An intelligent luggage obstacle avoidance method and system based on wireless sensors

Abstract

The present invention discloses an intelligent luggage obstacle avoidance method and system based on wireless sensors, which relates to the field of intelligent navigation technology. It includes inputting image data into a stacked hourglass network, using the obstacle area as a guiding clue, mapping the detection range of the ultra-wideband sensor to the corresponding ultra-wideband sector, using area data binding to bind the comprehensive time difference value with the obstacle area, encapsulating it as the focused time difference data, calculating the three-dimensional coordinate set of the obstacle and the direction of the obstacle; constructing a local grid map, using an exponential decay function to adjust the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate an optimal path; using the stacked hourglass network to mark the obstacle area as a guiding clue to reduce interference from irrelevant areas, constructing a local grid map and using an exponential decay function to adjust the pheromone value, binding the pheromone with the obstacle area to generate an optimized path, improving the obstacle avoidance ability and autonomous navigation performance of the intelligent luggage.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent navigation, and particularly to an intelligent luggage obstacle avoidance method and system based on wireless sensors. Background Art

[0002] As a convenient mobile tool, intelligent luggage is increasingly applied in daily life and logistics distribution fields. In the design of intelligent luggage, how to efficiently and safely avoid obstacles, especially in complex dynamic environments for path planning, has become an important challenge for its intelligence. With the rapid development of intelligent devices and robotics technologies, obstacle avoidance technologies based on wireless sensors have been widely applied in multiple fields, especially in the fields of intelligent navigation and autonomous mobile devices. The integration of wireless sensor technologies, such as ultra-wideband (UWB) sensors, ultrasonic sensors, and vision sensors, has become the key to achieving precise environmental perception and path planning.

[0003] Existing intelligent luggage obstacle avoidance technologies have some deficiencies. In scenarios with many obstacles and complex spaces, single sensor data may be difficult to provide sufficient accuracy and reliability, thus affecting the obstacle avoidance effect of the device. Existing path planning methods (such as DWA or artificial potential field method) rely on static weights or fixed thresholds in dynamic environments, and it is difficult to adapt to changes in the moving speed and distribution of obstacles, easily falling into local optima or frequently adjusting paths, which affects the navigation efficiency. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent luggage obstacle avoidance method and system based on wireless sensors, which solves the problem that existing path planning methods (such as DWA or artificial potential field method) rely on static weights or fixed thresholds in dynamic environments, are difficult to adapt to changes in the moving speed and distribution of obstacles, easily fall into local optima or frequently adjust paths, and affect the navigation efficiency.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent luggage obstacle avoidance method based on wireless sensors, which includes,

[0008] Collecting sensor data, including time difference, echo time, and image data;

[0009] Inputting the image data into a stacked hourglass network to mark the obstacle area, using the obstacle area as a guiding clue, mapping the detection range of the UWB sensor to the corresponding UWB sector, binding the comprehensive time difference value with the obstacle area using area data binding, encapsulating it into the focused time difference data, and calculating the three-dimensional coordinate set of the obstacle and the direction of the obstacle.

[0010] Construct a local grid map, adjust the pheromone value using an exponential decay function, combine the pheromone value, set weights using the analytic hierarchy process and perform weighted scoring to generate an optimal path;

[0011] Decompose the optimal path into small motion commands and execute them, monitor and adjust the execution results, and store and analyze the sensor data collected.

[0012] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: calculating the three-dimensional coordinate set of the obstacle and the direction of the obstacle includes:

[0013] The wireless sensors include ultra-wideband sensors, ultrasonic sensors, and cameras;

[0014] Use the acoustic distance formula to calculate the distance between the obstacle and the ultrasonic sensor, defined as the ultrasonic distance;

[0015] Input the image data into a stacked hourglass network, detect the feature area of the obstacle in the image through convolution and upsampling processing, use the semantic segmentation method to segment the feature area, and mark the obstacle area;

[0016] Use the obstacle area as a guiding clue, divide the detection range of the ultra-wideband sensor into sectors using the edge detection method, and map it to the corresponding ultra-wideband sector;

[0017] Use the weighted average method to calculate the comprehensive time difference value after weighted processing of all time difference data, use regional data binding to bind the comprehensive time difference value with the obstacle area, and encapsulate it as the focused time difference data;

