Mworks-based autonomous mobile robot path planning system

Abstract

The invention discloses an autonomous mobile robot path planning system based on Mworks, and relates to the technical field of autonomous mobile robots, and the system comprises an environment sensing data collection module which is used for starting a robot to collect the distance of the surrounding environment and the multi-dimensional data information of images in real time through a laser radar, a camera and an ultrasonic sensor which are carried by the robot, by constructing the Mworks-based autonomous mobile robot path planning system, efficient fusion and utilization of multi-source data are realized, the system can collect data of various sensors such as a laser radar, a camera and an ultrasonic sensor in real time, and a high-precision environment map is constructed through preprocessing, standardization and data fusion algorithms, so that the path planning system is suitable for the autonomous mobile robot. The multi-sensor data fusion method not only improves the accuracy and reliability of environmental perception, but also provides a solid foundation for subsequent path planning and task allocation.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous mobile robots, and specifically to an autonomous mobile robot path planning system based on Mworks. Background Art

[0002] With the rapid development of technology, autonomous mobile robots are increasingly widely used in many fields, such as industrial manufacturing, logistics distribution, and smart home. These robots can autonomously complete various complex tasks without human intervention, greatly improving work efficiency and automation levels. To achieve autonomous movement, path planning is one of the key technologies. The path planning system needs to be able to perceive the surrounding environment in real time, accurately construct an environmental map, and plan the optimal path according to task requirements. Currently, the data fusion technology based on multiple sensors has become a research hotspot in the field of autonomous mobile robot path planning. This technology aims to improve the accuracy and reliability of environmental perception by fusing data from multiple sensors, so as to provide more accurate information support for path planning.

[0003] With the rapid development of technology, autonomous mobile robots are increasingly widely used in many fields, such as industrial manufacturing, logistics distribution, and smart home. These robots can autonomously complete various complex tasks without human intervention, greatly improving work efficiency and automation levels. To achieve autonomous movement, path planning is one of the key technologies. The path planning system needs to be able to perceive the surrounding environment in real time, accurately construct an environmental map, and plan the optimal path according to task requirements. Currently, the data fusion technology based on multiple sensors has become a research hotspot in the field of autonomous mobile robot path planning. This technology aims to improve the accuracy and reliability of environmental perception by fusing data from multiple sensors, so as to provide more accurate information support for path planning.

[0004] Therefore, developing an autonomous mobile robot path planning system based on Mworks will significantly improve the path planning ability and task execution efficiency of autonomous mobile robots. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the prior art, and provide an autonomous mobile robot path planning system based on Mworks. This system integrates multiple modules such as environmental perception data acquisition, multi-source data fusion modeling, task and resource allocation, path planning decision-making, and execution and monitoring feedback. Through high-precision environmental map construction and intelligent task assignment and path planning, this system significantly improves the navigation ability and task execution efficiency of autonomous mobile robots in complex environments. At the same time, the real-time monitoring and feedback mechanism of the system ensures that the robot can quickly respond when encountering abnormal situations and re-plan the path, thus ensuring the smooth completion of the task.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: An autonomous mobile robot path planning system based on Mworks, which system includes:

[0007] Environmental perception data acquisition module: The robot is started to collect multi-dimensional data information of the distance and image of the surrounding environment in real time through the mounted lidar, camera, and ultrasonic sensor, and the collected data is transmitted to the multi-source data fusion and modeling module;

[0008] Multi-source data fusion and modeling module: Receiving multi-sensor data transmitted from the environmental perception data acquisition module, preprocessing and standardizing different types of sensor data through the Mworks platform, and using a data fusion algorithm to fuse the preprocessed multi-sensor data to construct a high-precision environmental map;

[0009] Task and resource allocation module: According to the received specific task information, combined with the environmental map generated by the multi-source data fusion and modeling module, the data of the current position and capabilities of the robot, and the task priority, a task allocation model established through the Mworks platform allocates tasks to the robot and transmits relevant information to the path planning decision module;

[0010] Path planning decision module: According to the tasks allocated by the task and resource allocation module and the environmental map provided by the multi-source data fusion and modeling module, creating a path planning model through the Mworks platform based on the safety and length of the path to plan the optimal path from the starting point to the target point for each robot, and transmitting the planned path information to the execution and monitoring module during the planning process;

[0011] Execution and monitoring feedback module: The robot executes tasks according to the path planned by the path planning decision module, and at the same time monitors the component status of the robot and changes in the surrounding environment in real time. Once it detects that a robot component fails or the environment changes abnormally, it immediately feeds back the information to the multi-source data fusion and modeling module and the path planning decision module to re-plan the path.

