A positioning method and system for linking an intelligent transfer vehicle with an electric hospital bed

Abstract

The present application relates to a positioning method and system for the linkage between an intelligent transfer vehicle and an electric bed, the method comprising obtaining identification information of the electric bed, the identification information including position information and posture information; collecting the posture data of the intelligent transfer vehicle itself, fusing the identification information, the posture data of the intelligent transfer vehicle itself and the surrounding environment information to generate a three-dimensional docking path; continuously adjusting the posture parameters of the intelligent transfer vehicle during movement; receiving the structural status information of the electric bed in real time during navigation and movement along the three-dimensional docking path to determine whether the preset docking conditions are met; after the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, sending a docking completion prompt signal; after receiving the docking completion prompt signal, executing the patient transfer operation to transfer the patient from the electric bed to the intelligent transfer vehicle platform. The present application has the effect of improving the safety of the patient transfer process.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent control and auxiliary docking of medical equipment, and in particular to a positioning method and system for the linkage between an intelligent transfer vehicle and an electric hospital bed. Background Art

[0002] Currently, in medical settings, patients in intensive care, postoperative recovery, or long-term bed rest often rely on dedicated intelligent transfer vehicles to transport patients between wards, operating rooms, and examination areas. To improve nursing efficiency, reduce physical exertion for medical staff, and ensure patient safety during transfers, intelligent transfer equipment is gaining widespread adoption.

[0003] However, in traditional solutions, the coordinated docking between intelligent transfer carts and electric beds still relies mainly on manual operation. Specifically, the navigation path planning of the transfer cart is usually based on a preset path or linear target control. It lacks spatial perception between the transfer cart and the bed, and cannot achieve accurate recognition and dynamic adjustment of the relative position and posture of the bed. During the docking process, if structural parameters such as bed height, bed surface angle or brake status are not manually corrected, there will be risks such as inaccurate docking position, posture misalignment, and even patient slippage.

[0004] The above-mentioned existing technical solutions have the following defects: the existing path planning method is difficult to deal with obstacles in complex scenarios in real time, lacks adaptive adjustment and task recovery capabilities for path interruptions, limits the continuity and safety of transfer tasks, and therefore there is room for improvement. Summary of the Invention

[0005] In order to improve the safety of the patient transfer process, the present application provides a positioning method and system for the linkage between an intelligent transfer vehicle and an electric bed.

[0006] The above-mentioned invention objective of this application is achieved through the following technical solutions:

[0007] A positioning method for an intelligent transfer vehicle linked to an electric bed, the positioning method comprising:

[0008] The positioning sensor group on the intelligent transfer vehicle collects surrounding environment information and identifies the positioning identification device on the electric bed to obtain identification information of the electric bed, wherein the identification information includes position information and posture information;

[0009] Collecting the posture data of the intelligent transfer vehicle, fusing the identification information, the posture data of the intelligent transfer vehicle and the surrounding environment information to generate a three-dimensional docking path for docking;

[0010] Controlling the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjusting the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed;

[0011] During navigation and movement along the three-dimensional docking path, the structural status information of the electric bed is received in real time, the structural status information including the bed height, bed surface angle and brake status, for determining whether the preset docking conditions are met;

[0012] If the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, the three-dimensional docking path is dynamically adjusted or the current task is suspended, and the three-dimensional docking path is continued after the path is restored;

[0013] After the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, a docking completion prompt signal is issued;

[0014] After receiving the docking completion prompt signal, the patient transfer operation is performed to transfer the patient from the electric bed to the intelligent transfer vehicle platform.

[0015] By adopting the above technical solution, by collecting the surrounding environment information collected by the positioning sensor group and identifying the positioning identification device on the electric bed, the bed identification information containing position information and posture information can be obtained, which can realize the accurate identification of the position and orientation of the bed by the intelligent transfer vehicle, thereby improving the accuracy of the initial posture assessment of the docking; by fusing the above identification information, the intelligent transfer vehicle's own posture data and the surrounding environment information to generate a three-dimensional docking path, the path planning can be carried out based on the consideration of the three-dimensional posture, environmental structure and traffic restrictions, thereby improving the spatial adaptability and task executability of the path generation; by controlling the intelligent transfer vehicle to continuously adjust the posture parameters during the path navigation process, it can ensure that the vehicle continuously corrects its own posture during dynamic driving , thereby improving the spatial alignment accuracy between the lathe and the bed; by receiving the structural status information of the electric bed in real time during the navigation movement and judging whether the preset docking conditions are met, the structural safety verification can be performed before the patient is transferred, thereby avoiding incorrect docking when the conditions are not met and improving the overall operation safety; by performing dynamic adjustments or task suspensions on obstacles detected in the path, and continuing navigation after the path is restored, the system's response capability to emergency scenarios can be improved, thereby ensuring the continuity and robustness of path execution; by sending a docking completion prompt signal and starting the patient transfer operation after the preset docking conditions are met and the adjustment is completed, an interactive closed loop between path control and manual operation can be achieved, thereby ensuring the safety and controllability of the patient transfer process.

[0016] In one example, the present application may be further configured as follows: the positioning sensor group on the intelligent transfer vehicle collects the surrounding environment information and identifies the positioning identification device on the electric bed to obtain the identification information of the electric bed, wherein the identification information includes the position information and the posture information.

[0017] Collecting a QR code image on the electric bed, performing image recognition on the QR code image information, extracting the bed identification, bed position information, and orientation angle information encoded in the QR code image, and obtaining an image recognition result;

[0018] Perform multi-point distance measurement on the area where the electric bed is located to construct a spatial depth map for determining the obstacle status and distance margin in front of the electric bed, and obtain a distance measurement result;

[0019] In combination with the environmental camera to collect images and depth data of the surrounding structure, the arrangement of the electric bed relative to the environment is analyzed to determine the boundary of the passable area;

[0020] Based on the image recognition result and the distance measurement result, the position information and posture information in the recognition information are constructed for subsequent path planning.

[0021] By adopting the above technical solution, by collecting QR code images and performing image recognition, the bed identification, location information and orientation angle in the code are extracted, which can achieve low-cost and accurate recognition of the static posture of the bed, thereby avoiding the risk of misidentification caused by error accumulation in traditional positioning methods; by constructing a spatial depth map through multi-point distance measurement of the area where the bed is located, it can intuitively reflect the status of the traversable space in front of the bed, thereby assisting in obstacle avoidance and steering decisions during path generation; by collecting surrounding structure images and depth data to analyze the relative layout of the bed and determine the boundaries of the traversable area, it can enhance the overall understanding of the environmental constraints of the bed, thereby providing a boundary basis for path planning.

[0022] In one example, the present application may be further configured as follows: collecting the posture data of the intelligent transfer vehicle includes:

[0023] The inertial measurement unit (IMU) configured on the intelligent transfer vehicle is used to collect acceleration data and angular velocity data of the vehicle body during navigation. The IMU includes a three-axis accelerometer and a three-axis gyroscope, which are used to obtain linear acceleration and angular velocity information of the vehicle body along the X, Y, and Z axes respectively.

[0024] The acceleration data and angular velocity data are fused and filtered based on the Kalman filter algorithm to remove instantaneous noise and zero drift error in the attitude estimation process;

[0025] The pitch angle, roll angle and yaw angle of the intelligent transfer vehicle are calculated based on the fused data and used as the posture data of the intelligent transfer vehicle for use in the posture planning part of the three-dimensional docking path.

[0026] By adopting the above technical solution, the acceleration and angular velocity data of the intelligent transfer vehicle during navigation are collected by configuring an inertial measurement unit, so that the real dynamic response of the vehicle during movement can be obtained, which is used for subsequent attitude solution and path control; by fusing sensor data through Kalman filtering to remove instantaneous noise and zero drift error, the stability and accuracy of the attitude solution results can be improved, thereby enhancing the reliability of attitude control; by calculating the pitch angle, roll angle and yaw angle as attitude input for use in three-dimensional path attitude planning, it is possible to realize the linkage planning of spatial path and attitude, thereby improving the spatial accuracy of path guidance.

[0027] In one example, the present application may be further configured as follows: fusing the identification information, the posture data of the intelligent transfer vehicle, and the surrounding environment information to generate a three-dimensional docking path for docking includes:

[0028] The bed position information and orientation angle information in the identification information, the pitch angle, roll angle and yaw angle parameters in the self-posture data, and the relative position of obstacles and the boundary data of the pass area in the surrounding environment information are input into the path generation module;

[0029] Based on the space vector modeling logic, a relative posture matrix between the electric bed and the intelligent transfer vehicle is constructed, wherein the relative posture matrix represents the displacement vector and angle difference between the two in a three-dimensional coordinate system;

[0030] The three-dimensional docking path is calculated and generated by a dynamic window path planning algorithm. The three-dimensional docking path includes a lateral movement section, a longitudinal feed section, and a height / posture compensation section, which are used to guide the intelligent transfer vehicle to complete the docking action.