[0018] Use the two-way ranging method to calculate the distance of the obstacle, defined as the time difference distance;

[0019] Use the maximum entropy threshold method to set the error threshold, use the absolute error method to calculate the error between the ultrasonic distance and the time difference distance. If the error is less than or equal to the error threshold, calculate the ultrasonic distance and the time difference distance as the final distance. If the error is greater than the error threshold, use the ultrasonic distance as the final distance;

[0020] Take the intelligent luggage as the origin, calculate the mean value of the ultra-wideband sector as the relative angle of the obstacle, and use the triangulation method to calculate the two-dimensional coordinates of the obstacle;

[0021] Use the Canny edge detection method to extract the contour of the obstacle, convert it to the two-dimensional map coordinates of the contour, take the two-dimensional coordinates of the obstacle as a reference, and use the parallax method to convert the two-dimensional map coordinates of the contour and the two-dimensional coordinates of the obstacle into three-dimensional coordinates to generate the three-dimensional coordinate set of the obstacle;

[0022] Calculate the circumscribed bounding box of the obstacle using the bounding box method, and extract the main axis direction using the principal component analysis method as the direction of the obstacle.

[0023] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: after the sensor data, preprocessing operations are first performed.

[0024] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: for constructing the local grid map, adjusting the pheromone value using the exponential decay function, combining the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate the optimal path, including:

[0025] Use the grid map method to place the intelligent luggage at the center of the map as the starting point and construct a local grid map.

[0026] Assign an initial pheromone value to each grid, set the initial value of all grids to "idle", and change the value of the grid corresponding to the three-dimensional coordinate set of the obstacle from "idle" to "occupied".

[0027] Calculate the center of the obstacle as a reference point using the coordinate averaging method, mark the grids in the sector as the influence area using the region expansion method, and adjust the pheromone using the exponential decay function to obtain the updated local grid map.

[0028] Calculate the current speed using the positioning algorithm.

[0029] Predict the end position of the path using the kinematic equation.

[0030] Generate candidate paths using the artificial potential field method, use the three-dimensional coordinate set of the obstacles in the local grid map to detect whether the grids corresponding to the candidate paths are marked as "occupied", and delete the overlapping "occupied" grids.

[0031] Select the remaining candidate paths, extract the pheromone corresponding to the end position of the path, calculate the path length using the Euclidean distance formula, calculate the minimum distance between the path and the obstacle using the closest point distance method, set the weight coefficient using the analytic hierarchy process, calculate the final score of the path using the weighted scoring method, and sort from high to low, and select the candidate path corresponding to the highest score as the optimal path.

[0032] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: decomposing the optimal path into small motion instructions and executing them, including:

[0033] Decompose the optimal path into small motion instructions using the path decomposition algorithm.

[0034] Use a wireless communication protocol to transmit small motion instructions to the intelligent luggage. The intelligent luggage receives the small motion instructions, converts the small motion instructions into control signals using a direct torque control algorithm, and executes them.

[0035] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: the monitoring and adjustment of the execution result includes:

[0036] Collect feedback position data and perform preprocessing, and set the target position using the YOLO deep learning algorithm;

[0037] Use the deviation method to calculate the difference between the target position and the preprocessed feedback position data;

[0038] Use the empirical rule to set the stop threshold, compare the difference with the stop threshold, and when the difference is less than the stop threshold, judge it as the normal state and continue to monitor;

[0039] When the difference is greater than or equal to the stop threshold, judge it as the abnormal state, calculate the control quantity by calculating the proportional, integral, and differential terms through the PID control algorithm, and adjust the optimal path.

[0040] As a preferred solution of the intelligent luggage obstacle avoidance method based on wireless sensors according to the present invention, wherein: the storage of the sensor data generated by collection and analysis includes:

[0041] Store the collected video data, the generated abnormal monitoring results, and the time index in the central database, and set security access measures. The central database backs up the stored data to the cloud, and regularly performs integrity detection on the stored data and the backup data. After the detection is completed, an integrity detection record is generated and synchronously stored in the central database.

