Intelligent luggage case obstacle avoidance method based on ultra wide band

Abstract

The invention discloses an intelligent luggage case obstacle avoidance method based on an ultra wide band, and relates to the technical field of intelligent mobile equipment control, and the method comprises the steps: collecting and preprocessing multi-source environment perception data, and generating an environment perception fusion data set; correcting projection in the ranging direction and constructing a local environment grid map based on grid division and grid occupation probability; comparing the bottom direction TOF height value with a ground reference height by adopting a threshold value judgment method, identifying a height difference risk area, and generating a path guide point through Cartesian coordinate conversion; based on the path guide points and the local environment grid map, generating a trajectory candidate set by adopting a dynamic window method, and removing impassable trajectories through a collision detection algorithm in combination with obstacle grid information; through a multi-target scoring function and a double-DIP controller, predicting a collision risk score of the feasible trajectory and generating an obstacle avoidance control instruction; the obstacle avoidance control instruction is executed, environment sudden change data are collected in real time, and the obstacle avoidance control instruction is adjusted in real time through a path replanning algorithm.

Description

Technical Field

[0001] This invention relates to the field of intelligent mobile device control technology, and in particular to an obstacle avoidance method for an intelligent suitcase based on ultra-wideband. Background Technology

[0002] With the development of smart hardware, mobile robots, and multi-source perception fusion technology, smart suitcases with path guidance and autonomous obstacle avoidance capabilities are gradually being applied to user travel assistance scenarios. Existing systems typically integrate a Time-of-Flight (TOF) ranging sensor for forward obstacle detection, an Induction Unit (IMU) for attitude measurement, and a UWB ranging unit to achieve user relative position perception. After constructing a local environment map based on the perceived information, a traversable trajectory is generated using a dynamic window method, and motion control is achieved through differential drive. This type of path control scheme has been validated in various types of portable autonomous devices, demonstrating good traversability and basic environmental adaptability in standard scenarios.

[0003] In dynamic environments, conventional path planning mechanisms often rely on obstacle grid determination for trajectory selection, which fails to adequately reflect the impact of speed execution constraints on path traversability. Furthermore, when sudden changes in user location or disturbances in elevation areas occur, path command updates frequently employ a single-cycle replacement strategy, lacking contextual analysis in the response process and making it difficult to guarantee the stability of the path following process and the continuity of obstacle avoidance commands. Summary of the Invention

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

[0005] Therefore, this invention provides an ultra-wideband-based intelligent luggage obstacle avoidance method to solve the problems of real-time adaptability of path control and dynamic obstacle avoidance response in sudden change scenarios.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent suitcase obstacle avoidance method based on ultra-wideband, comprising, Multi-source environmental perception data is collected and preprocessed to generate an environmental perception fusion dataset. Based on the environmental perception fusion dataset, the ranging direction projection is corrected, and a local environmental grid map is constructed using gridded spatial segmentation and occupied grid probability update methods. A threshold judgment method is used to compare the bottom-direction TOF height value with the ground reference height to identify elevation difference risk areas, and path guidance points are generated through Cartesian coordinate transformation. Based on the path guidance points and the local environmental grid map, a trajectory candidate set is generated using the dynamic window method, and inaccessible trajectories are eliminated by combining collision detection algorithms with obstacle grid information. Collision risk scores of feasible trajectories are predicted using a multi-objective scoring function and dual DIP controllers, and obstacle avoidance control commands are generated. The obstacle avoidance control commands are executed, and environmental change data is collected in real time. The obstacle avoidance control commands are adjusted in real time using a path replanning algorithm.

[0007] As a preferred embodiment of the intelligent luggage obstacle avoidance method based on ultra-wideband described in this invention, the multi-source environmental perception data includes forward TOF radar depth map data, bottom TOF ground height value, UWB ranging value and heading angle, pitch angle, roll angle, heading angle, linear velocity and angular velocity. The preprocessing includes timestamp synchronization, interpolation alignment, and encapsulation.