[0031] By adopting the above technical solution, by inputting the recognition information, its own posture parameters and environmental boundary data into the path generation module, a multi-source fusion input system for path generation can be constructed, thereby improving the integrity of the path generation decision; by constructing a relative posture matrix through spatial vector modeling, reflecting the relative displacement and angle difference of the lathe and machine in three-dimensional space, it can provide a clear spatial target difference basis for path docking, thereby improving the targetedness of path planning and terminal consistency; by calculating the horizontal, vertical and height / posture compensation path segments through the dynamic window path planning algorithm, it can solve the path requirements of different dimensions in a targeted manner, thereby improving the overall path execution efficiency and terminal docking quality.

[0032] In one example, the present application may be further configured as follows: generating the three-dimensional docking path by calculating the dynamic window path planning algorithm includes:

[0033] Constructing a two-dimensional velocity space window based on the current position coordinates, current posture angle and kinematic constraint parameters of the intelligent transfer vehicle;

[0034] In the two-dimensional velocity space window, for each combination of linear velocity and angular velocity, trajectory prediction simulation is performed based on the forward kinematics model to generate corresponding local candidate trajectories, each of which represents a sequence of reachable positions of the intelligent transfer vehicle within a preset short-term path planning cycle;

[0035] Determine, based on the position information and orientation angle information of the electric bed in the identification information, a planned target posture point of the three-dimensional docking path as a path guidance reference point;

[0036] For each local candidate trajectory, the trajectory offset, trajectory smoothness index, execution speed efficiency parameter, and minimum safe distance from the identified obstacles are calculated, and a path scoring function for path quality evaluation is constructed.

[0037] The various indicators of the path scoring function are weighted and synthesized according to the preset weighting coefficients, and the speed combination with the highest comprehensive score is selected as the linear speed and angular speed instructions for driving the intelligent transfer vehicle in the current control cycle;

[0038] Inversely deriving a path segment for navigation execution based on the selected speed combination, and executing the path segment as a continuous navigation path in the three-dimensional docking path;

[0039] The path segments corresponding to the selected speed combinations in multiple continuous control cycles are spliced ​​in an execution order to form the three-dimensional docking path.

[0040] By adopting the above technical solution, by constructing a two-dimensional speed space window to limit the candidate range of speed combinations, it is possible to ensure that path planning complies with vehicle dynamic constraints, thereby ensuring the feasibility of path execution and control safety; by simulating the forward kinematic model to generate local candidate trajectories, the motion results corresponding to each speed combination can be predicted in advance, thereby improving the foresight of path planning; by calculating the trajectory offset, smoothness, execution efficiency and obstacle avoidance safety to construct a scoring function, it is possible to evaluate the quality of the path in multiple dimensions, thereby improving the rationality of the final path segment selection; by selecting the speed combination with the highest comprehensive score to generate path segments and continuously splicing them to form a three-dimensional path, the path execution can be made continuous and stable, thereby improving the docking accuracy and dynamic adaptability.

[0041] In one example, the present application may be further configured as follows: if the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, dynamically adjusting the three-dimensional docking path or suspending the current task includes:

[0042] During the three-dimensional docking path navigation process performed by the intelligent transfer vehicle, obstacle information of the area ahead of the navigation path is collected in real time, wherein the obstacle information includes a minimum distance measurement value and a lateral offset angle of the obstacle relative to the intelligent transfer vehicle;

[0043] Comparing the obstacle information with a preset navigation safety distance threshold, and when the minimum distance measurement value is lower than the preset navigation safety distance threshold, triggering obstacle status judgment and entering a path adjustment process;

[0044] When an obstacle state is triggered, the two-dimensional velocity space window is reconstructed, velocity combinations pointing in the direction of the current obstacle are eliminated, and trajectory simulation, path scoring, and path segment selection operations are re-executed based on the remaining velocity combinations to generate new executable path segments;

[0045] If the obstacle state is repeatedly triggered within multiple consecutive path planning control cycles and no valid path segment is available after multiple speed combination reconstructions, the current navigation task state is switched to "path interruption waiting state" and the execution process of the current 3D docking path is suspended;

[0046] When the obstacle has been removed from the currently executed path segment in the three-dimensional docking path, or the minimum distance measurement value of the obstacle is again higher than the preset navigation safety distance threshold, the speed space construction and path segment generation operations are restarted to resume the subsequent navigation tasks of the three-dimensional docking path.

[0047] By adopting the above technical solution, by real-time collection of obstacle distance and offset angle information in the area ahead of the navigation path, the state of obstacles ahead can be perceived in real time, thereby providing input basis for path adjustment; by triggering obstacle state judgment by comparison with the preset navigation safety distance, potential risks can be responded to in a timely manner within a safe range, thereby enhancing the system's obstacle avoidance reaction capability; by eliminating obstacle direction and speed combinations and reconstructing path segments, the safety and dynamic responsiveness of path obstacle avoidance execution can be ensured; when multi-cycle reconstruction is invalid, the task is suspended and the path interruption waiting state is entered, which can avoid the risk of path loss of control caused by blind navigation, thereby ensuring the safety and controllability of navigation behavior; when the obstacle is removed, the path segment generation process is restarted, which can achieve task recovery and path continuity, thereby avoiding navigation termination caused by task interruption.

[0048] In one example, the present application may be further configured as follows: after the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, the docking completion prompt signal is sent, including:

[0049] After the three-dimensional docking path navigation is completed, the current posture parameters of the intelligent transfer vehicle and the structural status information of the electric bed are collected, the posture parameters including the final pitch angle, roll angle and yaw angle, and the structural status information including the bed height, bed surface angle and brake status;

[0050] Comparing the posture parameter with a preset posture threshold value to determine whether the spatial alignment error between the intelligent transfer vehicle and the electric bed meets the posture threshold range;

[0051] Comparing the structural state information with a preset docking safety condition to determine whether the structural state information satisfies the preset docking safety condition;

[0052] When the spatial alignment error satisfies the posture threshold range and the structural state information satisfies the preset docking safety condition, the docking completion prompt signal is generated to notify the operator to start the patient transfer operation.

[0053] By adopting the above technical solution, by collecting the terminal posture of the intelligent transfer vehicle and comparing it with the structural status information of the bed, an accuracy assessment mechanism can be established before the docking is completed, thereby ensuring that the spatial alignment meets the preset requirements; by judging whether the brake status, bed height, bed surface angle, etc. meet the safety conditions, multiple confirmations of the physical structure status can be achieved, thereby avoiding performing patient transfer operations in unsafe conditions; by generating a docking prompt signal after the posture and structural conditions are met, a logical interaction point between system control and manual operation can be established, thereby improving the safety and collaborative efficiency of the transfer process.

[0054] The second object of the present invention is achieved through the following technical solutions:

[0055] A positioning system for an intelligent transfer vehicle and an electric bed in linkage, the positioning system for an intelligent transfer vehicle and an electric bed in linkage comprising:

[0056] A data acquisition module is used to collect surrounding environment information through the positioning sensor group on the intelligent transfer vehicle and identify the positioning identification device on the electric bed to obtain identification information of the electric bed, wherein the identification information includes position information and posture information;

[0057] A path generation module is used to collect the posture data of the intelligent transfer vehicle, fuse the identification information, the posture data of the intelligent transfer vehicle and the surrounding environment information, and generate a three-dimensional docking path for docking;

[0058] A navigation module is used to control the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjust the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed;

[0059] a judgment module, configured to receive, in real time, structural status information of the electric bed during navigation and movement along the three-dimensional docking path, the structural status information including bed height, bed surface angle, and brake status, for determining whether a preset docking condition is met;

[0060] An adjustment module is configured to dynamically adjust the three-dimensional docking path or suspend the current task if the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, and continue to execute the three-dimensional docking path after the path is restored;

[0061] A docking prompt module is used to send a docking completion prompt signal after the preset docking conditions are met and the intelligent transfer vehicle is adjusted;

[0062] The transfer module is used to perform a patient transfer operation after receiving the docking completion prompt signal, and transfer the patient from the electric bed to the intelligent transfer vehicle platform.

[0063] By adopting the above technical solution and setting up a data acquisition module, a path generation module, a navigation module, a judgment module, an adjustment module, a docking prompt module and a transfer module, it is possible to form a functional closed loop of perception recognition, path planning, state judgment, obstacle avoidance control and patient transfer operations, thereby building an integrated medical transfer control system with automatic navigation, precise docking and intelligent collaboration capabilities, and improving the intelligence level, safety and automation level of transfer tasks in medical environments.

[0064] In summary, this application has the following beneficial technical effects:

[0065] 1. By collecting environmental information collected by the positioning sensor group and identifying the positioning markers on the electric bed, bed identification information containing position and posture information is obtained. This enables the intelligent transfer vehicle to accurately identify the bed's position and orientation, thereby improving the accuracy of the initial docking posture assessment. By integrating this identification information, the intelligent transfer vehicle's own posture data, and the surrounding environmental information to generate a three-dimensional docking path, path planning can be performed based on the consideration of the three-dimensional posture, environmental structure, and traffic restrictions, thereby improving the spatial adaptability and task executability of the path generation.