[0012] Furthermore, the data collected by each sensor in the environmental perception data acquisition module is as follows: The lidar collects the distance information between surrounding objects and the robot, the camera collects the visual image information of the surrounding environment, and the ultrasonic wave obtains the object distance within a short distance range.

[0013] Even further, in the multi-source data fusion and modeling module, a data fusion algorithm is used to fuse the preprocessed multi-sensor data, and the formula is: where M(t) represents the environmental map data generated after fusion, ω L 、ω C 、ω Uare the fusion coefficients of lidar, camera, and ultrasonic sensor data, respectively, and ω L + ω C + ω U = 1, and are dynamically adjusted according to different scenarios and tasks. D L (t) is the distance data collected by the lidar, is the vectorized representation of the image data collected by the camera, D U (t) is the ultrasonic data collected at close range.

[0014] Furthermore, in the multi-source data fusion modeling module, the calculation formula for the distance data D L (t) collected by the lidar is: The calculation formula for the image data I C (t) obtained by the camera is: The calculation formula for the ultrasonic data D U (t) collected at close range is: where are the basic perception data of the lidar, camera, and ultrasonic sensor, respectively, are the weight coefficients of the corresponding data, n L 、n C 、n U are the quantities of various types of data, respectively.

[0015] Furthermore, in the task and resource allocation module, for the construction of the task allocation model, let the task priority be P T , the distance between the robot and the task target be D R-T , the remaining resources of the robot be R R , and the task allocation index be A T-R , and the calculation formula is: where λ is the distance influence coefficient, and the task with the largest task allocation index is assigned to the robot.

[0016] Furthermore, in the path planning and decision-making module, for the creation of the path planning model, let the safety of the path be S P , the length of the distance be L P , the traffic efficiency be E P , and the path evaluation value be V P , and the calculation formula is: where μ, ν, and ξ are the weight indices of each factor, which are determined by analyzing historical data.

[0017] Furthermore, in the execution and monitoring feedback module, for the determination of environmental changes, let the environmental change index be C E , and the calculation formula is: where m is the number of environmental map feature points used for comparison, f jis the status or attribute information of the j-th feature point in the current environment map, f jinit is the status or attribute information of the j-th feature point in the initial environment map, diff(f j , f jinit ) is a custom difference value function used to measure the difference degree between the information of two feature points. When the information of two feature points is exactly the same, diff(f j , f jinit ) = 0. The greater the difference, the larger the function value. When C E > C Eth , it indicates that the environment has an anomaly; C E < C Eth , it indicates that the environment has no anomaly, where C Eth is a pre-set environment anomaly threshold.

[0018] Furthermore, for the calculation of the environment change threshold C Eth in the execution and monitoring feedback module, it is assumed that in the past period of time, the robot has recorded the changes in the environment map for M times. Each time a change occurs, an environment change index is calculated, l = 1, 2,..., M. Calculate the mean μ CE and standard deviation σ CE of the historical environment change indices. The formula is: The formula for calculating the threshold C Eth of the environment change index is: C Eth = μ CE + α1·σ CE , where α1 is an adjustment coefficient, and the value range is α1 > 1, which is determined according to the adaptability and stability requirements of the robot system to environmental changes.

[0019] Furthermore, for the determination of the status of the robot components in the execution and monitoring feedback module, the calculation formula is: where n represents the number of key components used for calculation, w i is the weight coefficient of the i-th component, and the value range is between [0, 1], and is the current status parameter value of the i-th component, is the reference value of the normal status parameter of the i-th component. When it indicates that the robot component has a fault; it indicates that the robot component has no fault, where is the robot component fault threshold.