[0042] In a second aspect, the present invention provides an intelligent luggage obstacle avoidance system based on wireless sensors, including,

[0043] A collection and calculation module for collecting sensor data, including time difference, echo time, and image data, inputting the image data into a stacked hourglass network to mark the obstacle area, using the obstacle area as a guiding clue, mapping the detection range of the ultra-wideband sensor to the corresponding ultra-wideband sector, binding the comprehensive time difference value with the obstacle area using area data binding, encapsulating it as the focused time difference data, and calculating the three-dimensional coordinate set and the direction of the obstacle;

[0044] A path construction module for constructing a local grid map, adjusting the pheromone value using an exponential decay function, combining the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate an optimal path;

[0045] The monitoring and storage module is used to decompose the optimal path into small motion instructions and execute them, monitor and adjust the execution results, and store the sensor data collected, analyzed, and generated.

[0046] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the intelligent luggage obstacle avoidance method based on wireless sensors as described in the first aspect of the present invention is implemented.

[0047] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the intelligent luggage obstacle avoidance method based on wireless sensors as described in the first aspect of the present invention is implemented.

[0048] The beneficial effects of the present invention are as follows: The present invention uses a stacked hourglass network to mark the obstacle area as a guiding clue, reduces the interference of irrelevant areas, generates an optimized path by constructing a local grid map and using an exponential decay function to adjust the pheromone value and binding the pheromone to the obstacle area, and improves the obstacle avoidance ability and autonomous navigation performance of intelligent luggage. Description of the Drawings

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0050] Figure 1 It is a flowchart of the intelligent luggage obstacle avoidance method based on wireless sensors in Embodiment 1.

[0051] Figure 2 It is a structural diagram of the intelligent luggage obstacle avoidance system based on wireless sensors in Embodiment 1. Detailed Embodiments

[0052] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention with reference to the drawings in the specification.

[0053] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0054] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0055] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides an intelligent luggage obstacle avoidance method based on wireless sensors, including the following steps:

[0056] S1. Collect sensor data, including time difference, echo time, and image data;

[0057] Input the image data into a stacked hourglass network, mark the obstacle area, use the obstacle area as a guiding clue, map the detection range of the ultra-wideband sensor to the corresponding ultra-wideband sector, use area data binding to bind the comprehensive time difference value with the obstacle area, encapsulate it as the focused time difference data, and calculate the three-dimensional coordinate set and direction of the obstacle;

[0058] Specifically, calculating the three-dimensional coordinate set and direction of the obstacle includes:

[0059] The wireless sensor includes an ultra-wideband sensor, an ultrasonic sensor, and a camera;

[0060] Use the acoustic distance formula to calculate the distance between the obstacle and the ultrasonic sensor, defined as the ultrasonic distance;

[0061] Use linear interpolation method to synchronize the collected sensor data in time, use median filtering to denoise the time difference and echo time data, use Z-score detection to identify abnormal data in the time difference and echo time data, use mean interpolation method to fill in the missing data in the time difference and echo time data, use Gaussian blur to denoise the image data, use histogram equalization to enhance the image data, and use min-max normalization method to normalize the sensor data;

[0062] Input the image data into a stacked hourglass network, detect the feature area of the obstacle in the image through convolution and upsampling processing, use semantic segmentation method to segment the feature area, and mark the area where the obstacle is dominant;

[0063] Use the obstacle area as a guiding clue, use edge detection method to divide the detection range of the ultra-wideband sensor into sectors and map it to the corresponding ultra-wideband sector;

[0064] Calculate the comprehensive time difference value by using the weighted average method to process all time difference data, and bind the comprehensive time difference value with the obstacle area through regional data binding to encapsulate the focused time difference data;

[0065] Calculate the distance of the obstacle by using the two-way ranging method, which is defined as the time difference distance, and the formula is:

[0066] ,

[0067] where d is the distance of the obstacle, c is the signal propagation speed, is the focused time difference data;

[0068] Set the error threshold by using the maximum entropy threshold method, and calculate the error between the ultrasonic distance and the time difference distance by using the absolute error method. If the error is less than or equal to the error threshold, calculate the ultrasonic distance and the time difference distance as the final distance. If the error is greater than the error threshold, use the ultrasonic distance as the final distance;

[0069] Take the intelligent luggage as the origin, calculate the mean value of the ultra-wideband sector as the relative angle of the obstacle, and calculate the two-dimensional coordinates of the obstacle by using the triangulation method;

[0070] Use the Canny edge detection method to extract the contour of the obstacle, convert it into the two-dimensional map coordinates of the contour, take the two-dimensional coordinates of the obstacle as a reference, and use the parallax method to convert the two-dimensional map coordinates of the contour and the two-dimensional coordinates of the obstacle into three-dimensional coordinates to generate a set of three-dimensional coordinates of the obstacle;

[0071] Calculate the circumscribed bounding box of the obstacle by using the bounding box method, determine the width, height and depth of the bounding box, and use the principal component analysis method to extract the main axis direction as the direction of the obstacle.