[0066] 2. By controlling the intelligent transfer vehicle to continuously adjust its posture parameters during path navigation, the vehicle can continuously correct its own posture during dynamic driving, thereby improving the spatial alignment accuracy between the lathe and the bed; by receiving the structural status information of the electric bed in real time during the navigation movement and judging whether the preset docking conditions are met, the structural safety verification can be performed before the patient is transferred, thereby avoiding incorrect docking when the conditions are not met and improving the overall operation safety; by performing dynamic adjustments or task suspension on obstacles detected in the path and continuing navigation after the path is restored, the system's response capability to sudden scenarios can be improved, thereby ensuring the continuity and robustness of path execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a flow chart of a positioning method for linking an intelligent transfer vehicle with an electric hospital bed in one embodiment of the present application;

[0068] Figure 2 This is a flowchart for implementing step S10 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0069] Figure 3 This is a flowchart for implementing step S20 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0070] Figure 4 This is another implementation flowchart of step S20 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0071] Figure 5 This is a flowchart for implementing step S26 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0072] Figure 6 This is a flowchart for implementing step S50 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0073] Figure 7 This is a flowchart for implementing step S60 in a positioning method for linking an intelligent transfer vehicle with an electric bed in one embodiment of the present application;

[0074] Figure 8 This is a principle block diagram of a positioning system for linking an intelligent transfer vehicle with an electric hospital bed in one embodiment of the present application. DETAILED DESCRIPTION

[0075] The present application is further described in detail below with reference to the accompanying drawings.

[0076] In one embodiment, if Figure 1As shown, the present application discloses a positioning method for linking an intelligent transfer vehicle with an electric hospital bed, which specifically includes the following steps:

[0077] S10: The surrounding environment information is collected through the positioning sensor group on the intelligent transfer vehicle, and the positioning identification device on the electric bed is identified to obtain identification information of the electric bed, where the identification information includes position information and posture information.

[0078] Specifically, the structural image data and QR code feature image in the field of view in front of the intelligent transfer vehicle are obtained, and the image processing algorithm is used to perform contour enhancement, edge extraction and shape matching operations on the collected video frames. The positioning QR code area pasted on the electric bed is locked, the positioning QR code information is decoded, and the bed's unique identification number, current positioning coordinates and forward orientation angle parameters contained therein are extracted and cached as the identification information of the electric bed. At the same time, the ultrasonic ranging task is performed after the visual inspection is completed. The passable depth range in front of the bed is estimated through the multi-point ranging results and a depth image structure is formed to assist in subsequent path judgment and planning.

[0079] S20: collecting the posture data of the intelligent transfer vehicle, fusing the recognition information, the posture data of the intelligent transfer vehicle and the surrounding environment information, and generating a three-dimensional docking path for docking.

[0080] Specifically, the inertial measurement unit (IMU) interface is called to obtain the acceleration data and angular velocity data of the current transfer vehicle along the three axes, and the above data are fused in real time through the Kalman filter method to estimate the pitch angle, roll angle and yaw angle parameters of the current vehicle posture. At the same time, the cached bed position information and posture information are combined and used as the target reference posture. The path planning interface is called through the path construction engine, and the spatial posture relationship between the current position, current position posture and target bed is used as input parameters to perform the trajectory calculation task in three-dimensional space, forming a path structure containing multiple transition segments, where each path records a triplet of position control points, expected posture angles and speed instructions, which are finally combined to generate a continuous and feasible three-dimensional docking path.

[0081] S30: Control the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjust the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed.

[0082] Specifically, the path segment data in the three-dimensional docking path is parsed segment by segment, and the linear velocity and angular velocity control instructions are generated according to the spatial error between the current vehicle posture state and the next control point. The instructions drive the steering motor and the wheel motor to respond jointly. As the vehicle moves along the path, the posture control algorithm will continuously compare the difference between the current IMU feedback posture and the expected posture in the path segment. If errors are found in pitch, roll or yaw, the differential value of the motor output is adjusted to achieve angle correction. As the control loop is iteratively executed, the intelligent transfer vehicle will gradually approach the electric bed and achieve the posture overlap target, forming a spatial alignment state between the lathes.

[0083] S40: During navigation and movement along the three-dimensional docking path, structural status information of the electric bed is received in real time. The structural status information includes bed height, bed surface angle, and brake status, and is used to determine whether the preset docking conditions are met.

[0084] Specifically, a communication connection is established with the electric bed, and status query instructions are continuously sent during the movement of the transfer vehicle, and real-time feedback data is received from the bed end. The feedback information includes the position feedback of the current bed lifting motor, the inclination encoder data of the bed adjustment mechanism, and the status value of the brake control signal. The above three data are mapped to bed height, bed angle and brake status parameters respectively, and logically compared with the preset docking condition values. When the system detects that all three states are within the allowable range, for example, the bed height is between 650mm and 700mm, the bed angle is less than 3 degrees, and the brake signal is in a locked state, the internal flag is updated to "structural status satisfied", otherwise it is maintained in the "unsatisfied" state and continues to monitor.

[0085] S50: If the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, the three-dimensional docking path is dynamically adjusted or the current task is suspended, and the three-dimensional docking path is continued after the path is restored.

[0086] Specifically, the laser ranging device and depth camera are continuously called to obtain the obstacle point cloud or distance image within 2 meters in front of the vehicle in each control cycle. When the distance of a certain area is found to be less than the set safety threshold (for example, 500mm) and coincides with the current driving path direction, it is immediately determined that there is an obstacle and the path execution status is switched to "abnormal". The path correction engine then reconstructs the current speed feasible space, eliminates the path segments that may cause collisions, and then regenerates the local trajectory segments based on the remaining speed combination and attempts to resume navigation. If no valid trajectory is formed after three consecutive rounds of construction, the navigation task is suspended and the vehicle is kept stationary. After the obstacle area is cleared, the path generation and execution process is restarted to ensure navigation continuity and obstacle avoidance safety.

[0087] S60: After the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, a docking completion prompt signal is issued.

[0088] Specifically, after the navigation path is executed, the current vehicle posture data and the bed structure status cache value are immediately obtained, and the posture error evaluation function is called to compare whether the deviation between the current posture angle and the path end posture is within the allowable range, for example, the three-axis posture deviation is less than 2 degrees. At the same time, it is confirmed that the bed structure status in the previous step S40 has been marked as "structural status satisfied". If both conditions are met, the "docking achieved" flag is set to True, and then the prompt control module is called to send a docking completion notification to the control interface, and the buzzer or light module is triggered to send a prompt signal to prompt the nursing staff to perform the patient transfer operation.

[0089] S70: After receiving the docking completion prompt signal, perform the patient transfer operation to transfer the patient from the electric bed to the intelligent transfer vehicle platform.

[0090] Specifically, after the docking completion prompt is issued, the transfer preparation state is entered, and the authorization signal for the transfer action is triggered after waiting for confirmation from the nursing staff. Then, the transfer slide or lifting platform on the vehicle is controlled to align with the docking area of ​​the bed, and the push motor is started to slowly slide the patient pallet from the side of the bed into the transfer vehicle platform. During the process, the load changes are monitored in real time through the pressure sensor installed under the pallet, and the motor propulsion speed is dynamically adjusted through the speed regulation module to ensure stability. After the transfer is completed, the pallet is locked to the fixed structure of the vehicle platform, and the transfer status is set to "completed". The subsequent executable task queue is updated and the evacuation process preparation stage is entered.

[0091] By adopting the above technical solution, by collecting the surrounding environment information collected by the positioning sensor group and identifying the positioning identification device on the electric bed, the bed identification information containing position information and posture information can be obtained, which can realize the accurate identification of the position and orientation of the bed by the intelligent transfer vehicle, thereby improving the accuracy of the initial posture assessment of the docking; by fusing the above identification information, the intelligent transfer vehicle's own posture data and the surrounding environment information to generate a three-dimensional docking path, the path planning can be carried out based on the consideration of the three-dimensional posture, environmental structure and traffic restrictions, thereby improving the spatial adaptability and task executability of the path generation; by controlling the intelligent transfer vehicle to continuously adjust the posture parameters during the path navigation process, it can ensure that the vehicle continuously corrects its own posture during dynamic driving , thereby improving the spatial alignment accuracy between the lathe and the bed; by receiving the structural status information of the electric bed in real time during the navigation movement and judging whether the preset docking conditions are met, the structural safety verification can be performed before the patient is transferred, thereby avoiding incorrect docking when the conditions are not met and improving the overall operation safety; by performing dynamic adjustments or task suspensions on obstacles detected in the path, and continuing navigation after the path is restored, the system's response capability to emergency scenarios can be improved, thereby ensuring the continuity and robustness of path execution; by sending a docking completion prompt signal and starting the patient transfer operation after the preset docking conditions are met and the adjustment is completed, an interactive closed loop between path control and manual operation can be achieved, thereby ensuring the safety and controllability of the patient transfer process.