[0008] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the steps for correcting the ranging direction projection and constructing a local environment grid map through gridded spatial segmentation and grid occupancy probability update are as follows: Based on the environmental perception fusion dataset, the projection error of the ranging direction is identified by the projection residual matching method, and the attitude supplementation function is used to dynamically correct the projection error of the ranging direction to obtain the corrected ranging point. Use Cartesian coordinate transformation to convert each distance measurement point into a two-dimensional coordinate system with the center of the suitcase as the origin; By dividing the space into grids, the two-dimensional coordinate system is divided into several grid cells according to a fixed grid size; Based on grid cells, the ranging path is marked using the Bresenham linear grid filling algorithm, and the obstacle probability of each grid cell in the current period is updated using the Bayesian inference-based occupancy probability update method to obtain a local environment grid map for obstacle avoidance path judgment.

[0009] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the steps for identifying elevation difference risk areas are as follows: The bottom-oriented TOF distance measurement value is obtained by a bottom-oriented TOF distance measurement sensor and compared with a preset ground reference height to identify the degree of deviation. Based on historical ground height observation data, a deviation threshold is defined. When the deviation is greater than or equal to the deviation threshold, the current location is marked as a height difference risk point.

[0010] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the step of generating path guidance points through Cartesian coordinate transformation is as follows: This is executed once per control cycle. If a point is marked as a height difference risk point twice in a row, then several grid areas will be expanded forward and marked as height difference danger zones. The relative distance to the user is obtained by using a UWB ranging sensing unit and the user's heading angle is obtained by using the AOA angle measurement method. The main control processing unit combines the user's heading angle and the heading angle to obtain the user's relative polar coordinate information. Based on the user's relative polar coordinate information, the relative position coordinates of the user in the local map are calculated by converting polar coordinates to Cartesian coordinates, and the position and orientation angle are encapsulated as path guidance points.

[0011] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the steps of generating a candidate trajectory set using a dynamic window method and eliminating impassable trajectories by combining collision detection algorithm with obstacle grid information are as follows. Based on linear velocity and angular velocity, a control command space is constructed by combining the current velocity state, acceleration constraints, and control cycle; The dynamic window method is used to sample multiple velocity combinations in the control command space, and the trajectory prediction method based on differential drive constraints is combined to simulate the pose change trajectory of the suitcase in the future time. At the same time, a fixed time step is used to perform integral calculation to predict the trajectory points of the suitcase. By using coordinate mapping and grid indexing, the luggage track points are mapped to the local environment grid map. If the luggage track points fall into the grid of the elevation difference risk area, they are considered as impassable tracks and are removed.

[0012] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the steps of predicting the collision risk score of feasible trajectories and generating obstacle avoidance control commands through a multi-objective scoring function and dual DIP controllers are as follows: Using geometric feature extraction methods, the end position, running direction angle and path length of feasible trajectories are extracted, excluding infeasible trajectories. The feasibility status marker of the feasible trajectory in the current cycle is obtained through wheel speed constraint judgment and feasibility judgment methods. Based on the executability status marker, the path scoring method is adopted to identify the target distance error between the end point of each feasible trajectory and the path guide point, the cosine value of the angle between the trajectory direction and the target direction, and the minimum obstacle spacing within the trajectory. After normalization, they are weighted to form a multi-objective scoring function. The DIP1 controller obtains the ideal linear velocity based on the target distance error; The DIP2 controller obtains the ideal angular velocity through the directional angle error, and then fuses it with the trajectory velocity using a weighted average method to generate obstacle avoidance control commands.