[0072] Traditional UWB obstacle avoidance systems (such as separate TDOA processing) usually scan the entire detection range without discrimination and cannot optimize signal processing for specific areas. This step uses the visual area ("30% on the right") as a clue to dynamically adjust the focus of TDOA data processing. Compared with traditional full-range ranging technologies, it significantly reduces interference in irrelevant areas (such as signal noise on the left), improves the perception efficiency and accuracy. The visual-guided wireless signal focusing is similar to the human perception mechanism of "looking first and then touching". This cross-sensing collaboration has no precedent in intelligent luggage obstacle avoidance. Existing technologies mostly process visual and wireless data in parallel, rather than letting vision actively intervene in the priority of wireless signals, breaking through the limitations of traditional independent operations. Traditional temporal consistency analysis (such as target tracking in robot navigation) usually relies on a single sensor (such as a camera or radar), while this step integrates visual contour changes and wireless distance changes to form a cross-sensing collaborative dynamic inference. Compared with using only visual tracking (which is vulnerable to occlusion) or wireless ranging (lacking morphological information) alone, this method can more accurately judge the behavior of obstacles. The joint analysis of the time series of visual and wireless data realizes the accurate prediction of the movement trend of obstacles, especially in dynamic scenarios (such as crowded pedestrian areas) with remarkable effects. Traditional obstacle avoidance systems have not tried to deduce dynamic behavior through the synchronous changes of visual and wireless signals. This cross-domain integration breaks through the limitations of single technologies;

[0073] The multi-scale feature extraction ability of the stacked hourglass network, combined with the pixel-level classification of semantic segmentation, ensures the accuracy of obstacle area marking. Compared with traditional edge detection or threshold segmentation, it can more accurately distinguish targets (such as pedestrians and luggage) in complex backgrounds. The marked area provides visual cues for ultra-wideband sector division, breaking through the limitation of traditional wireless sensors relying on signal strength. The vision-guided sector division enables the ultra-wideband detection range to dynamically focus on the obstacle area, avoiding signal confusion in the case of multiple targets with fixed sectors. Compared with traditional uniform division methods, it significantly improves the positioning pertinence. The regional data binding associates the time difference with specific obstacles to form "focused data", reducing redundant signal interference and ensuring the accuracy of subsequent distance calculations. The fusion of vision and wireless signals makes full use of the spatial resolution of images and the time-domain accuracy of ultra-wideband, achieving deep cooperation between multiple sensors. The maximum entropy threshold method dynamically optimizes the threshold according to the error distribution. Compared with the fixed threshold method, it can better adapt to environmental noise (such as acoustic wave reflection and electromagnetic interference), improving the reliability of distance fusion. The conversion from two-dimensional to three-dimensional, combined with contour and depth information, generates a complete spatial representation of the obstacle. Compared with traditional methods with only two-dimensional positioning, it can more accurately describe the position and shape of the obstacle. The main axis direction extracted by PCA reflects the movement trend of the obstacle (such as the pedestrian's orientation), providing a prediction basis for dynamic obstacle avoidance and making up for the deficiency of static positioning. It not only improves the accuracy and robustness of obstacle detection but also provides high-quality input for subsequent path planning. Through vision-guided ultra-wideband sector division, regional data binding, and maximum entropy threshold optimization, the multi-sensor fusion is elevated to a new level.

[0074] S2. Construct a local grid map, adjust the pheromone value using an exponential decay function, combine the pheromone value, set weights using the analytic hierarchy process and perform weighted scoring to generate the optimal path;

[0075] Specifically, constructing a local grid map, adjusting the pheromone value using an exponential decay function, combining the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate the optimal path includes:

[0076] Use the grid map method to place the intelligent luggage at the center of the map as the starting point and construct a local grid map.

[0077] Assign an initial pheromone value to each grid. The initial values of all grids are set to "idle". Each grid represents a position in the map. At initialization, there are no obstacles. Change the value of the grid corresponding to the three-dimensional coordinate set of the obstacle from "idle" to "occupied".