[0092] In one embodiment, if Figure 2 As shown, in step S10, the positioning sensor group on the intelligent transfer vehicle collects the surrounding environment information and identifies the positioning identification device on the electric bed to obtain the identification information of the electric bed. The identification information includes position information and posture information, specifically including:

[0093] S11: Collecting a QR code image on the electric bed, performing image recognition on the QR code image information, extracting the bed identification, bed position information, and orientation angle information encoded in the QR code image, and obtaining an image recognition result.

[0094] Specifically, after collecting image data in the front-view screen, the candidate extraction and structure verification of the QR code pattern area in the image are performed, the edge position of the QR code is extracted by binarization enhancement processing and border detection of the feature area, and then the internal pattern of the recognition area is decoded to restore the encoded data content. The parsed bed number is used as a unique identifier to mark the target bed entity. At the same time, the spatial posture of the QR code plane in the three-dimensional space is calculated based on the position offset and tilt angle of the QR code in the image frame combined with the camera internal parameters, and the center position information and orientation angle parameters of the bed are extrapolated based on the center position of the QR code combined with the angle. The above image recognition results are recorded as the initial recognition information of the positioning process in the basic input set of the path reasoning stage.

[0095] S12: Perform multi-point distance measurement on the area where the electric bed is located and construct a spatial depth map to determine the obstacle status and distance space margin in front of the electric bed and obtain a distance measurement result.

[0096] Specifically, the multi-point ranging device is controlled to continuously collect distance data of the area in front of the electric bed within a preset angle range. The distance value obtained by each sampling is bound to the corresponding scanning angle to form a spatial polar coordinate data set. After completing continuous collection within a certain angle range, the data set is converted into a two-dimensional depth matrix and a spatial depth map is constructed to visualize the obstacle distribution and the position of the passage gap in the area. The specific distance from the front of the electric bed to the nearest obstacle is calculated according to the minimum distance value represented in the depth map, and the passage margin of the front area is calculated according to the boundary point and the maximum passable angle. When it is detected that the minimum distance is less than the set safety distance threshold, the current area is marked as "obstacle existence state", otherwise it is marked as "sufficient passage state". The distance measurement result will be used for risk assessment and guidance correction before path planning.

[0097] S13: Combined with the environmental camera to collect surrounding structure images and depth data, analyze the layout of the electric bed relative to the environment and determine the boundary of the passable area.

[0098] Specifically, a structural image captured by an environmental camera fixed inside the room is obtained and aligned with the depth data obtained from the perspective of the bed. The environmental structural plan of the area where the bed is located is restored by image stitching, and an environmental grid map is constructed in combination with the spatial entity distance points identified in the depth data. The position distribution of structural obstructions or static equipment in the grid map is compared with the spatial interval between the current position of the bed to deduce the movable physical boundary of the bed. The maximum free path range for the bed to move forward, left, and right is represented in the environmental map as a closed polygonal boundary area. This boundary is used as a safety restriction range in subsequent path construction to exclude inaccessible directions.

[0099] S14: Based on the image recognition result and the distance measurement result, the position information and posture information in the recognition information are constructed for subsequent path planning.

[0100] Specifically, based on the acquired bed identification number, spatial center position and orientation angle, forward space margin data and passable status flag, three-dimensional spatial information fusion is completed within the same time window. The position and posture values ​​obtained by image recognition are used as basic reference points, and the indicators representing the obstacle status and passable margin in the ranging information are used as additional weight parameters. The above multi-source data are assembled into a unified structured identification information data packet, which contains the coordinates, orientation, front obstacle judgment information and passable boundary description of the bed in three-dimensional space. The identification information data packet is used as the target input in the path reasoning stage for function calls such as path starting point and target point setting, posture adjustment judgment and obstacle avoidance direction selection.

[0101] In one embodiment, if Figure 3 As shown, in step S20, the posture data of the intelligent transfer vehicle is collected, which specifically includes:

[0102] S21: Use the inertial measurement unit configured on the intelligent transfer vehicle to collect the acceleration data and angular velocity data of the vehicle body during the navigation process. The inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope, which are used to obtain the linear acceleration and angular velocity information of the vehicle body along the X, Y, and Z axes respectively.

[0103] Specifically, after navigation is initialized, the data collection task is started by calling the inertial measurement interface. Real-time linear acceleration data along the three axes of the vehicle coordinate system is obtained from the acceleration sensing component configured inside the vehicle body. At the same time, the angular velocity information generated by the rotation around the X-axis, Y-axis and Z-axis measured by the gyroscope is synchronously read. The six-dimensional sensor data is cached in the attitude calculation queue at a fixed frequency. Each set of joint sampling values ​​of acceleration and angular velocity is marked with a collection timestamp for subsequent time series processing. The X-axis represents the vehicle's forward motion axis, the Y-axis represents the lateral displacement axis, and the Z-axis represents the reference axis for the vertical acceleration and rotation angle changes. During the collection process, the sensor stability and numerical drift trend can be visualized through a graphical interface.

[0104] S22: Based on the Kalman filter algorithm, the acceleration data and angular velocity data are fused and filtered to remove the instantaneous noise and zero drift error in the attitude estimation process.

[0105] Specifically, the state prediction mechanism is activated in each data sampling cycle to construct a state estimation model of the vehicle's current posture. First, the posture state variables at the previous moment are used as prediction input to estimate the expected acceleration and angular velocity change trends at the current moment. Then, the actual acquired acceleration and angular velocity values ​​are compared to calculate the observation error and establish a covariance update matrix. The predicted state and the observed state are fused through the Kalman filter main function to output the optimal estimation result. In this process, the weighting coefficient is adjusted in real time to cope with the changes in measurement uncertainty under different working conditions. Through continuous iterative filtering, the posture estimation anomalies caused by road vibration, inertia error or sensor drift can be effectively eliminated, and a smooth and continuous posture basic data sequence can be obtained for subsequent solution and call.

[0106] S23: The pitch angle, roll angle, and yaw angle of the intelligent transfer vehicle are calculated based on the fused data and used as the intelligent transfer vehicle's own posture data for use in the posture planning part of the three-dimensional docking path.

[0107] Specifically, the combination of three-axis acceleration and angular velocity data after fusion filtering is read, and the attitude solution process is performed within a unified time window. The sensor data is converted into a triplet of the vehicle's pitch angle, roll angle, and yaw angle at the current time point through the three-dimensional Euler angle transformation model, where the pitch angle is used to describe the tilt state of the vehicle in the front and rear directions, the roll angle is used to indicate the degree of vehicle body roll, and the yaw angle is used as a reference for the direction of the vehicle head relative to the environmental coordinate system. The above attitude angle values ​​are written into the attitude state buffer and transmitted to the path planning module together with the position state as the attitude reference information of the path control point. In the path adjustment and terminal attitude alignment control stage, the attitude angle is used as the judgment and control input reference to achieve precise spatial registration control.

[0108] In one embodiment, if Figure 4 As shown, in step S20, the identification information, the posture data of the intelligent transfer vehicle and the surrounding environment information are integrated to generate a three-dimensional docking path for docking, which specifically includes:

[0109] S24: The bed position information and orientation angle information in the identification information, the pitch angle, roll angle and yaw angle parameters in the self-posture data, and the relative position of the obstacle and the boundary data of the pass area in the surrounding environment information are input into the path generation module.

[0110] Specifically, the bed position information and orientation angle information obtained in the image recognition stage are used as the target posture input, combined with the current vehicle posture parameters derived in the previous inertial solution, including pitch angle, roll angle and yaw angle, and the relative position distribution of environmental obstacles and the boundary wireframe of the passage area collected at the initial moment of navigation are read, and the above multi-source information is structured into a path input data frame in a unified format. The data frame contains the current vehicle posture state in three-dimensional space, the target posture parameters of the bed terminal, the walkable space in the navigation area, and the absolute coordinates of the currently detected obstacles. It is written into the initial input buffer of the path construction logic through a one-time parameter transfer method, providing a complete reference data source for subsequent posture alignment, obstacle avoidance prediction and path segmentation construction.

[0111] S25: Based on the spatial vector modeling logic, a relative posture matrix is ​​constructed between the electric bed and the intelligent transfer vehicle. The relative posture matrix represents the displacement vector and angle difference between the two in the three-dimensional coordinate system.

[0112] Specifically, the central spatial coordinates of the bed are extracted as the target point position and its orientation angle is obtained as the target attitude angle. At the same time, the current position coordinates and current orientation angle triplet of the vehicle are read, and the coordinate difference between the center point of the vehicle and the center point of the bed is calculated as the target displacement vector. The attitude angle difference between the current vehicle orientation and the bed orientation is calculated using a three-dimensional rotation difference model. The above position offset vector and angle difference are combined into a six-dimensional space state vector and expanded into a homogeneous transformation matrix form. The matrix stores the relative position and relative attitude relationship between the bed and the vehicle in a standard 4×4 structure. During the path planning process, the relative position and attitude matrix is ​​used as the target matching condition to verify and adjust the end attitude of each navigation node, so that the end of the path can achieve double closure with the target attitude in position and angle.