[0020] Furthermore, for the setting of the fault threshold in the execution and monitoring feedback module, it is assumed that when the robot was operating normally in the past period of time, N groups of status parameter values Calculate the mean value μ of the state parameters of each component during normal operation 1i and the standard deviation σ 1i , and the formula is: Comprehensively consider all key components and calculate the failure threshold of the robot components by weighted average The calculation formula is: where n is the number of key components, w i is the weight coefficient of the i-th component, and its value is determined according to the importance of the component to the overall operation of the robot. α2 is an adjustment coefficient, which is adjusted according to actual needs and the sensitivity to fault warning, and the value range is α2≥0.

[0021] Compared with the prior art, the path planning system of the autonomous mobile robot based on Mworks has the following beneficial effects:

[0022] First, by constructing a path planning system of an autonomous mobile robot based on Mworks, the present invention realizes the efficient fusion and utilization of multi-source data. The system can collect data of multiple sensors such as lidar, camera, and ultrasonic sensor in real time, and through preprocessing, standardization, and data fusion algorithms, construct a high-precision environmental map. This method of multi-sensor data fusion not only improves the accuracy and reliability of environmental perception, but also provides a solid foundation for subsequent path planning and task allocation. In addition, the system can also allocate the optimal task for the robot through the task allocation model according to specific task information and the current position and ability data of the robot, thereby realizing the reasonable configuration and efficient utilization of resources, and significantly improving the adaptability and intelligent level of the autonomous mobile robot in complex environments.

[0023] Second, by introducing methods for judging environmental changes and determining the state of robot components in the execution and monitoring feedback module, the present invention further enhances the stability and reliability of the system. By comparing the feature point information of the current environmental map and the initial environmental map, the system can timely detect environmental anomalies and re-plan the path, thereby avoiding path errors or failures of the robot caused by environmental changes. At the same time, the system can also monitor the state parameters of the key components of the robot in real time and give fault warnings according to the preset fault threshold, which not only improves the safety and stability of the autonomous mobile robot during operation, but also provides strong support for the maintenance and repair of the robot. Through continuous optimization and upgrading, this system is expected to be widely applied and promoted in more fields, injecting new vitality into the intelligent development of autonomous mobile robots.

[0024] Other advantages, objects, and features of the present invention will be set forth in part in the following description, and in part will be obvious to those skilled in the art from a study of the following, or may be learned from practice of the invention. Brief Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0026] Figure 1 It is a flowchart of an autonomous mobile robot path planning system based on Mworks;

[0027] Figure 2 It is a framework diagram of an autonomous mobile robot path planning system based on Mworks. Detailed Embodiments

[0028] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific embodiments, structures, features, and effects according to the present invention.

[0029] Embodiment 1:

[0030] Warehouse logistics scenario.

[0031] In a modern large-scale warehouse center, thousands of goods need to be transported and sorted every day, and the autonomous mobile robot based on Mworks plays a crucial role.

[0032] Environmental perception data collection: In the early morning, when the warehouse starts its operation, an autonomous mobile robot is awakened and prepares to execute the task of transporting goods. The lidar on the top of the robot starts to rotate at high speed, continuously emitting laser beams and receiving reflected signals to accurately measure the distance information between itself and the surrounding shelves, stacks of goods, and other obstacles. These distance data are like building a "spatial skeleton" for the robot, giving it a preliminary understanding of the general outline of the surrounding environment. At the same time, the high-definition camera mounted on the front end of the robot also starts to work. It quickly captures the visual images of the surrounding environment. Through advanced image recognition algorithms, the robot can clearly distinguish the identification of different goods, the numbers of the shelves, and various warning signs, providing key information for accurately finding the target goods later. In addition, the ultrasonic sensors distributed around the robot are constantly monitoring the objects within a short distance. Once another robot, staff, or temporarily placed items approach, the ultrasonic sensors will immediately respond and collect the corresponding distance data, effectively avoiding the occurrence of collision accidents. These multi-dimensional data collected from the lidar, camera, and ultrasonic sensors are transmitted in real time to the multi-source data fusion and modeling module through high-speed data transmission lines.