[0013] As a preferred embodiment of the ultra-wideband-based intelligent suitcase obstacle avoidance method of the present invention, the steps of adjusting the obstacle avoidance control command in real time using a path replanning algorithm are as follows: Execute obstacle avoidance control commands and collect real-time environmental change data; Based on environmental mutation data, a multi-source sensing fusion and time-series change detection and analysis method is adopted to construct a mutation state judgment logic and identify dynamic environmental change events such as path blockage, elevation difference risk, and user leaving the guidance area. Based on dynamic environmental change events, the path replanning algorithm dynamically adjusts the path guidance points and updates obstacle avoidance control commands in real time.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the ultra-wideband-based intelligent suitcase obstacle avoidance method described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the ultra-wideband-based intelligent suitcase obstacle avoidance method described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By generating a candidate trajectory set using a dynamic window method based on path guidance points and a local environment grid map, and eliminating impassable trajectories through a collision detection algorithm combined with obstacle grid information, a unified screening of path feasibility and system controllability is achieved, enhancing the execution stability of control outputs. Furthermore, by using a multi-objective scoring function and a dual DIP controller, collision risk scores of feasible trajectories are predicted and obstacle avoidance control commands are generated, achieving comprehensive evaluation and fusion control of path direction, target distance, and safety distance, thus improving adaptability to dynamic environmental changes. Ultimately, this results in obstacle avoidance control with higher path feasibility and more timely control response. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an obstacle avoidance method for an intelligent suitcase based on ultra-wideband.

[0019] Figure 2 This is a flowchart of a multi-source fusion gridded dynamic window obstacle avoidance planning method. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0023] Reference Figures 1-2 This is one embodiment of the present invention, which provides an intelligent suitcase obstacle avoidance method based on ultra-wideband, comprising the following steps: S1. Collect multi-source environmental perception data and preprocess it to generate an environmental perception fusion dataset; Multi-source environmental perception data includes forward TOF radar depth map data, bottom TOF ground height values, UWB ranging values ​​and heading angles, pitch angles, roll angles, yaw angles, linear velocity, and angular velocity; Furthermore, forward-facing TOF radar depth map data acquires spatial distribution information of obstacles in the environment ahead; bottom-facing TOF radar detects the relative height change between the bottom of the suitcase and the ground, identifying terrain elevation differences and areas at risk of sinking; UWB ranging values ​​and heading angles acquire the relative spatial relationship between the user and the suitcase, especially the generation of path guidance points; IMU acquires pitch, roll, and heading angles for attitude estimation and ranging direction correction; linear velocity and angular velocity are obtained by the main control processing unit through decoding encoder feedback information to identify the actual motion state of the suitcase within the current control cycle.

[0024] Preprocessing includes timestamp synchronization, interpolation alignment, and encapsulation.

[0025] Furthermore, the preprocessing includes timestamp synchronization, interpolation alignment, and encapsulation. Timestamp synchronization addresses the inconsistency in data timelines caused by differences in sampling frequencies and response delays among different sensors during data acquisition. This invention employs a unified time reference mechanism to align various types of data. Interpolation alignment involves using methods such as linear interpolation on the timeline to fill in misaligned data nodes, ensuring the continuity of data sequence during the fusion process. Encapsulation involves uniformly constructing the synchronized heterogeneous data into a structured environmental perception fusion dataset, ensuring standardization, timeliness, and repeatable parsing.

[0026] S2. Based on the environmental perception fusion dataset, the ranging direction projection is corrected and a local environmental grid map is constructed through gridded spatial segmentation and grid occupancy probability update method. Based on the environmental perception fusion dataset, the projection error of the ranging direction is identified by the projection residual matching method, and the attitude supplementation function is used to dynamically correct the projection error of the ranging direction to obtain the corrected ranging point. Use Cartesian coordinate transformation to convert each corrected distance measurement point into a two-dimensional coordinate system with the center of the suitcase as the origin; By dividing the space into grids, the two-dimensional coordinate system is divided into several grid cells according to a fixed grid size; Based on grid cells, the ranging path is marked using the Bresenham linear grid filling algorithm, and the obstacle probability of each grid cell in the current period is updated using the Bayesian inference-based occupancy probability update method to obtain a local environment grid map for obstacle avoidance path judgment.

[0027] Furthermore, depth map data from the forward-facing TOF radar is acquired within the current period, and combined with the pitch and roll angles provided by the IMU, the attitude deviation in the ranging direction is identified using the projection residual matching method. The original projection direction vector is calculated, expressed as: ; in, It is the original projection direction vector. It is the first The horizontal scanning angle of the ranging ray.