[0113] S26: A three-dimensional docking path is calculated and generated through a dynamic window path planning algorithm. The three-dimensional docking path includes a lateral movement segment, a longitudinal feed segment, and a height / posture compensation segment, which is used to guide the intelligent transfer vehicle to complete the docking action.

[0114] Specifically, based on the dynamic window path planning method, a two-dimensional velocity space is constructed and the current position of the vehicle is used as the starting point of the path, and the center posture of the bed is used as the target point of the path. In each path planning control cycle, different linear velocity and angular velocity combinations are iteratively generated to form a candidate local trajectory set. Each trajectory generates a displacement prediction curve for a future period of time through forward motion model simulation. Among all candidate trajectories, their trajectory offset, obstacle spacing and posture change trend are evaluated in turn and path scoring operations are performed. The trajectory with the highest score is selected as the valid path segment in the current control cycle and executed. The path segments generated in multiple cycles are spliced ​​in the navigation order to form a complete path. The front section of the path mainly uses lateral movement to complete the initial parallel alignment with the bed, the middle section performs longitudinal feeding to gradually approach the side of the bed, and the last section contains height matching and pitch and roll angle adjustment logic to complete the posture closure docking, guiding the vehicle platform boundary and the bed platform to maintain strict alignment to ensure the safety and accuracy of the patient transfer operation.

[0115] The implementation of the dynamic window path planning algorithm in path planning is described as follows: in each control cycle, a two-dimensional velocity space window is constructed based on the current position and posture state of the vehicle. This velocity space defines all optional linear velocity and angular velocity combinations. Trajectory simulation operations are performed on each set of velocity combinations in the velocity space to generate a candidate sequence of reachable paths within a short period of time. For each candidate path, its offset distance to the target posture, path smoothness, forward efficiency, and obstacle distance safety value are evaluated in turn, and a comprehensive path score is calculated. After the path scores are sorted, the velocity combination with the highest score is selected for control execution. The path segment corresponding to this velocity combination is reversed as the navigation path of the current cycle, and the path segments in consecutive control cycles are sequentially spliced ​​into a complete three-dimensional docking path.

[0116] In one embodiment, if Figure 5 As shown, in step S26, a three-dimensional docking path is generated by calculating using a dynamic window path planning algorithm, specifically including:

[0117] S261: Construct a two-dimensional velocity space window based on the current position coordinates, current posture angle, and kinematic constraint parameters of the intelligent transfer vehicle.

[0118] Specifically, the velocity space construction function is called to extract the current position coordinates and attitude angle parameters of the intelligent transfer vehicle, including the plane position X and Y coordinates and the current yaw angle θ. The maximum linear velocity, maximum angular velocity, minimum linear velocity, and angular velocity increment preset in the vehicle structure configuration are used as constraint boundaries. A two-dimensional velocity grid window is constructed in the velocity space with linear velocity as the vertical axis and angular velocity as the horizontal axis. For each grid point, a set of available speed combinations is generated as the candidate speed set for this cycle. Speed ​​combinations that are outside the dynamic safety range in the velocity space are marked and eliminated, and only speed combinations that meet the vehicle turning radius limit, acceleration and deceleration capabilities, and minimum safe distance constraints are retained for trajectory generation.

[0119] S262: In the two-dimensional velocity space window, for each combination of linear velocity and angular velocity, trajectory prediction simulation is performed based on the forward kinematics model to generate corresponding local candidate trajectories. The local candidate trajectories represent a sequence of reachable positions of the intelligent transfer vehicle within a preset short-term path planning cycle.

[0120] Specifically, for each set of linear velocity and angular velocity parameters in the velocity space, the differential kinematics model is used to calculate the vehicle's position evolution at consecutive moments within a preset time window Δt. The displacement change in each small time step is recursively estimated based on the vehicle's current posture angle θ and the selected velocity pair (v, ω). The X and Y displacements of each step are accumulated to generate a continuous position sequence. This sequence is recorded as a local candidate trajectory in a list structure and the starting and ending postures of the path are marked. Each trajectory corresponds to a hypothetical driving path for the vehicle in the future for several control cycles. This step does not actually drive the vehicle and is only used for path evaluation.

[0121] S263: Determine the planned target position point of the three-dimensional docking path based on the position information and orientation angle information of the electric bed in the identification information, and use it as a reference point for path guidance.

[0122] Specifically, the center position coordinates and orientation angle data of the electric bed in the world coordinate system are read from the identification information, and the target position coordinates and orientation angle are combined to form the target posture point of the three-dimensional docking path. The target posture point is defined as the spatial position and posture state that the intelligent transfer vehicle ultimately needs to reach. This point serves as the end point reference benchmark for path evaluation, and is used to calculate the error between the trajectory end point and the target posture in subsequent path scoring. It also serves as a global constraint condition for the navigation path to ensure that the end point posture and spatial position of the navigation process meet the accuracy requirements.

[0123] S264: For each local candidate trajectory, the trajectory offset between the target docking position point, the trajectory smoothness index, the execution speed efficiency parameter, and the minimum safe distance between the local candidate trajectory and the identified obstacle are calculated, and a path scoring function for path quality evaluation is constructed.

[0124] Specifically, a path performance analysis process is performed on each local candidate trajectory. First, the Euclidean distance between the end position of the trajectory and the target pose point is calculated as the trajectory offset. Then, the uniformity of the angle change between each segment is analyzed according to the trend of path curvature change to obtain the trajectory smoothness index. At the same time, the average speed and maximum acceleration changes required for path execution are statistically analyzed to obtain the execution efficiency parameter. Combined with the obstacle map, the minimum distance between the trajectory path point and the obstacle boundary is calculated as the safety margin. Based on the scores of all dimensions, a normalized path scoring function is constructed, and different performance items are synthesized into a single comprehensive score in a weighted linear combination manner for path ranking.

[0125] S265: Perform weighted synthesis on the various indicators of the path scoring function according to the preset weighting coefficients, and select the speed combination with the highest comprehensive score as the linear speed and angular speed instructions for driving the intelligent transfer vehicle in the current control cycle.

[0126] Specifically, all candidate path scoring results are bound to the corresponding speed combinations one by one, and each scoring item is linearly combined according to the set scoring function weighting coefficient to obtain the final score. All speed combinations and corresponding path scores are sorted, and the speed combination with the highest score is selected as the optimal control instruction for controlling the forward direction and angle change of the intelligent transfer vehicle in the current cycle. The linear velocity and angular velocity values ​​in this speed combination are written into the controller instruction buffer as the navigation execution input signal within this cycle, thereby ensuring that the transfer vehicle completes posture matching and path advancement in the optimal direction.

[0127] S266: Inversely derive a path segment for navigation execution based on the selected speed combination, and execute the path segment as a continuous navigation path in the three-dimensional docking path.

[0128] Specifically, based on the trajectory information recorded in the previous path simulation, the path data corresponding to the current optimal speed combination is retrieved from the candidate trajectory set, and all the position points and attitude angle points of the path segment are extracted to form a short-term continuous path segment, which is used as the navigation path of the current control cycle. When entering the path execution phase, the path point sequence is sent to the navigation controller for step-by-step tracking. The path segment is executed as a continuous navigation path in the three-dimensional docking path. During the trajectory tracking process, the vehicle's current position feedback and the path segment target point are continuously read to calculate the deviation and make control corrections to ensure that the vehicle accurately advances along the path.

[0129] S267: splicing the path segments corresponding to the selected speed combinations in multiple continuous control cycles in an execution order to form a three-dimensional docking path.

[0130] Specifically, after the end of each control cycle, the executed path segments of the current cycle are appended to the path record cache. The path segments obtained by sequentially splicing in multiple control cycles are arranged in chronological order to form a complete navigation path sequence. The path sequence is defined as the three-dimensional docking path under the current task, which includes all navigation trajectories between the vehicle's current position and the target bed posture point. After splicing is completed, the path structure can be used for path backtracking, navigation task log generation, docking posture error statistics and other functions, and is also used as input to control the final path closure accuracy judgment.

[0131] In one embodiment, if Figure 6 As shown, in step S50, if the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, the three-dimensional docking path is dynamically adjusted or the current task is suspended, specifically including:

[0132] S51: During the three-dimensional docking path navigation process performed by the intelligent transfer vehicle, obstacle information of the area ahead of the navigation path is collected in real time. The obstacle information includes a minimum distance measurement value and a lateral offset angle of the obstacle relative to the intelligent transfer vehicle.