[0033] Multi-source data fusion and modeling: After receiving a large amount of data from the environmental perception data collection module, the multi-source data fusion and modeling module, with the powerful data processing ability of the Mworks platform, begins to preprocess and standardize different types of sensor data. This step is like organizing information from different places and in different formats into a unified "language" for subsequent fusion analysis. During the data fusion process, the system will dynamically adjust the fusion coefficients of the lidar, camera, and ultrasonic sensor data according to the actual situation of the current warehousing environment. For example, in areas with complex shelf layouts and dense goods placements, the accuracy and importance of the lidar data are higher, and its fusion coefficient will increase accordingly; while in areas with higher requirements for identifying goods labels and types, the fusion coefficient of the camera data will be increased. Through this dynamic adjustment, the preprocessed multi-sensor data are deeply fused using data fusion algorithms. The formula is: where M(t) represents the environmental map data generated after fusion; ω L 、ω C 、ω U are the fusion coefficients of the lidar, camera, and ultrasonic sensor data respectively, and ω L +ω C +ω U =1, and are dynamically adjusted according to different scenarios and tasks. D L (t) is the distance data collected by the lidar, is the vectorized representation of the image data collected by the camera, D U(t) uses ultrasonic waves to collect short - range data and constructs a high - precision warehouse environment map. This map not only accurately presents the positions of the shelves, the widths and directions of the aisles, but also marks the storage locations of the goods and the key information of the handling priorities, providing an accurate "navigation map" for the subsequent actions of the robot.

[0034] Task and resource allocation: The warehouse management system issues an urgent task to move a batch of high - priority electronic products from Area A to the shipping point in Area B of the warehouse. After receiving this task information, the task and resource allocation module quickly combines the environment map generated by the multi - source data fusion modeling module, the current position of the robot, and its own ability data (such as maximum load, handling speed) to start task allocation. First, the module determines the task priority according to the urgency of the task and the importance of the goods. For this batch of electronic products, due to their strong timeliness, they are given a high priority. Then, it calculates the distance between the robot and the task target, and at the same time evaluates the remaining resources of the robot, such as battery power and the number of available handling times. Then, through the task allocation model established by Mworks, let the task priority be P T , the distance between the robot and the task target be D R-T , the remaining resources of the robot be R R , construct the task allocation index as A T-R , and the calculation formula is: where λ is the distance influence coefficient. The task with the largest task allocation index is assigned to the robot. After a series of complex calculations and comparisons, the system finds that among the robots that can execute tasks, the current robot has the largest task allocation index and is most suitable for executing this task. Therefore, the task and resource allocation module assigns this task to this robot and passes the detailed information of the task, including the location of the target goods, the handling destination, and the task priority, to the path planning and decision - making module.

[0035] Path planning and decision - making: After receiving the task information and the environment map from the task and resource allocation module, the path planning and decision - making module starts to plan the optimal path for the robot. During the planning process, the module fully considers key factors such as the safety of the path, the length of the distance, and the passing efficiency. Since there are frequent movements of goods handling vehicles and other robots in the warehouse, the safety of the path is of crucial importance. The robot cannot collide with other objects or enter restricted areas. At the same time, the length of the distance directly affects the handling efficiency, and the shortest path should be selected as much as possible. In addition, the passing efficiency cannot be ignored, such as avoiding congested aisles and choosing unobstructed routes. Through the path planning model created by Mworks, let the safety of the path be S P , the length of the distance be L P , the passing efficiency be E P , the path evaluation value be VP , the calculation formula is: where μ, ν, and ξ are the weight indices of each factor, determined by historical data analysis. Let the safety of the path be S P , the length of the distance be L P , the traffic efficiency be E P , the path evaluation value be V P , the calculation formula is: where μ, ν, and ξ are the weight indices of each factor, determined by historical data analysis. After repeated calculations and comparisons of multiple path plans, an optimal path is finally planned for the robot, starting from the current position, along a spacious passage, avoiding other obstacles and congested areas, first reaching the pick-up point in Area A, and then quickly going to the delivery point in Area B. During the planning process, every time a section of the path is planned, the path planning decision module will timely transmit the planned path information to the execution and monitoring module.