[0028] Attitude compensation is performed by combining the pitch and roll angles output by the IMU, and the rotation-corrected direction vector is calculated as follows: ; in, It is the rotation-corrected direction vector. It is a rotation matrix about the pitch axis. It is the rotation matrix about the roll axis, and the compensated ranging direction vector; Based on the original distance measurement, the corrected distance measurement point is obtained, expressed as: ; in, It is the compensation after the first The coordinate vectors of a ranging point in three-dimensional space. It is the first The original distance measurement value of the ray; Using Cartesian coordinate transformation Mapped to a two-dimensional coordinate system with the center of the suitcase as the origin, and discarding the height component, the expression is: ; in, It is a three-dimensional ranging point Projecting the horizontal and vertical coordinates onto a local two-dimensional coordinate system. It is the first after attitude correction The x-axis and y-axis coordinates of a distance measuring point in a local three-dimensional coordinate system.

[0029] By using a gridded spatial partitioning method, the two-dimensional coordinate space is divided into a uniform grid with a side length of Δ, and the corresponding grid index is: ; in, It is a distance measuring point In the local environment raster map, the raster index coordinates, Δ is the side length of the raster cell.

[0030] To fill the grid along the ranging path, the Bresenham straight line algorithm is used, from the sensor origin to each... Perform integer grid step fill.

[0031] It should also be noted that, to achieve dynamic occupancy probability modeling for each grid cell, a Bayesian inference-based update method is employed. The update formula is: ; in, exist Time Grid The posterior probability of being occupied. yes real time for grid Prior probability of being occupied Is Observed in occupied state The probability, It is the first in the local environment raster map grid pairs The observation results , indicating empty and occupied, It is the first in the local environment raster map One grid cell, It is a normalization constant.

[0032] Bayesian inference can achieve dynamic fusion of uncertainties in sensor observations, avoiding distortion of environmental mapping caused by occasional ranging errors.

[0033] S3. Use the threshold judgment method to compare the bottom TOF height value with the ground reference height to identify the elevation difference risk area, and generate path guidance points through Cartesian coordinate transformation; The bottom-oriented TOF ranging value is obtained by a bottom-oriented TOF ranging sensor and compared with a preset ground reference height to identify the degree of deviation. Based on historical ground height observation data, a deviation threshold is defined. When the degree of deviation is greater than or equal to the deviation threshold, the current location is marked as a height difference risk point.

[0034] Furthermore, the bottom-oriented TOF ranging sensor acquires the bottom-oriented TOF ranging value for the current period, compares it with the preset ground reference height, and calculates the degree of deviation between the two: ; in, It is a highly deviated value. This is the ranging value collected by the TOF ranging sensor from bottom to top during the current cycle. It is the elevation difference deviation threshold; To quantify the risk identification boundary, a statistical model is constructed based on historical multi-period ground height observation data, and a height deviation threshold is defined. The height deviation value within the current period satisfies: ; in, It is the elevation difference deviation threshold; Then mark the current luggage compartment location grid where the ranging point is located as an elevation difference risk point.

[0035] This is executed once per control cycle. If a point is marked as a height difference risk point twice in a row, then several grid areas will be expanded forward and marked as height difference danger zones. The relative distance to the user is obtained by using a UWB ranging sensing unit, and the user's heading angle is obtained by using the AOA angle measurement method. The user's heading angle and the heading angle are combined to obtain the user's relative polar coordinate information. Based on the user's relative polar coordinate information, the relative position coordinates of the user in the local map are calculated by converting polar coordinates to Cartesian coordinates, and the position and orientation angle are encapsulated as path guidance points.

[0036] Furthermore, a deviation judgment logic is executed once in each control cycle. If two consecutive detection results meet the elevation difference risk point, then based on the current heading direction, several forward grid areas are expanded along the direction of travel in the local environmental grid map and uniformly marked as elevation difference danger areas.