[0133] Specifically, during the navigation process, the forward ranging sensor continuously scans the obstacle distribution information in the area directly in front of the vehicle, extracts the centroid position coordinates of each obstacle point cloud cluster and performs Euclidean distance calculation with the vehicle's current position to obtain the minimum ranging value. At the same time, based on the vehicle's current posture, the lateral offset angle of the obstacle's centroid in the vehicle coordinate system is calculated. The above two parameters are written into the obstacle perception cache as the obstacle state input of the current period, and are continuously updated for subsequent path evaluation and safety judgment.

[0134] S52: Compare the obstacle information with the preset navigation safety distance threshold. When the minimum distance measurement value is lower than the preset navigation safety distance threshold, the obstacle status judgment is triggered and the path adjustment process is entered.

[0135] Specifically, the minimum ranging value in the obstacle perception cache is read and compared with the preset navigation safety distance threshold. When the ranging value is lower than the threshold, it is judged as a potential collision risk, and the path validity state of the current control cycle is set to an unsafe state and enters the path re-evaluation phase. At the same time, the obstacle avoidance flag is set in the control thread to instruct the subsequent path generation process to enable the obstacle avoidance adjustment logic.

[0136] S53: When the obstacle state is triggered, the two-dimensional velocity space window is reconstructed, the velocity combinations pointing to the direction of the current obstacle are eliminated, and the trajectory simulation, path scoring and path segment selection operations are re-executed based on the remaining velocity combinations to generate a new executable path segment.

[0137] Specifically, after detecting that the obstacle avoidance state is activated, the speed space window reconstruction operation is re-executed. The relative lateral offset angle of the obstacle is read to determine the fan-shaped area in the vehicle coordinate system. The speed combination pointing to the fan-shaped area in the speed space is determined to be a high-risk combination and is eliminated. Subsequently, the trajectory simulation is re-performed based on the remaining speed combinations to generate a set of candidate path segments. The offset error, safety distance and execution efficiency score between each path segment and the target posture are calculated in turn. The path scoring function is constructed to complete the path sorting and the path segment with the highest score is selected as the executable path for the current cycle to replace the original path segment.

[0138] S54: If the obstacle state is repeatedly triggered within multiple consecutive path planning control cycles and no valid path segment is available after multiple speed combination reconstructions, the current navigation task state is switched to the "path interruption waiting state" and the execution process of the current three-dimensional docking path is suspended.

[0139] Specifically, an obstacle status cumulative counter is maintained in the path evaluation module. When the counter continuously records the obstacle avoidance state for multiple control cycles and no safe path segment is generated after all reconstructed speed combinations, the path interruption judgment logic is executed to switch the navigation state to the "path interruption waiting state", synchronously interrupting the current three-dimensional path execution thread and keeping the vehicle's current posture locked. At the same time, a pause flag signal is sent to the upper-level control logic to wait for the obstacle to dynamically disappear before reactivating the path navigation task.

[0140] S55: When the obstacle has been removed from the currently executed path segment in the three-dimensional docking path, or the minimum distance measurement value of the obstacle is higher than the preset navigation safety distance threshold again, the speed space construction and path segment generation operations are restarted to resume the subsequent navigation tasks of the three-dimensional docking path.

[0141] Specifically, obstacle perception operations are continuously performed at a low frequency during the path interruption waiting state. When it is detected that the obstacle target in the currently executed path segment has moved out of the path space or the minimum obstacle distance value is greater than the safety threshold, the obstacle status flag is reset to the non-obstacle avoidance state and the speed space construction function is restarted. At the same time, the path segment generation process is started to recalculate the feasible path and resume the control thread to execute path tracking. The remaining three-dimensional docking path is continued from the interruption point to complete the navigation task closed loop.

[0142] In one embodiment, if Figure 7 As shown, in step S60, that is, after the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, a docking completion prompt signal is issued, which specifically includes:

[0143] S61: After the three-dimensional docking path navigation is completed, the current posture parameters of the intelligent transfer vehicle and the structural status information of the electric bed are collected. The posture parameters include the final pitch angle, roll angle and yaw angle, and the structural status information includes the bed height, bed surface angle and brake status.

[0144] Specifically, after arriving at the navigation endpoint, the vehicle's current posture state is obtained, and the angular velocity integral value of the latest output of the inertial measurement unit is converted into three-axis angle values ​​of pitch angle, roll angle and yaw angle through the attitude solution function. At the same time, the current status feedback data of the bed is requested through the wireless communication link, and the position value of the bed lifting motor encoder, the output angle value of the bed pitch sensor and the status bit of the electronic brake controller in the received data frame are parsed. The three types of data are combined to form a structural status information structure for subsequent docking accuracy judgment module to call.

[0145] S62: Compare the posture parameter with a preset posture threshold to determine whether the spatial alignment error between the intelligent transfer vehicle and the electric bed meets the posture threshold range.

[0146] Specifically, the current pitch, roll and yaw angle values ​​in the attitude parameters are read and the difference calculation is performed with the preset attitude target values. The calculated three-axis angle error values ​​are compared item by item with the angle error thresholds set in the configuration file. When the pitch error is less than the threshold P, the roll error is less than the threshold R, and the yaw error is less than the threshold Y, the attitude alignment status flag is set to "qualified", otherwise it is marked as "misaligned". This flag will be used together with the structural status result to determine whether to trigger the docking completion signal output.

[0147] S63: Compare the structural state information with the preset docking safety condition to determine whether the structural state information meets the preset docking safety condition.

[0148] Specifically, the bed height value in the structural status information is analyzed and compared with the docking height window of the intelligent transfer vehicle platform to determine whether it falls within the tolerance range. The bed surface angle value is further compared to see whether it is within the horizontal posture error of ±3°, and the brake status bit is checked to see whether it is activated. When all three conditions are met, the structural status verification flag is set to "qualified" to jointly determine whether the basic safety conditions for executing patient transfer are met.

[0149] S64: When the spatial alignment error meets the posture threshold range and the structural state information meets the preset docking safety conditions, a docking completion prompt signal is generated to notify the operator to start the patient transfer operation.

[0150] Specifically, when the posture alignment status flag and the structural status verification flag are both in the "qualified" state, the prompt signal generation function is activated and the message content is constructed to include the current timestamp, vehicle location ID and docking status code. The generated prompt signal will be used to prompt the operator in the form of a local sound and light alarm. At the same time, the prompt "Docking completed, please start the transfer" is pushed through the display interface, and the docking success event log is recorded in the background, waiting for manual confirmation input of the subsequent patient transfer operation.

[0151] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0152] In one embodiment, a positioning system for an intelligent transfer vehicle and an electric bed is provided, and the positioning system for an intelligent transfer vehicle and an electric bed corresponds to the positioning method for an intelligent transfer vehicle and an electric bed in the above embodiment. Figure 8 As shown, the positioning system for the intelligent transfer vehicle and electric bed linkage includes a data acquisition module, a path generation module, a navigation module, a judgment module, an adjustment module, a docking prompt module and a transfer module. The functional modules are described in detail as follows:

[0153] A data acquisition module is used to collect surrounding environment information through the positioning sensor group on the intelligent transfer vehicle and identify the positioning identification device on the electric bed to obtain identification information of the electric bed, including position information and posture information;

[0154] The path generation module is used to collect the posture data of the intelligent transfer vehicle, integrate the recognition information, the posture data of the intelligent transfer vehicle and the surrounding environment information, and generate a three-dimensional docking path for docking;

[0155] The navigation module is used to control the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjust the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed;

[0156] A judgment module is used to receive real-time structural status information of the electric bed during navigation and movement along the three-dimensional docking path. The structural status information includes bed height, bed surface angle, and brake status, and is used to determine whether the preset docking conditions are met;

[0157] An adjustment module is used to dynamically adjust the three-dimensional docking path or suspend the current task if the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, and continue the three-dimensional docking path after the path is restored;

[0158] The docking prompt module is used to send a docking completion prompt signal after the preset docking conditions are met and the intelligent transfer vehicle has completed the adjustment;

[0159] The transfer module is used to perform the patient transfer operation after receiving the docking completion prompt signal, and transfer the patient from the electric bed to the intelligent transfer vehicle platform.

[0160] Optionally, the data acquisition module includes:

[0161] The QR code image recognition submodule is used to collect the QR code image on the electric bed, perform image recognition on the QR code image information, extract the bed identification, bed position information and orientation angle information encoded in the QR code image, and obtain the image recognition result;

[0162] The multi-point ranging and mapping submodule is used to perform multi-point distance measurement on the area where the electric bed is located, construct a spatial depth map, and use it to determine the obstacle status and distance margin in front of the electric bed to obtain the ranging result;

[0163] The surrounding environment analysis submodule is used to combine the environmental camera to collect surrounding structure images and depth data, analyze the layout of the electric bed relative to the environment, and determine the boundaries of the passable area;

[0164] The recognition information generation submodule is used to construct the position information and posture information in the recognition information based on the image recognition results and the ranging results, which are used for subsequent path planning.