[0036] Execution and monitoring feedback: After receiving the path information transmitted by the path planning decision module, the robot immediately starts to perform the handling task according to the planned path. During the movement, the execution and monitoring feedback module of the robot will continuously monitor the status of its own components and the changes in the surrounding environment. On the one hand, it continuously monitors the status of the key components of the robot through a specific calculation formula. The formula is: where n represents the number of key components used for calculation, w i is the weight coefficient of the i-th component, with a value range between [0, 1], and is the current status parameter value of the i-th component, is the reference value of the normal status parameter of the i-th component. For example, the rotation speed and temperature of the motor, the wear degree of the wheels, and the battery power. If the status parameter of a certain component deviates from the normal range, such as too high motor temperature or serious wheel wear, the execution and monitoring feedback module will immediately calculate the component status value. When determines that a fault has occurred in the robot component. At this time, the execution and monitoring feedback module will quickly feedback the fault information to the multi-source data fusion modeling module and the path planning decision module. The multi-source data fusion modeling module will re-evaluate the environment according to the new situation, and the path planning decision module will re-plan the path to guide the robot to the maintenance area or pause the task and wait for the maintenance personnel to handle it. On the other hand, the execution and monitoring feedback module will also monitor the changes in the surrounding environment through a formula. The formula is: where m is the number of environmental map feature points used for comparison, f j is the status or attribute information of the j-th feature point in the current environmental map, f jinit is the status or attribute information of the j-th feature point in the initial environmental map, diff(f j , f jinit) is a custom differential value function used to measure the degree of difference between the information of two feature points. When the information of the two feature points is exactly the same, diff(f j , f jinit ) = 0; the greater the difference, the greater the function value. For example, when warehouse workers temporarily adjust the position of the shelves, or new goods are stacked on the passage, these changes will cause the state of the feature points in the environmental map to change. The execution and monitoring feedback module will calculate the environmental change index in a timely manner. When C E > C Eth , it indicates that the environment is abnormal. Similarly, the execution and monitoring feedback module will feed back the environmental abnormality information to the multi-source data fusion modeling module and the path planning decision-making module to re-plan the path to ensure that the robot can complete the handling task safely and efficiently.

[0037] In summary, in the warehousing and logistics scenario, the autonomous mobile robot path planning system based on Mworks demonstrates efficient and intelligent characteristics. It obtains multi-dimensional information through the environmental perception data acquisition module, and the multi-source data fusion modeling module constructs an accurate environmental map, providing a solid foundation for task allocation and path planning. The task and resource allocation module reasonably allocates tasks, and the path planning decision-making module plans the optimal path that takes into account safety, distance, and efficiency. The execution and monitoring feedback module monitors in real time. Once component failures or environmental abnormalities occur, it quickly feedbacks and re-plans the path to ensure accurate and efficient cargo handling, improve the overall operation efficiency of warehousing and logistics, reduce labor costs, and has significant application value.

[0038] Embodiment 2:

[0039] Indoor cleaning scenario.

[0040] In a modern office building, after work every day, the cleaning robot will automatically start working to provide a clean space for the next day's office environment.

[0041] Environmental perception data collection: In the evening, after a busy day in the office building, the cleaning robot slowly drives out of the charging base and prepares to perform cleaning tasks. The lidar installed on the robot starts to scan the indoor environment omnidirectionally, quickly collecting the distance data between the surrounding desks, filing cabinets, partition obstacles and itself. These data are continuously updated to build a real-time indoor space model for the robot, enabling it to clearly know which areas are passable and which areas need to be avoided. At the same time, the camera at the front end of the robot is constantly taking pictures of the surrounding environment. With the help of advanced image recognition technology, the robot can accurately identify stains, garbage on the ground, and different types of furniture, providing a basis for targeted cleaning in the follow-up. In addition, the ultrasonic sensor is also closely monitoring objects within a short distance. Once it detects an object approaching suddenly, such as an item left by a staff member or a temporarily placed cleaning tool, the ultrasonic sensor will promptly collect the distance data and issue a warning to prevent the robot from colliding. These multi-source data collected by the lidar, camera and ultrasonic sensor are transmitted in real time to the multi-source data fusion and modeling module through a stable data transmission link.