[0037] It should also be noted that real-time guidance and control target determination based on user location information are performed. The real-time relative distance between the user and the suitcase is obtained through a UWB ranging sensing unit, and the user's heading angle relative to the suitcase is obtained using an AOA angle measurement method. The main control processing unit simultaneously receives the current heading angle of the suitcase from the IMU, and combines the two angles to form the user's relative polar coordinate information: ; in, It is the synthesized user relative polar radius. It is the synthesized user global orientation angle. It is the real-time relative distance between the user and the suitcase. It is the user orientation angle. It is the current heading angle of the suitcase; The user's position in the local two-dimensional Cartesian coordinate system is: ; in, It represents the user's position coordinates in a local two-dimensional Cartesian coordinate system with the suitcase as the origin; The relative coordinate position and the corresponding direction angle are encapsulated together as a path guide point, which serves as the target point for local path planning, and subsequent actions such as direction guidance, dynamic following and path control based on the path guide point are completed.

[0038] S4. Based on the path guidance points and the local environment grid map, a dynamic window method is used to generate a trajectory candidate set, and an impassable trajectory is eliminated by combining the collision detection algorithm with obstacle grid information. Based on linear velocity and angular velocity, a control command space is constructed by combining the current velocity state, acceleration constraints, and control cycle; The dynamic window method is used to sample multiple velocity combinations in the control command space, and the trajectory prediction method based on differential drive constraints is combined to simulate the pose change trajectory of the suitcase in the future time. At the same time, a fixed time step is used to perform integral calculation to predict the trajectory points of the suitcase. By using coordinate mapping and grid indexing, the luggage track points are mapped to the local environment grid map. If the luggage track points fall into the grid of the elevation difference risk area, they are considered as impassable tracks and are removed.

[0039] Furthermore, based on the path guidance points and the local environment grid map, a control command space with linear velocity and angular velocity as two-dimensional parameters is constructed. The boundary of the control command space is dynamically adjusted according to the instantaneous linear velocity and angular velocity of the current luggage, combined with the maximum linear acceleration and maximum angular acceleration limits. By fixing the control period, the achievable velocity range of linear velocity and angular velocity within the current period is calculated. ; ; in, It refers to the linear velocity in the control command. It is the actual linear velocity of the suitcase during the current cycle. It is the maximum linear acceleration. It is a control cycle. It refers to the angular velocity in the control command. It is the actual angular velocity of the suitcase during the current period. It is the maximum angular acceleration; In a two-dimensional velocity space, linear velocity and angular velocity are sampled using a dynamic window method to obtain several sets of... Control pairs. For each control pair, a trajectory prediction method based on differential drive constraints is used to calculate the pose change trajectory of the suitcase within a future time window. Specifically, a fixed time step is adopted. Integrating the motion process yields a sequence of trajectory points: ; ; ; in, It is the current The x-coordinate of the time trajectory point in the two-dimensional coordinate system. yes The x-coordinate of the predicted point at time +1 It is the first in sampling control Correct linear velocity, It is the current The heading angle at any moment, It is the integration time step. It is the current The ordinate of the time trajectory point in the two-dimensional coordinate system. yes The predicted ordinate of the point at time +1 yes The heading angle of the predicted point at time +1 It is the sampling control number The correct angular velocity; The obtained trajectory points The trajectory is transformed into a local environment raster map using a coordinate mapping function, and the grid cell corresponding to each trajectory point is located using the raster indexing method. The raster cell is then queried to determine if it is marked as an obstacle raster or an elevation risk area. If any point in the trajectory falls into an elevation risk area, the trajectory is marked as impassable and removed from the candidate set.

[0040] All remaining trajectories constitute a trajectory candidate set, providing basic input for the subsequent path scoring and control instruction generation module.

[0041] S5. Through a multi-objective scoring function and dual DIP controllers, predict the collision risk score of feasible trajectories and generate obstacle avoidance control commands. Using geometric feature extraction methods, the end position, running direction angle and path length of feasible trajectories are extracted, excluding infeasible trajectories. The feasibility status marker of the feasible trajectory in the current cycle is obtained through wheel speed constraint judgment and feasibility judgment methods. Based on the executability status marker, the path scoring method is adopted to identify the target distance error between the end point of each feasible trajectory and the path guide point, the cosine value of the angle between the trajectory direction and the target direction, and the minimum obstacle spacing within the trajectory. After normalization, they are weighted to form a multi-objective scoring function. The DIP1 controller obtains the ideal linear velocity based on the target distance error; The DIP2 controller obtains the ideal angular velocity through the directional angle error, and then fuses it with the trajectory velocity using a weighted average method to generate obstacle avoidance control commands.