[0165] Optionally, the path generation module includes:

[0166] The attitude sensing data acquisition submodule is used to collect acceleration data and angular velocity data of the vehicle body during navigation using the inertial measurement unit configured on the intelligent transfer vehicle. The inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope, which are used to obtain linear acceleration and angular velocity information of the vehicle body along the X, Y, and Z axes respectively;

[0167] The attitude data filtering submodule is used to perform fusion filtering on acceleration data and angular velocity data based on the Kalman filter algorithm to remove instantaneous noise and zero drift error during attitude estimation.

[0168] The attitude angle solver module is used to calculate the pitch angle, roll angle and yaw angle of the intelligent transfer vehicle based on the fused data, which is used as the intelligent transfer vehicle's own attitude data for the attitude planning part of the three-dimensional docking path.

[0169] The path data input submodule is used to input the bed position information and orientation angle information in the identification information, the pitch angle, roll angle and yaw angle parameters in the self-attitude data, and the relative position of obstacles and the boundary data of the pass area in the surrounding environment information into the path generation module;

[0170] The posture modeling submodule is used to construct the relative posture matrix between the electric bed and the intelligent transfer vehicle based on the space vector modeling logic. The relative posture matrix represents the displacement vector and angle difference between the two in the three-dimensional coordinate system;

[0171] The preliminary path generation submodule is used to calculate and generate a three-dimensional docking path through a dynamic window path planning algorithm. The three-dimensional docking path includes a lateral movement segment, a longitudinal feed segment, and a height / posture compensation segment, which are used to guide the intelligent transfer vehicle to complete the docking action.

[0172] Optionally, the preliminary path generation submodule includes:

[0173] A speed space construction unit is used to construct a two-dimensional speed space window based on the current position coordinates, current posture angle and kinematic constraint parameters of the intelligent transfer vehicle;

[0174] A trajectory prediction unit is used to perform trajectory prediction simulation based on the forward kinematics model for each combination of linear velocity and angular velocity in the two-dimensional velocity space window, and generate corresponding local candidate trajectories. The local candidate trajectories represent the sequence of reachable positions of the intelligent transfer vehicle within a preset short-term path planning cycle.

[0175] The target posture confirmation unit is used to determine the planned target posture point of the three-dimensional docking path based on the position information and orientation angle information of the electric bed in the identification information, which serves as a reference point for path guidance;

[0176] The trajectory scoring calculation unit is used to calculate the trajectory offset between each local candidate trajectory and the target docking pose point, trajectory smoothness index, execution speed efficiency parameter, and minimum safe distance from the identified obstacles, and construct a path scoring function for path quality evaluation;

[0177] The optimal path selection unit is used to weight and synthesize the various indicators of the path scoring function according to the preset weighting coefficients, and select the speed combination with the highest comprehensive score as the linear speed and angular speed instructions for driving the intelligent transfer vehicle in the current control cycle;

[0178] A path segment back-calculation unit is used to back-calculate a path segment for navigation execution based on the selected speed combination, and execute the path segment as a continuous navigation path in the three-dimensional docking path;

[0179] The complete path splicing unit is used to splice the path segments corresponding to the speed combinations selected in multiple continuous control cycles in an execution order to form a three-dimensional docking path.

[0180] Optionally, the adjustment module includes:

[0181] The obstacle perception submodule is used to collect obstacle information in real time in the area ahead of the navigation path during the intelligent transfer vehicle's three-dimensional docking path navigation. The obstacle information includes the minimum distance measurement value and lateral offset angle of the obstacle relative to the intelligent transfer vehicle;

[0182] The obstacle avoidance judgment submodule is used to compare obstacle information with the preset navigation safety distance threshold. When the minimum distance measurement value is lower than the preset navigation safety distance threshold, the obstacle status judgment is triggered and the path adjustment process is entered;

[0183] The path reconstruction submodule is used to reconstruct the two-dimensional velocity space window when the obstacle state is triggered, eliminate the velocity combinations pointing in the direction of the current obstacle, and re-execute the trajectory simulation, path scoring and path segment selection operations based on the remaining velocity combinations to generate new executable path segments;

[0184] The path interruption management submodule is used to switch the current navigation task state to the "path interruption waiting state" and suspend the execution process of the current 3D docking path if the obstacle state is repeatedly triggered within multiple consecutive path planning control cycles and no valid path segment is available after multiple speed combination reconstructions;

[0185] The navigation recovery judgment submodule is used to restart the velocity space construction and path segment generation operations and resume subsequent navigation tasks of the 3D docking path when the obstacle has moved out of the currently executed path segment in the 3D docking path or the obstacle minimum distance value is higher than the preset navigation safety distance threshold again.

[0186] Optionally, the docking prompt module includes:

[0187] The posture state acquisition submodule is used to collect the current posture parameters of the intelligent transfer vehicle and the structural state information of the electric bed after the three-dimensional docking path navigation is completed. The posture parameters include the final pitch angle, roll angle and yaw angle, and the structural state information includes the bed height, bed surface angle and brake status;

[0188] The posture alignment comparison submodule is used to compare the posture parameters with the preset posture threshold value to determine whether the spatial alignment error between the intelligent transfer vehicle and the electric bed meets the posture threshold range;

[0189] The structural status comparison submodule is used to compare the structural status information with the preset docking safety conditions to determine whether the structural status information meets the preset docking safety conditions;

[0190] The docking completion prompt submodule is used to generate a docking completion prompt signal when the spatial alignment error meets the posture threshold range and the structural status information meets the preset docking safety conditions, so as to notify the operator to start the patient transfer operation.

[0191] Regarding the specific definition of a positioning system for the linkage of an intelligent transfer vehicle and an electric bed, please refer to the definition of a positioning method for the linkage of an intelligent transfer vehicle and an electric bed above, which will not be repeated here. Each module in the above-mentioned positioning system for the linkage of an intelligent transfer vehicle and an electric bed can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0192] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0193] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A positioning method for linking an intelligent transfer vehicle with an electric hospital bed, characterized in that: The positioning method of the intelligent transfer vehicle and the electric hospital bed in linkage includes: The positioning sensor group on the intelligent transfer vehicle collects surrounding environment information and identifies the positioning identification device on the electric bed to obtain identification information of the electric bed, wherein the identification information includes position information and posture information; Collecting the posture data of the intelligent transfer vehicle, fusing the identification information, the posture data of the intelligent transfer vehicle and the surrounding environment information to generate a three-dimensional docking path for docking; The collecting of the posture data of the intelligent transfer vehicle includes: The inertial measurement unit (IMU) configured on the intelligent transfer vehicle is used to collect acceleration data and angular velocity data of the vehicle body during navigation. The IMU includes a three-axis accelerometer and a three-axis gyroscope, which are used to obtain linear acceleration and angular velocity information of the vehicle body along the X, Y, and Z axes respectively. The acceleration data and angular velocity data are fused and filtered based on the Kalman filter algorithm to remove instantaneous noise and zero drift error in the attitude estimation process; The pitch angle, roll angle and yaw angle of the intelligent transfer vehicle are calculated based on the fused data and used as the posture data of the intelligent transfer vehicle for the posture planning part of the three-dimensional docking path; The step of fusing the identification information, the posture data of the intelligent transfer vehicle, and the surrounding environment information to generate a three-dimensional docking path for docking includes: The bed position information and orientation angle information in the identification information, the pitch angle, roll angle and yaw angle parameters in the self-posture data, and the relative position of obstacles and the boundary data of the pass area in the surrounding environment information are input into the path generation module; Based on the space vector modeling logic, a relative posture matrix between the electric bed and the intelligent transfer vehicle is constructed, wherein the relative posture matrix represents the displacement vector and angle difference between the two in a three-dimensional coordinate system; The three-dimensional docking path is calculated and generated by a dynamic window path planning algorithm, wherein the three-dimensional docking path includes a lateral movement section, a longitudinal feed section, and a height / posture compensation section, for guiding the intelligent transfer vehicle to complete the docking action; The calculating and generating the three-dimensional docking path by a dynamic window path planning algorithm includes: Based on the current position coordinates, current posture angle and kinematic constraint parameters of the intelligent transfer vehicle, a two-dimensional velocity space window is constructed, wherein a set of available velocity combinations is generated for each grid point in the two-dimensional velocity space window as each set of linear velocity and angular velocity combinations; In the two-dimensional velocity space window, for each combination of linear velocity and angular velocity, trajectory prediction simulation is performed based on the forward kinematics model to generate corresponding local candidate trajectories, each of which represents a sequence of reachable positions of the intelligent transfer vehicle within a preset short-term path planning cycle; Determine, based on the position information and orientation angle information of the electric bed in the identification information, a planned target posture point of the three-dimensional docking path as a path guidance reference point; For each local candidate trajectory, the trajectory offset between the target docking position, trajectory smoothness index, execution speed efficiency parameter, and minimum safe distance from the identified obstacles are calculated, and a path scoring function for path quality evaluation is constructed. The various indicators of the path scoring function are weighted and synthesized according to the preset weighting coefficients, and the speed combination with the highest comprehensive score is selected as the linear speed and angular speed instructions for driving the intelligent transfer vehicle in the current control cycle; Inversely deriving a path segment for navigation execution based on the selected speed combination, and executing the path segment as a continuous navigation path in the three-dimensional docking path; splicing the path segments corresponding to the selected speed combinations in multiple continuous control cycles in an execution order to form the three-dimensional docking path; Controlling the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjusting the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed; During navigation and movement along the three-dimensional docking path, the structural status information of the electric bed is received in real time, the structural status information including the bed height, bed surface angle and brake status, for determining whether the preset docking conditions are met; If the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, the three-dimensional docking path is dynamically adjusted or the current task is suspended, and the three-dimensional docking path is continued after the path is restored; After the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, a docking completion prompt signal is issued; After receiving the docking completion prompt signal, the patient transfer operation is performed to transfer the patient from the electric bed to the intelligent transfer vehicle platform.

