based on its historical performance, accuracy, and reliability; The data fusion and decision-making subsystem includes: A fusion algorithm module is used to fuse the processed data through an algorithm; The navigation decision module is used to make navigation decisions based on the fusion results, taking into account map information, target setting and environmental conditions.

2. A mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The weight distribution formula in the performance evaluation and weight distribution module is: W=k1P+k2R+k3S, where k1, k2, k3 are coefficients set according to importance, P is the accuracy of the sensor, R is the reliability of the sensor, and S is the historical performance score.

3. A mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The performance categories of the sensor performance classification module when classifying the performance categories of the sensor include but are not limited to complex illumination type, dense obstacle type, and stable terrain type.

4. The mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The initial weight setting module sets an initial weight for each sensor based on the general performance and accuracy of the sensor.

5. The mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The environmental data collected by the sensor module includes but is not limited to distance, image, and obstacle information.

6. A mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The calibration formula for calibrating the sensor by the calibration and preprocessing module is: T=aM+b, where M is the measured value of the sensor, the true value is T, and a and b are coefficients determined through multiple calibration experiments; The filtering is median filtering: for a series of sensor data x1, x2, ... x n , the result y after median filtering i is the data value at the median position after sorting; Denoising is mean denoising: for the original data x i , the denoised value Where m is the window size; Normalization is linear normalization: its formula is where x i is the original data value, x min and x max are the minimum and maximum values ​​of the data respectively.

7. A mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The algorithm used by the fusion algorithm module to fuse data is Kalman filtering, particle filtering, or a fusion method in deep learning.

8. The mobile navigation system for an intelligent service robot according to claim 1, characterized in that: The optimization and guarantee subsystem includes: The learning and optimization module is used to record the robot's navigation process and results, as well as sensor data, and use reinforcement learning algorithms to optimize weight adjustment strategies and navigation strategies; The uncertainty processing module is used to continuously monitor sensor data and temporarily adjust the weights of related sensors when continuous anomalies occur; A map building and updating module that creates and maintains a map of the robot’s working environment based on prior knowledge of sensor data; The target setting module is used to specify the robot's target location or task; The motion control module is used to convert navigation decisions into specific motor control instructions to control the robot's speed, steering and movement.

9. A mobile navigation system for an intelligent service robot according to claim 8, characterized in that: The external interaction and support subsystem includes: Communication module, used for data transmission and interaction between the robot and external devices or systems; Safety monitoring module for real-time detection of possible dangerous situations; The power management module is used to monitor the power status of the robot.

10. A navigation method for a mobile navigation system of an intelligent service robot according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1: The sensor module collects environmental data and transmits it to the calibration and preprocessing module; S2: The calibrated and preprocessed data enters the environment perception and evaluation module, the sensor performance classification module, and the initial weight setting module respectively; S3: The environment perception and assessment module calculates the environment complexity score, the sensor performance classification module determines the sensor performance category, and the initial weight setting module sets the initial weight for the sensor. All this information is input into the weight dynamic adjustment module. S4: The weight dynamic adjustment module adjusts the sensor weight according to the input information, and then passes the result to the performance evaluation and weight allocation module; S5: After performance evaluation and weight allocation, the sensor data is input into the fusion algorithm module for fusion processing; S6: The fusion result enters the navigation decision module. At the same time, the map construction and update module provides map information, and the target setting module determines the target, which jointly support the formulation of navigation decisions; S7: The output of the navigation decision module is passed to the learning and optimization module, the uncertainty processing module and the motion control module; S8: The learning and optimization module optimizes the system strategy based on historical data, the uncertainty processing module handles abnormal sensor data, and the motion control module controls the actual movement of the robot; S9: The communication module is responsible for data interaction with the outside world during the entire process, while the power management module and security monitoring module ensure the stable and safe operation of the system.

Images