[0042] Furthermore, based on the trajectory candidate set, feature extraction and executability analysis are performed on each trajectory. Geometric feature extraction methods are used to extract the end position coordinates of each trajectory. End direction angle and the overall path length of the trajectory .

[0043] Simultaneously, combining the trajectory prediction method with wheel speed upper and lower limits and differential drive constraints, and control cycle constraints, a wheel speed constraint judgment and executability determination method is executed to determine whether the trajectory can be executed in the current control cycle. Trajectories that meet the constraints are marked with an executability status flag and proceed to the subsequent scoring and control stages.

[0044] Based on the executability status markers, the execution path scoring method is used for each executable trajectory to calculate the following three performance metrics: (1) Target distance error : Euclidean distance error between the endpoint of the trajectory and the path guide point; (2) Cosine error of direction angle : Cosine of the angle between the trajectory terminal direction angle and the guide point direction angle; (3) Minimum obstacle spacing The minimum distance between all trajectory points and the obstacle grid.

[0045] Each indicator is normalized to its maximum value and then multiplied by a set weighting coefficient. This forms the final multi-objective scoring function: ; in, It is the total score of the multi-objective scoring function. It is the difference between the direction angle at the end of the trajectory and the direction angle at the path guide point. It is the cosine of the direction angle error; the closer the value is to 1, the higher the direction consistency. It is the minimum distance between all trajectory points and obstacle grids in the trajectory. It is the Euclidean distance error between the end point of the path and the guide point. , , These are the weight coefficients of each indicator in the multi-objective scoring function, corresponding to the weights of target distance error, directional consistency, and obstacle spacing, respectively, satisfying: After obtaining the scores of all tracks, the track with the highest score is selected as the optimal path for this cycle, and the dual DIP controller is invoked to calculate obstacle avoidance control instructions based on its score.

[0046] The DIP1 controller obtains the ideal linear velocity based on the target distance error and a speed mapping relationship based on proportional adjustment. ; The DIP2 controller outputs the ideal angular velocity based on the angle difference between the direction angle at the end of the trajectory and the direction angle at the path guide point. .

[0047] Ideal speed The actual velocities corresponding to the optimal trajectory are weighted and averaged to form the final obstacle avoidance control command: ; ; in, It is the ideal linear velocity output by the DIP1 controller. It is the ideal angular velocity output by the DIP2 controller. , These are the average linear velocity and angular velocity of the corresponding trajectory in trajectory prediction. , These are the final fused linear and angular velocity control commands. It is the linear velocity fusion coefficient. It is the angular velocity fusion coefficient; S6. Execute obstacle avoidance control commands and collect environmental change data in real time, and adjust the obstacle avoidance control commands in real time through path replanning algorithm.

[0048] Execute obstacle avoidance control commands and collect real-time environmental change data; Based on environmental mutation data, a multi-source sensing fusion and time-series change detection and analysis method is adopted to construct a mutation state judgment logic and identify dynamic environmental change events such as path blockage, elevation difference risk and user leaving the guidance area. Based on dynamic environmental change events, the path replanning algorithm dynamically adjusts the path guidance points and updates obstacle avoidance control commands in real time.

[0049] Furthermore, after generating the obstacle avoidance control command in step S5, the final output obstacle avoidance control command is executed immediately. The instructions have been obtained through fusion using dual DIP controllers, specifically in the following form: ; in, , The linear velocity and angular velocity are derived from the end of the optimal trajectory. While executing instructions, real-time environmental monitoring is performed, and sensor data is collected to form an environmental mutation dataset. ; in, This is the current control cycle. The collection of environmental mutation data collected internally. It represents the change in distance in the forward TOF depth map relative to the previous period. It is the change in altitude of the bottom-view TOF ranging value relative to the previous cycle. It is the change in altitude of the UWB ranging value relative to the previous cycle. It is the change in the angular value relative to the previous cycle; right A sudden change judgment logic function is established by employing multi-source sensing fusion and temporal change detection and analysis methods. It is used to identify three types of dynamic events: Path obstruction judgment: If a new obstacle point appears in the end region of the trajectory (by... > (trigger), then set =1; Elevation risk assessment: If < And if the condition is satisfied for two consecutive periods, then assume =2; User deselection judgment: If ,set up =3.