[0078] The center of the obstacle is calculated as a reference point using the coordinate averaging method. The cells within the sector are marked as the affected area using the region expansion method. The pheromone is adjusted using an exponential decay function, such that the pheromone value decays faster the closer it is to the obstacle, resulting in an updated local grid map.

[0079] The current speed is calculated using a positioning algorithm, with the formula:

[0080] ,

[0081] where \(v\) is the current speed, is the distance of the intelligent luggage's displacement, is the time interval;

[0082] The end position of the path is predicted using the kinematic equation, with the formula:

[0083] ,

[0084] ,

[0085] where and are the end positions of the path, and are the current positions of the path respectively, is the direction of the obstacle;

[0086] A candidate path is generated using the artificial potential field method. Using the three-dimensional coordinate set of the obstacles in the local grid map, it is detected whether the cells corresponding to the candidate path are marked as "occupied", and the overlapping "occupied" cells are deleted.

[0087] The remaining candidate paths are selected, the pheromone corresponding to the end position of the path is extracted, the length of the path is calculated using the Euclidean distance formula, the minimum distance between the path and the obstacle is calculated using the closest point distance method, the weight coefficients are set using the analytic hierarchy process, the final score of the path is calculated using the weighted scoring method, and they are sorted from high to low. The candidate path corresponding to the highest score is selected as the optimal path.

[0088] In the traditional ant colony algorithm, pheromone update is mostly based on path selection (such as increment) or uniform evaporation (such as time decay), without directly considering the dynamic influence of the distance to obstacles. By introducing the exponential decay function (commonly used in the field of signal processing) across domains into the pheromone mechanism, a "distance-sensitive pheromone distribution" is formed. This makes the pheromone value not only reflect the path preference but also directly encode the threat level of obstacles (the closer, the more dangerous), providing more fine-grained environmental information for subsequent path selection. Compared with the traditional uniform or linear decay, this exponential decay method simulates the characteristic that danger rapidly decreases with distance in reality (such as the threat increasing sharply when a pedestrian approaches a luggage case), improving the dynamic adaptability of obstacle avoidance. Using the pheromone value (reflecting historical experience and the influence of obstacles) as a scoring factor and combining it with the path length and the minimum distance, a multi-dimensional evaluation system is formed. AHP is usually used for strategic decision-making. The technical solution of the present invention introduces it into real-time path selection and adjusts the weight according to the environmental dynamics (such as obstacle density, moving speed), breaking through the limitation of the traditional fixed weight. Compared with the static scoring of DWA, the technical solution of the present invention realizes "environment-adaptive scoring" through pheromone and AHP, making path selection more intelligent (such as being more inclined to safe paths in crowded pedestrian areas), improving the obstacle avoidance ability of the luggage case in complex dynamic environments (such as airports), and avoiding falling into local traps;

[0089] The local grid map centered on the luggage case provides a high-resolution spatial representation, directly marking "occupied" by combining three-dimensional coordinates. Compared with the traditional two-dimensional grid, it can more realistically reflect the distribution of obstacles and improve the environmental perception ability. Initially assigned as "idle" and dynamically updated as "occupied", it simplifies the computational complexity and is suitable for the real-time processing requirements of portable devices. The standardized grid structure is convenient for integration with other modules (such as path planning, pheromone update), enhancing the scalability of the solution. The exponential decay function makes the pheromone value rapidly decrease with the distance to obstacles (such as approaching 0 near and tending to the initial value of 1 far away). Compared with the traditional uniform or linear decay, it more realistically simulates the threat distribution (such as the sharp risk when a pedestrian approaches). Combining with the region expansion method, pheromone adjustment is not limited to "occupied" grids but also extends to the influence area, refining the granularity of threat encoding. Combining the pheromone value (historical experience), path length (efficiency), and minimum distance (safety), a comprehensive scoring framework is formed. Compared with the traditional single index (such as the shortest distance), it significantly improves the decision-making intelligence. The analytic hierarchy process adjusts the weight according to the environmental dynamics (such as the safety weight increasing to 0.6 when the obstacle density is high), breaking through the limitation of the fixed weight and realizing adaptive path selection. Through the pheromone adjustment of the exponential decay function and the path scoring of AHP combined with pheromone, it demonstrates significant creativity. The former introduces the concept of signal processing across domains to achieve the spatial encoding of dynamic threats, and the latter combines the management decision-making method with the bio-inspired mechanism to construct an environment-adaptive scoring system.