2. The positioning method for the linkage between an intelligent transfer vehicle and an electric hospital bed according to claim 1 is characterized in that: The positioning sensor group on the intelligent transfer vehicle collects the surrounding environment information and identifies the positioning identification device on the electric bed to obtain the identification information of the electric bed, and the identification information includes position information and posture information, including: Collecting a QR code image on the electric bed, performing image recognition on the QR code image information, extracting the bed identification, bed position information, and orientation angle information encoded in the QR code image, and obtaining an image recognition result; Perform multi-point distance measurement on the area where the electric bed is located to construct a spatial depth map for determining the obstacle status and distance margin in front of the electric bed, and obtain a distance measurement result; In combination with the environmental camera to collect images and depth data of the surrounding structure, the arrangement of the electric bed relative to the environment is analyzed to determine the boundary of the passable area; Based on the image recognition result and the distance measurement result, the position information and posture information in the recognition information are constructed for subsequent path planning.

3. The positioning method of the intelligent transfer vehicle and the electric bed according to claim 1 is characterized in that: If the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, the three-dimensional docking path is dynamically adjusted or the current task is suspended, including: During the three-dimensional docking path navigation process performed by the intelligent transfer vehicle, obstacle information of the area ahead of the navigation path is collected in real time, wherein the obstacle information includes a minimum distance measurement value and a lateral offset angle of the obstacle relative to the intelligent transfer vehicle; Comparing the obstacle information with a preset navigation safety distance threshold, and when the minimum distance measurement value is lower than the preset navigation safety distance threshold, triggering obstacle status judgment and entering a path adjustment process; When an obstacle state is triggered, the two-dimensional velocity space window is reconstructed, velocity combinations pointing in the direction of the current obstacle are eliminated, and trajectory simulation, path scoring, and path segment selection operations are re-executed based on the remaining velocity combinations to generate new executable path segments; If the obstacle state is repeatedly triggered within multiple consecutive path planning control cycles and no valid path segment is available after multiple speed combination reconstructions, the current navigation task state is switched to "path interruption waiting state" and the execution process of the current 3D docking path is suspended; When the obstacle has been removed from the currently executed path segment in the three-dimensional docking path, or the minimum distance measurement value of the obstacle is again higher than the preset navigation safety distance threshold, the speed space construction and path segment generation operations are restarted to resume the subsequent navigation tasks of the three-dimensional docking path.

4. The method for positioning a smart transfer vehicle in conjunction with an electric hospital bed according to claim 1, characterized in that: After the preset docking conditions are met and the intelligent transfer vehicle completes the adjustment, a docking completion prompt signal is issued, including: After the three-dimensional docking path navigation is completed, the current posture parameters of the intelligent transfer vehicle and the structural state information of the electric bed are collected, the posture parameters including the final pitch angle, roll angle and yaw angle; Comparing the posture parameter with a preset posture threshold value to determine whether the spatial alignment error between the intelligent transfer vehicle and the electric bed meets the posture threshold range; Comparing the structural state information with a preset docking safety condition to determine whether the structural state information satisfies the preset docking safety condition; When the spatial alignment error satisfies the posture threshold range and the structural state information satisfies the preset docking safety condition, the docking completion prompt signal is generated to notify the operator to start the patient transfer operation.

5. A positioning system for an intelligent transfer vehicle linked to an electric hospital bed, characterized in that: The positioning system for linking an intelligent transfer vehicle with an electric hospital bed comprises: A data acquisition module is used to collect surrounding environment information through the positioning sensor group on the intelligent transfer vehicle and identify the positioning identification device on the electric bed to obtain identification information of the electric bed, wherein the identification information includes position information and posture information; A path generation module is used to collect the posture data of the intelligent transfer vehicle, fuse the identification information, the posture data of the intelligent transfer vehicle and the surrounding environment information, and generate a three-dimensional docking path for docking; A navigation module is used to control the intelligent transfer vehicle to navigate and move according to the three-dimensional docking path, and continuously adjust the posture parameters of the intelligent transfer vehicle during the movement to achieve spatial alignment between the transfer vehicle and the electric bed; a judgment module, configured to receive, in real time, structural status information of the electric bed during navigation and movement along the three-dimensional docking path, the structural status information including bed height, bed surface angle, and brake status, for determining whether a preset docking condition is met; An adjustment module is configured to dynamically adjust the three-dimensional docking path or suspend the current task if the intelligent transfer vehicle detects an obstacle during the execution of the three-dimensional docking path, and continue to execute the three-dimensional docking path after the path is restored; A docking prompt module is used to send a docking completion prompt signal after the preset docking conditions are met and the intelligent transfer vehicle is adjusted; A transfer module is used to perform a patient transfer operation after receiving the docking completion prompt signal, and transfer the patient from the electric bed to the intelligent transfer vehicle platform; The path generation module includes: A posture sensing data acquisition submodule is used to collect acceleration data and angular velocity data of the vehicle body during navigation using an inertial measurement unit configured on the intelligent transfer vehicle. The inertial measurement unit includes a three-axis accelerometer and a three-axis gyroscope, which are used to obtain linear acceleration and angular velocity information of the vehicle body along the X, Y, and Z axes respectively; The attitude data filtering submodule is used to perform fusion filtering on the acceleration data and angular velocity data based on the Kalman filter algorithm to remove instantaneous noise and zero drift error in the attitude estimation process; An attitude angle solver module is used to calculate the pitch angle, roll angle and yaw angle of the intelligent transfer vehicle based on the fused data, and use them as the attitude data of the intelligent transfer vehicle for the attitude planning part of the three-dimensional docking path; a path data input submodule, configured to input the bed position information and orientation angle information in the identification information, the pitch angle, roll angle and yaw angle parameters in the self-attitude data, and the relative position of obstacles and the boundary data of the passable area in the surrounding environment information into the path generation module; A posture modeling submodule is used to construct a relative posture matrix between the electric bed and the intelligent transfer vehicle based on space vector modeling logic, wherein the relative posture matrix represents the displacement vector and angle difference between the two in a three-dimensional coordinate system; A preliminary path generation submodule is used to calculate and generate the three-dimensional docking path through a dynamic window path planning algorithm. The three-dimensional docking path includes a lateral movement segment, a longitudinal feed segment, and a height / posture compensation segment, and is used to guide the intelligent transfer vehicle to complete the docking action. The preliminary path generation submodule includes: A velocity space construction unit is configured to construct a two-dimensional velocity space window based on the current position coordinates, current posture angle, and kinematic constraint parameters of the intelligent transfer vehicle, wherein a set of available velocity combinations is generated for each grid point in the two-dimensional velocity space window as each set of linear velocity and angular velocity combinations; a trajectory prediction unit configured to perform trajectory prediction simulation based on a forward kinematics model for each combination of linear velocity and angular velocity in the two-dimensional velocity space window to generate a corresponding local candidate trajectory, wherein the local candidate trajectory represents a sequence of reachable positions of the intelligent transfer vehicle within a preset short-term path planning period; a target posture confirmation unit, configured to determine, based on the position information and orientation angle information of the electric bed in the identification information, a planned target posture point of the three-dimensional docking path as a path guidance reference point; The trajectory scoring calculation unit is used to calculate the trajectory offset between each local candidate trajectory and the target docking pose point, trajectory smoothness index, execution speed efficiency parameter, and minimum safe distance from the identified obstacles, and construct a path scoring function for path quality evaluation; An optimal path selection unit is used to weight and synthesize the various indicators of the path scoring function according to a preset weighting coefficient, and select the speed combination with the highest comprehensive score as the linear speed and angular speed instructions for driving the intelligent transfer vehicle in the current control cycle; a path segment back-calculation unit, configured to back-calculate a path segment for navigation execution based on the selected speed combination, and execute the path segment as a continuous navigation path in the three-dimensional docking path; The complete path splicing unit is used to splice the path segments corresponding to the speed combinations selected in multiple continuous control cycles in an execution order to form the three-dimensional docking path.

Images