[0050] When it exists When the value is a valid value in {1,2,3}, the path replanning mechanism is triggered.

[0051] Path replanning based on local environment grid map With the newly generated target vector Then, re-execute the control flow of steps S4 and S5. That is: by As a path guiding point, a control command space is constructed, and the trajectory is resampled in the velocity space using the dynamic window method to calculate the scoring function: ; After selecting the optimal trajectory based on the scoring function, the ideal velocity output by the dual DIP controller is weighted and fused with the original trajectory velocity to calculate and generate new control commands based on the linear and angular velocity control commands executed. .

[0052] The update command overwrites and replaces the original. This enables real-time adjustment and closed-loop response of obstacle avoidance control behavior, ensuring that the suitcase maintains stable and adaptive navigation capabilities even when the path is obstructed, there are elevation disturbances, or user guidance is interrupted. This embodiment also provides a computer device applicable to the ultra-wideband-based intelligent suitcase obstacle avoidance method, comprising: 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 ultra-wideband-based intelligent suitcase obstacle avoidance method proposed in the above embodiment.

[0053] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0054] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the obstacle avoidance method for an ultra-wideband intelligent suitcase as proposed in the above embodiments. 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), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0055] In summary, this invention achieves a unified screening of path feasibility and system controllability by: generating a candidate trajectory set using a dynamic window method based on path guidance points and a local environment grid map; and eliminating impassable trajectories using a collision detection algorithm combined with obstacle grid information, thereby enhancing the execution stability of control outputs; and by using a multi-objective scoring function and a dual DIP controller to predict the collision risk score of feasible trajectories and generate obstacle avoidance control commands, achieving comprehensive evaluation and fusion control of path direction, target distance, and safety distance, thus improving adaptability to dynamic environmental changes. Ultimately, this invention achieves obstacle avoidance control with higher path feasibility and more timely control response.

[0056] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An obstacle avoidance method for an intelligent suitcase based on ultra-wideband, characterized in that: include, Collect multi-source environmental perception data and preprocess it to generate an environmental perception fusion dataset; Based on the environmental perception fusion dataset, the ranging direction projection is corrected and a local environmental grid map is constructed through gridded spatial segmentation and grid occupancy probability update method. A threshold judgment method is used to compare the bottom TOF height value with the ground reference height to identify areas with elevation difference risks, and path guidance points are generated through Cartesian coordinate transformation. Based on path guidance points and local environment grid maps, a dynamic window method is used to generate a trajectory candidate set, and an impassable trajectory is eliminated by combining a collision detection algorithm with obstacle grid information. By using a multi-objective scoring function and a dual DIP controller, collision risk scores for feasible trajectories are predicted and obstacle avoidance control commands are generated. It executes obstacle avoidance control commands and collects real-time environmental change data, and adjusts the obstacle avoidance control commands in real time through path replanning algorithm.

2. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 1, characterized in that: The multi-source environmental perception data includes forward TOF radar depth map data, bottom TOF ground height value, UWB ranging value and heading angle, pitch angle, roll angle, heading angle, linear velocity and angular velocity; The preprocessing includes timestamp synchronization, interpolation alignment, and encapsulation.

3. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 2, characterized in that: The process involves correcting the ranging direction projection and constructing a local environment grid map using a gridded spatial segmentation and grid occupancy probability update method. The steps are as follows. Based on the environmental perception fusion dataset, the projection error of the ranging direction is identified by the projection residual matching method, and the attitude supplementation function is used to dynamically correct the projection error of the ranging direction to obtain the corrected ranging point. Use Cartesian coordinate transformation to convert each distance measurement point into a two-dimensional coordinate system with the center of the suitcase as the origin; By dividing the space into grids, the two-dimensional coordinate system is divided into several grid cells according to a fixed grid size; Based on grid cells, the ranging path is marked using the Bresenham linear grid filling algorithm, and the obstacle probability of each grid cell in the current period is updated using the Bayesian inference-based occupancy probability update method to obtain a local environment grid map for obstacle avoidance path judgment.

4. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 3, characterized in that: The steps for identifying areas with elevation differences and risks are as follows. The bottom-oriented TOF distance measurement value is obtained by a bottom-oriented TOF distance measurement sensor and compared with a preset ground reference height to identify the degree of deviation. Based on historical ground height observation data, a deviation threshold is defined. When the deviation is greater than or equal to the deviation threshold, the current location is marked as a height difference risk point.

5. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 4, characterized in that: The steps for generating path guide points through Cartesian coordinate transformation are as follows. Each control cycle performs a deviation assessment. If a point is marked as a height difference risk point twice in a row, then several grid areas are extended forward and marked as height difference risk areas. The relative distance to the user is obtained by using a UWB ranging sensing unit, and the user's heading angle is obtained by using the AOA angle measurement method. The main control processing unit combines the user's heading angle and the heading angle to obtain the user's relative polar coordinate information. Based on the user's relative polar coordinate information, the relative position coordinates of the user in the local map are calculated by converting polar coordinates to Cartesian coordinates, and the relative position coordinates and direction angles are encapsulated as path guidance points.

6. The obstacle avoidance method for an ultra-wideband intelligent suitcase as described in claim 5, characterized in that: The process involves generating a candidate trajectory set using a dynamic window method and then eliminating impassable trajectories using a collision detection algorithm combined with obstacle grid information. The steps are as follows: Based on linear velocity and angular velocity, a control command space is constructed by combining the current velocity state, acceleration constraints, and control cycle; A dynamic window method is used to sample multiple velocity combinations in the control command space, and a trajectory prediction method based on differential drive constraints is combined to simulate the pose change trajectory of the suitcase in the future. Simultaneously, based on the position and direction of the path guidance point, the endpoint deviation and directional angle between each trajectory and the path guidance point are calculated to predict the luggage trajectory point; By using coordinate mapping and grid indexing, the luggage track points are mapped to the local environment grid map. If the luggage track points fall into the grid of the elevation difference risk area, they are considered as impassable tracks and are removed.

7. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 6, characterized in that: The steps for predicting collision risk scores for feasible trajectories and generating obstacle avoidance control commands using a multi-objective scoring function and dual DIP controllers are as follows: The geometric feature extraction method is used to extract the end position, running direction angle and path length of the feasible trajectory, and the executability status mark of the feasible trajectory in the current cycle is obtained through wheel speed constraint judgment and executability judgment method. Based on the executability status marker, the path scoring method is adopted to identify the target distance error between the end point and the path guide point of each feasible trajectory, the cosine value of the angle between the trajectory direction and the target direction, and the minimum obstacle spacing within the trajectory. After normalization, they are weighted to form a multi-objective scoring function. Based on the multi-objective scoring function, the optimal trajectory is selected from the feasible trajectories, and the linear velocity and angular velocity are extracted. The DIP1 controller obtains the ideal linear velocity based on the target distance error; The DIP2 controller obtains the ideal angular velocity through the azimuth error, and then fuses it with the ideal linear velocity and ideal angular velocity using a weighted average method to generate obstacle avoidance control commands.

8. The obstacle avoidance method for an intelligent suitcase based on ultra-wideband as described in claim 7, characterized in that: The steps for adjusting obstacle avoidance control commands in real time using a path replanning algorithm are as follows: Execute obstacle avoidance control commands and collect real-time environmental change data; Based on environmental mutation data, a multi-source sensing fusion and time-series change detection and analysis method is adopted to construct a mutation state judgment logic and identify dynamic environmental change events such as path blockage, elevation difference risk, and user leaving the guidance area. Based on dynamic environmental change events, the path replanning algorithm dynamically adjusts the path guidance points and updates obstacle avoidance control commands in real time.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the obstacle avoidance method for an ultra-wideband intelligent suitcase as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the obstacle avoidance method for an ultra-wideband intelligent suitcase as described in any one of claims 1 to 8.

Images