[0090] S3. Decompose the optimal path into small motion instructions and execute them, monitor and adjust the execution results, and store and analyze the sensor data collected.

[0091] Specifically, decomposing the optimal path into small motion instructions and executing them includes:

[0092] Use a path decomposition algorithm to decompose the optimal path into small motion instructions;

[0093] Use a wireless communication protocol to transmit the small motion instructions to the intelligent luggage. The intelligent luggage receives the small motion instructions, uses a direct torque control algorithm to convert the small motion instructions into control signals, and executes them.

[0094] The path decomposition algorithm divides the continuous path into small-step instructions. Compared with directly executing the overall path, it reduces the cumulative error, ensures that the luggage strictly follows the planned trajectory. The decomposed small instructions are convenient for real-time interruption or correction (such as pausing a certain segment of instructions when detecting a new obstacle), enhancing the flexibility of obstacle avoidance and being more adaptable than the traditional execution of the entire path. The wireless communication protocol ensures low latency and high stability of instruction transmission. The direct torque control algorithm realizes the fast execution of small instructions by adjusting the motor torque in real time. DTC avoids complex vector transformations and directly adjusts the voltage based on torque requirements, reducing motor power consumption. The refined design of the path decomposition algorithm transforms traditional static path planning into a dynamically controllable instruction sequence. Combined with the real-time transmission of the wireless communication protocol, it breaks through the limitations of traditional wired or entire-segment execution. The introduction of the direct torque control algorithm applies advanced technologies in the field of motor control across fields to intelligent luggage, achieving the unity of high precision and high efficiency.

[0095] Furthermore, monitoring and adjusting the execution results includes:

[0096] Collect and preprocess the feedback position data, and use the YOLO deep learning algorithm to set the target position;

[0097] Use the deviation method to calculate the difference between the target position and the preprocessed feedback position data;

[0098] Use an empirical rule to set a stop threshold, compare the difference with the stop threshold. When the difference is less than the stop threshold, it is judged as a normal state and continue to monitor;

[0099] When the difference is greater than or equal to the stop threshold, it is judged as an abnormal state. Calculate the control quantity by using the PID control algorithm to obtain the proportional, integral, and differential terms, and adjust the optimal path.

[0100] The YOLO algorithm sets the target position through real-time image analysis. Compared with traditional static target setting, it can dynamically adapt to environmental changes (such as pedestrian movement), improve the pertinence of monitoring, and through deep learning to fuse visual feedback, it can not only detect at the end of the path, but also detect potential threats (such as falling objects), providing forward-looking information for subsequent adjustments. The rule of thumb sets the stop threshold (such as 0.015m) according to environmental characteristics (such as ground smoothness), avoiding the limitations of fixed thresholds. The feedback mechanism of the PID can still execute stably under sensor noise or external disturbances, making up for the vulnerability of on-off control. The introduction of the YOLO deep learning algorithm upgrades traditional static target setting to dynamic visual perception. Combining the deviation method and the rule of thumb forms an efficient state monitoring mechanism. The feedback regulation of the PID control algorithm changes path adjustment from passive correction to active optimization, integrating target detection and control theory across domains.

[0101] Furthermore, store the sensor data generated by collection and analysis, including:

[0102] Store the collected video data, the generated anomaly monitoring results, and the time index in the central database, and set security access measures. The central database backs up the stored data to the cloud, and regularly conducts integrity detection on the stored data and the backup data. After the detection is completed, an integrity detection record is generated and synchronously stored in the central database.

[0103] It simplifies the process of data storage and retrieval, and also helps with real-time data analysis and monitoring. Through the central database, the system can achieve efficient retrieval, analysis, and further data mining of data, thereby improving the working efficiency of intelligent devices or monitoring systems. By backing up data to the cloud, it not only improves the availability of data storage but also enhances the disaster recovery ability of the system. Integrity detection can ensure that the data in the system always remains consistent and accurate, avoiding system failures or decision-making mistakes caused by data errors. The combination of video data and anomaly monitoring results enables the system to accurately judge whether there are problems or abnormal behaviors and take measures in a timely manner.