[0042] Multi-source data fusion and modeling: After receiving the data transmitted from the environmental perception data collection module, the multi-source data fusion and modeling module uses the Mworks platform to preprocess and standardize the data. According to the environmental characteristics of the office building, the system will dynamically adjust the fusion coefficients of the lidar, camera, and ultrasonic sensor data according to the actual situation. For example, in the office area with dense desks and chairs, the lidar data is more important for determining the passage path, and its fusion coefficient will be increased accordingly; while in the public area with more stains on the ground, the camera data is more critical for identifying cleaning targets, and its fusion coefficient will be increased. The preprocessed multi-sensor data are fused through a data fusion algorithm, and the formula is: Build a high-precision indoor environment map, which details the layout of each indoor area, including the location of furniture, the direction of passageways, and the location of trash cans, providing accurate basic data for the cleaning robot to plan the cleaning path.

[0043] Task and resource allocation: The cleaning management system has issued a task to carry out comprehensive cleaning of the entire office floor. After receiving the task information, the task and resource allocation module combines the environmental map generated by the multi-source data fusion modeling module, the current position of the robot, and the remaining power resource data to start task allocation. First, the cleaning task priority of each area is determined based on the cleaning difficulty, area size, and time interval of the last cleaning in different areas. Public areas with frequent personnel activities and more stains, such as tea rooms and corridors, will be given a higher priority; while for some relatively tidy independent offices, the priority is relatively low. Then, the distance between the robot and the target position of each cleaning area is calculated, while considering the resource situation of the robot's remaining power. Then, through the task allocation model established by Mworks, the task priority is set as P T , the distance between the robot and the task target is D R-T , the remaining resources of the robot are R R , the construction task allocation index is A T-R , the calculation formula is: After comprehensive calculation and comparison, the task and resource allocation module rationally divides the cleaning area among different robots, ensuring that each robot can efficiently complete the cleaning tasks in its area of responsibility. For the current cleaning robot, the task and resource allocation module assigns it the cleaning tasks of several adjacent offices and a corridor, and passes the task details, including the location and priority of the cleaning area, to the path planning decision module.

[0044] Path planning decision: After receiving the task information and environmental map from the task and resource allocation module, the path planning decision module begins to plan the cleaning path for the cleaning robot. During the planning process, the module will comprehensively consider the safety of the path, the length of the distance, and the cleaning efficiency. Since there are various wires and socket facilities in the office, the safety of the path must be guaranteed. The robot cannot collide with these facilities to avoid damage or safety accidents. At the same time, in order to improve the cleaning efficiency, it is necessary to plan the shortest path as much as possible to reduce the robot's movement time between different areas. In addition, cleaning efficiency is also an important consideration. For example, it is necessary to arrange the cleaning order reasonably to avoid repeated cleaning of the same area. Through the path planning model created by Mworks, these factors are combined and the evaluation value of different paths is calculated using the formula: After in-depth analysis and comparison of multiple path plans, an optimal path was finally planned for the cleaning robot, starting from the current position, cleaning each office in turn in the order of difficult first and then easy, and finally cleaning the corridor. During the planning process, the path planning decision module will transmit the planned path information to the execution and monitoring module in real time.

[0045] Execution and Monitoring Feedback: The cleaning robot starts to perform the cleaning task according to the path planned by the path planning and decision-making module. During the cleaning process, the execution and monitoring feedback module will continuously monitor the status of the robot's own components and the changes in the surrounding environment. On the one hand, it will continuously monitor the status of the robot's key components through specific calculation formulas. The formula is: For example, the wear degree of the cleaning brush, the working status of the vacuum motor, and the battery power. If the status parameters of a certain component are abnormal, such as the serious wear of the cleaning brush affecting the cleaning effect, or the suction of the vacuum motor decreasing, the execution and monitoring feedback module will immediately calculate the component status value. When indicates that a fault has occurred in the robot component. At this time, the execution and monitoring feedback module will quickly feedback the fault information to the multi-source data fusion and modeling module and the path planning and decision-making module. The multi-source data fusion and modeling module will re-evaluate the environment, and the path planning and decision-making module will re-plan the path to guide the robot to the maintenance area or pause the task waiting for the maintenance personnel to handle. On the other hand, the execution and monitoring feedback module will also monitor the changes in the surrounding environment through a formula. The formula is: For example, when a staff member returns to the office to pick up something after work and moves the positions of the desks and chairs, or new garbage is discarded on the ground, these changes will cause the status of the feature points on the environmental map to change. The execution and monitoring feedback module will calculate the environmental change index in a timely manner. When C E > C Eth , it indicates that the environment has an abnormality. Similarly, the execution and monitoring feedback module will feedback the environmental abnormality information to the multi-source data fusion and modeling module and the path planning and decision-making module to re-plan the path to ensure that the cleaning task can be completed efficiently and safely.

[0046] To sum up, in the indoor cleaning scenario, the system gives full play to the advantages of autonomous cleaning. The environmental perception data acquisition module helps the robot comprehensively understand the indoor environment. The multi-source data fusion and modeling module generates a detailed environmental map for subsequent task arrangements. The task and resource allocation module allocates cleaning tasks according to the actual situation. The path planned by the path planning and decision-making module is efficient and safe. The execution and monitoring feedback module keeps real-time control of the robot components and environmental changes, promptly feedbacks in case of abnormalities, and re-plans the path to ensure that the cleaning work is uninterrupted and without omission, providing a stable and high-quality cleaning service for the indoor environment, improving the cleaning effect and user experience, and promoting the intelligent development of the indoor cleaning industry.

[0047] The above are only the preferred embodiments of the present invention and do not impose any formal limitations on the present invention. Although the present invention has been disclosed above in its preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments by using the disclosed technical content without departing from the technical solution of the present invention. However, as long as it does not depart from the technical solution content of the present invention, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An autonomous mobile robot path planning system based on Mworks, characterized in that, The system includes: Environmental perception data acquisition module: The robot is activated to collect multi-dimensional data information of distances and images of the surrounding environment in real time through the mounted lidar, camera, and ultrasonic sensor, and transmits the collected data to the multi-source data fusion and modeling module; Multi-source data fusion and modeling module: Receives multi-sensor data transmitted from the environmental perception data acquisition module, preprocesses and standardizes different types of sensor data through the Mworks platform, and fuses the preprocessed multi-sensor data using a data fusion algorithm to construct a high-precision environmental map; Task and resource allocation module: Based on the received specific task information, combined with the environmental map generated by the multi-source data fusion and modeling module, the data of the robot's current position, capabilities, and task priorities, uses the task allocation model established through the Mworks platform to allocate tasks to the robot and transmits relevant information to the path planning and decision-making module; Path planning and decision-making module: According to the tasks allocated by the task and resource allocation module and the environmental map provided by the multi-source data fusion and modeling module, creates a path planning model through the Mworks platform based on the safety and length of the path to plan the optimal path from the starting point to the target point for each robot. During the planning process, transmits the planned path information to the execution and monitoring module; Execution and monitoring feedback module: The robot executes tasks according to the path planned by the path planning and decision-making module, and simultaneously monitors the component status of the robot and changes in the surrounding environment in real time. Once it detects a failure of the robot components or an abnormality in the environment, immediately feeds back the information to the multi-source data fusion and modeling module and the path planning and decision-making module to re-plan the path.

2. The path planning system of the autonomous mobile robot based on Mworks according to claim 1, characterized in that, The data collected by each sensor in the environmental perception data acquisition module are: The lidar collects the distance information between the surrounding objects and the robot, the camera collects the visual image information of the surrounding environment, and the ultrasonic wave obtains the object distance within a short distance range.

3. The path planning system for an autonomous mobile robot based on Mworks according to claim 1, characterized in that, In the multi-source data fusion modeling module, a data fusion algorithm is used to fuse the preprocessed multi-sensor data. The formula is as follows: where M(t) represents the environmental map data generated after fusion, ω L , ω C , ω U are the fusion coefficients of the lidar, camera, and ultrasonic sensor data respectively, and ω L +ω C +ω U = 1, which is dynamically adjusted according to different scenarios and tasks. D L (t) is the distance data collected by the lidar, is the vectorized representation of the image data collected by the camera, and D U (t) is the short-range data collected by the ultrasonic wave.