[0104] This embodiment also provides an intelligent luggage obstacle avoidance system based on wireless sensors, including:

[0105] A collection and calculation module, used to collect sensor data, including time difference, echo time, and image data. Input the image data into a stacked hourglass network to mark the obstacle area. Use the obstacle area as a guiding clue to map the detection range of the ultra-wideband sensor to the corresponding ultra-wideband sector. Use regional data binding to bind the comprehensive time difference value with the obstacle area, encapsulate it as the focused time difference data, and calculate the three-dimensional coordinate set of the obstacle and the direction of the obstacle.

[0106] A path construction module for constructing a local grid map, adjusting pheromone values using an exponential decay function, combining pheromone values, setting weights using the analytic hierarchy process and performing weighted scoring to generate an optimal path;

[0107] A monitoring and storage module for decomposing the optimal path into small motion instructions and executing them, monitoring and adjusting the execution results, and storing the sensor data collected, analyzed and generated.

[0108] This embodiment also provides a computer device applicable to the case of an intelligent luggage obstacle avoidance method based on wireless sensors, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent luggage obstacle avoidance method based on wireless sensors as proposed in the above embodiment.

[0109] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad provided on the outer shell of the computer device, or an external keyboard, a touchpad or a mouse, etc.

[0110] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for obstacle avoidance of an intelligent luggage based on wireless sensors as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, a magnetic disk or an optical disc.

[0111] In summary, the present invention uses a stacked hourglass network to mark the obstacle area as a guiding clue, reduces the interference of irrelevant areas, generates an optimized path by constructing a local grid map and using an exponential decay function to adjust the pheromone value and binding the pheromone to the obstacle area, thereby improving the obstacle avoidance ability and autonomous navigation performance of the intelligent luggage.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An intelligent obstacle avoidance method for smart luggage based on wireless sensors, characterized in that: including collecting sensor data, including time difference, echo time, and image data; inputting the image data into a stacked hourglass network to mark the obstacle area, using the obstacle area as a guiding clue, mapping the detection range of the ultra-wideband sensor to the corresponding ultra-wideband sector, binding the integrated time difference value with the obstacle area using area data binding, encapsulating it as the focused time difference data, and calculating the three-dimensional coordinate set and direction of the obstacle; constructing a local grid map, adjusting the pheromone value using an exponential decay function, combining the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate the optimal path; decomposing the optimal path into small motion instructions and executing them, monitoring and adjusting the execution results, and storing the sensor data collected and analyzed; the calculating of the three-dimensional coordinate set and direction of the obstacle includes: the wireless sensors include ultra-wideband sensors, ultrasonic sensors, and cameras; calculating the distance between the obstacle and the ultrasonic sensor using the acoustic wave distance formula, defined as the ultrasonic distance; inputting the image data into a stacked hourglass network, detecting the feature area of the obstacle in the image through convolution and upsampling processing, segmenting the feature area using semantic segmentation method, and marking the obstacle area; using the obstacle area as a guiding clue, dividing the detection range of the ultra-wideband sensor into sectors using the edge detection method and mapping it to the corresponding ultra-wideband sector; calculating the integrated time difference value after weighted processing of all time difference data using the weighted average method, binding the integrated time difference value with the obstacle area using area data binding, and encapsulating it as the focused time difference data; calculating the distance of the obstacle using the two-way ranging method, defined as the time difference distance; setting the error threshold using the maximum entropy threshold method, calculating the error between the ultrasonic distance and the time difference distance using the absolute error method, if the error is less than or equal to the error threshold, calculating the ultrasonic distance and the time difference distance as the final distance, if the error is greater than the error threshold, using the ultrasonic distance as the final distance; taking the intelligent luggage as the origin, calculating the mean of the ultra-wideband sector as the relative angle of the obstacle, and calculating the two-dimensional coordinates of the obstacle using the triangulation method; extracting the contour of the obstacle using the Canny edge detection method, converting it to the two-dimensional map coordinates of the contour, using the two-dimensional coordinates of the obstacle as a reference, and converting the two-dimensional map coordinates of the contour and the two-dimensional coordinates of the obstacle to three-dimensional coordinates using the parallax method to generate the three-dimensional coordinate set of the obstacle; calculating the circumscribed bounding box of the obstacle using the bounding box method, and extracting the main axis direction using the principal component analysis method as the direction of the obstacle.