4. The path planning system of the autonomous mobile robot based on Mworks according to claim 3, wherein In the multi-source data fusion modeling module, the lidar collects distance data D L (t) is calculated by the formula: The camera obtains image data I C (t) is calculated by the formula: The ultrasonic sensor collects short-distance data D U (t) is calculated by the formula: Where are the basic perception data of the lidar, camera, and ultrasonic sensor respectively, are the weight coefficients of the corresponding data, n L 、n C 、n U are the quantities of various types of data respectively.

5. The path planning system for an autonomous mobile robot based on Mworks according to claim 1, wherein The construction of the task allocation model in the task and resource allocation module, where the task priority is P T , the distance between the robot and the task target is D R-T , the remaining resources of the robot are R R , the constructed task allocation index is A T-R , and the calculation formula is: where λ is the distance influence coefficient, and the task with the largest task allocation index is assigned to the robot.

6. The path planning system of the autonomous mobile robot based on Mworks according to claim 1, wherein In the creation of the path planning model in the path planning decision module, let the safety of the path be S P , the length of the distance be L P , the passing efficiency be E P , the path evaluation value be V P , and the calculation formula is: where μ, ν, and ξ are the weight indices of each factor, which are determined through historical data analysis.

7. The path planning system for an autonomous mobile robot based on Mworks according to claim 1, wherein The determination of environmental changes in the execution and monitoring feedback module is made by setting the environmental change index as C E , and the calculation formula is as follows: where m is the number of environmental map feature points for comparison, and f j is the status or attribute information of the j-th feature point in the current environmental map, and f jinit is the status or attribute information of the j-th feature point in the initial environmental map. diff(f j , f jinit ) is a custom difference value function used to measure the difference degree between the information of two feature points. When the information of two feature points is exactly the same, diff(f j , f jinit ) = 0. The greater the difference, the larger the function value. When C E > C Eth , it indicates that the environment has an anomaly; when C E < C Eth , it indicates that the environment has no anomaly, where C Eth is a pre-set environmental anomaly threshold.

8. The path planning system for an autonomous mobile robot based on Mworks according to claim 7, wherein The environmental change threshold C in the execution and monitoring feedback module Eth is calculated as follows. Suppose that in the past period of time, the robot has recorded the changes in the environmental map for M times. Each time a change occurs, an environmental change index is calculated, where l = 1, 2, …, M. Calculate the mean μ CE and the standard deviation σ CE of the historical environmental change indices. The formulas are as follows: The formula for calculating the threshold C Eth of the environmental change index is: C Eth = μ CE + α1·σ CE , where α1 is an adjustment coefficient with a value range of α1 > 1, which is determined according to the adaptability and stability requirements of the robot system to environmental changes.

9. The path planning system for an autonomous mobile robot based on Mworks according to claim 1, wherein The determination of the status of the robot components in the execution and monitoring feedback module is calculated by the following formula: where n represents the number of key components used for calculation, w i is the weight coefficient of the i-th component, and its value range is between [0, 1], and x i is the current status parameter value of the i-th component, is the reference value of the normal status parameter of the i-th component. When it indicates that a fault has occurred in the robot component; it indicates that no fault has occurred in the robot component, where is the fault threshold of the robot component.

10. The path planning system for an autonomous mobile robot based on Mworks according to claim 9, wherein For the setting of the fault threshold in the execution and monitoring feedback module, when the robot operates normally in the past period, N groups of state parameter values are collected for each key component i k = 1, 2, …, N, calculate the mean value μ of the state parameters when each component operates normally 1i and the standard deviation σ 1i , and the formula is: Taking all key components into comprehensive consideration, calculate the fault threshold of the robot components by the weighted average method The calculation formula is: where n is the number of key components, w i is the weight coefficient of the i-th component, and its value is determined according to the importance of the component to the overall operation of the robot. α2 is an adjustment coefficient, which is adjusted according to actual requirements and the sensitivity to fault warning, and the value range is α2 ≥ 0

Images