2. The intelligent luggage obstacle avoidance method based on wireless sensors according to claim 1, wherein: the constructing of the local grid map, adjusting the pheromone value using an exponential decay function, combining the pheromone value, setting weights using the analytic hierarchy process and performing weighted scoring to generate the optimal path includes: using the grid map method to place the intelligent luggage at the center of the map as the starting point and constructing a local grid map; assigning an initial pheromone value to each grid, setting the initial value of all grids to "idle", and changing the value of the grid corresponding to the three-dimensional coordinate set of the obstacle from "idle" to "occupied"; The center of the obstacle is calculated as a reference point using the coordinate averaging method, the grids in the sector are marked as the affected area using the region expansion method, and the pheromone is adjusted using the exponential decay function to obtain the updated local grid map; The current speed is calculated using the positioning algorithm; The end position of the path is predicted using the kinematic equation; The candidate path is generated using the artificial potential field method, and the three-dimensional coordinate set of the obstacles in the local grid map is used to detect whether the grid corresponding to the candidate path is marked as "occupied", and the overlapping "occupied" grids are deleted; The remaining candidate paths are selected, the pheromone corresponding to the end position of the path is extracted, the path length is calculated using the Euclidean distance formula, the minimum distance between the path and the obstacle is calculated using the nearest point distance method, the weight coefficient is set using the analytic hierarchy process, the final score of the path is calculated using the weighted scoring method, and they are sorted from high to low. The candidate path corresponding to the highest score is selected as the optimal path.

3. The intelligent luggage obstacle avoidance method based on wireless sensors according to claim 2, characterized in that: The decomposition of the optimal path into small motion instructions and their execution include: The optimal path is decomposed into small motion instructions using the path decomposition algorithm; The small motion instructions are transmitted to the intelligent luggage using the wireless communication protocol. The intelligent luggage receives the small motion instructions, converts the small motion instructions into control signals using the direct torque control algorithm, and executes them.

4. The intelligent luggage obstacle avoidance method based on a wireless sensor according to claim 3, characterized in that: The monitoring and adjustment of the execution results include: The feedback position data is collected and preprocessed, and the target position is set using the YOLO deep learning algorithm; The deviation method is used to calculate the difference between the target position and the preprocessed feedback position data; The stop threshold is set using the empirical rule, and the difference is compared with the stop threshold. When the difference is less than the stop threshold, it is judged as the normal state and the monitoring continues; When the difference is greater than or equal to the stop threshold, it is judged as the abnormal state, and the control quantity is calculated by the PID control algorithm to obtain the proportional, integral, and differential terms, and the optimal path is adjusted.

5. The intelligent luggage obstacle avoidance method based on wireless sensors according to claim 1, characterized in that: After the sensor data, preprocessing operations are first performed.

6. The intelligent luggage obstacle avoidance method based on a wireless sensor according to claim 4, characterized in that: The storage of the sensor data collected, analyzed, and generated includes: The collected video data, the generated abnormal monitoring results, and the time index are stored in the central database, and security access measures are set. The central database backs up the stored data to the cloud, and regularly performs integrity detection on the stored data and the backup data. After the detection is completed, an integrity detection record is generated and synchronously stored in the central database.

7. An intelligent luggage obstacle avoidance system based on wireless sensors, based on the intelligent luggage obstacle avoidance method based on wireless sensors according to any one of claims 1 to 6, characterized in that: Including, The collection and calculation module is used to collect sensor data, including time difference, echo time, and image data. The image data is input into the stacked hourglass network to mark the obstacle area. The obstacle area is used as a guiding clue, the detection range of the ultra-wideband sensor is mapped to the corresponding ultra-wideband sector, and the integrated time difference value is bound to the obstacle area using the regional data binding to encapsulate the focused time difference data, and the three-dimensional coordinate set and the direction of the obstacle are calculated; The path construction module is used to construct the local grid map, adjust the pheromone value using the exponential decay function, combine the pheromone value, set the weight using the analytic hierarchy process and perform weighted scoring to generate the optimal path; The monitoring and storage module is used to decompose the optimal path into small motion instructions and execute them, monitor and adjust the execution results, and store the sensor data collected, analyzed and generated.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the intelligent luggage obstacle avoidance method based on wireless sensors according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the intelligent luggage obstacle avoidance method based on wireless sensors according to any one of claims 1 to 6 are implemented.

Images