Home care robot and system thereof

Abstract

The present invention discloses a home care robot, comprising a bed, a driving mechanism disposed under the bed, a robotic arm connected to the bed for lifting a patient, a bed tilting device for tilting the bed toward the robotic arm to cooperate with the robotic arm in lifting the patient, and a stabilizing mechanism for adjusting the robot's posture to prevent the robot from tipping over. On this basis, a home care robot system is also disclosed, comprising an identification and positioning system, a multi-axis collaborative motion system that can simultaneously control the driving mechanism and extend the robotic arm to lift the fallen person, and a stabilizing system. When the identification and positioning system recognizes that a patient has fallen, the robot can move to the area around the fallen person without collision, and deploy the robotic arm in advance, tilt the bed board, and cooperate with the robotic arm to lift the fallen person onto the bed board. The flexible end effector avoids secondary damage to the fallen person, and the stabilizing system enables the robot to remain stable during the process of lifting the fallen person, thereby achieving fully automatic and highly reliable execution of rescue actions in a home environment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent nursing robots, and in particular to a home nursing robot and a system thereof. Background Art

[0002] With the aging population and strained medical resources, home care robots have evolved from a concept to practical use, entering the intelligent development stage. They can perform tasks such as monitoring vital signs, assisting with getting up, transferring, feeding, administering medication, and conducting rehabilitation training. However, current home care robots are primarily used for basic care or carrying items, with functions focused on voice interaction and remote monitoring. Numerous challenges remain in the field of elderly rescue. Existing technologies still rely on manual labor or simple stretchers, which are inefficient and risky. Existing robots lack specialized transfer structures for collapsed elderly people, making them prone to secondary injuries due to excessive mechanical rigidity during handling and difficult to adapt to elderly people of varying body shapes and postures. Furthermore, given the often diverse and complex nature of home environments, traditional robots rely on simple obstacle avoidance algorithms, which can lead to instability in dynamic path planning and maneuvering across complex terrain. Summary of the Invention

[0003] Purpose of the invention: The purpose of the present invention is to provide a home care robot that avoids secondary injury to the person who has fallen during the transfer process and has high stability during transfer. Another purpose of the present invention is to provide a home care robot system that can safely transfer the person who has fallen.

[0004] Technical solution: The home care robot described in the present invention includes a bed, a driving mechanism arranged under the bed, a robotic arm connected to the bed for lifting a patient, a bed tilting device for tilting the bed toward the side of the robotic arm to cooperate with the robotic arm to lift the patient, and a stabilizing mechanism for adjusting the robot's posture to prevent the robot from tipping over. The robotic arm has an end effector, and the end effector is made of flexible material.

[0005] Furthermore, the bed body is composed of a bed board and an outer frame arranged on the edge of the bed board. The bed board is segmented and divided into three sections. The middle section of the bed board can move up and down, and the bed boards at both ends are fixed.

[0006] Preferably, the bed tilting device is arranged under the bed, including a connecting frame connected to the outer frame at one end and the bed at the other end, a slide rail arranged along the tilting direction of the bed board, a slider used in conjunction with the slide rail and fixedly connected to the back of the bed board at the top, and a slider motor for driving the movement of the slider. The connecting frame can drive the bed to tilt toward the side of the robotic arm.

[0007] Furthermore, the driving mechanism includes a crotch frame disposed below the bed, a rotating crotch symmetrically disposed on both sides of the crotch frame, a leg assembly connected below the rotating crotch, a foot assembly connected below the leg assembly, and a motor module disposed inside the crotch frame. The motor module provides power for the movement of the rotating crotch and leg assemblies. The leg assembly includes an upper arm connected to the rotating crotch, a lower arm rotatably connected below the upper arm, and a first rotating shaft for connecting the upper and lower arms. The foot assembly includes a wheel hub connected to the lower end of the lower arm and used to drive the robot to walk, and a foot motor module disposed inside the wheel hub for controlling the movement of the wheel hub. The wheel hub is connected to the lower arm via a wheel hub rotating shaft. The foot motor module can control the state of the wheel hub.

[0008] Furthermore, the stabilization mechanism includes a three-axis gyroscope arranged on the hip frame and used to monitor the inclination angle of the bed, and a retractable auxiliary arm whose upper end is fixedly connected to the hip frame and whose lower end is fixedly connected to the lower arm. The retractable auxiliary arm keeps the robot stable by controlling the lower arm.

[0009] The home care robot system described in the present invention is applied to a home care robot. The system includes an identification and positioning system for detecting whether a patient has fallen and determining the location of the fall, a multi-axis collaborative motion system that can simultaneously control the drive mechanism and extend the robotic arm to pick up the fallen person, and a stabilization system for maintaining smooth movement of the robot.

[0010] Preferably, the recognition and positioning system constructs a 3D occupancy grid map based on RGB color image and depth image data, and marks obstacles in the map; detects the distribution of key points on the human body, determines the location of the fall and calls for help at the same time; extracts the feasible area within a 1.5m radius centered on the location where the patient falls, and generates a safe operating space boundary.

[0011] Furthermore, the multi-axis collaborative motion system uses a two-layer path planning algorithm of global planning and local planning to navigate to the fallen person. While navigating the path, the drive mechanism and the robotic arm are linked and controlled to trigger the robotic arm to extend before the robot reaches the falling position, and the end effector of the robotic arm is aligned with the normal vector of the human body; the bed tilting device cooperates with the robotic arm to carry the patient onto the bed board.

[0012] Furthermore, when the bed is tilted, the stabilization system controls the wheel hub to detect the ground slip rate. When it detects that the torque fluctuation of the wheel hub exceeds the preset threshold, the foot motor module controls the wheel hub to get stuck. When the bed tilting device retracts to a horizontal position, when the three-axis gyroscope detects that the pitch angle deviation of the bed is greater than the preset threshold, the retractable auxiliary arm compensates to ensure the stability of the robot.

[0013] Beneficial effects: Compared with the existing technology, the present invention has the following significant advantages: (1) In the process of picking up the fallen person, the flexible end effector picks up the fallen person in a retracted manner, with low impact force, and the tilt of the bed enables the robotic arm to place the fallen person steadily on the bed board, avoiding secondary damage to the fallen person; (2) During the process of the bed tilting and the bed returning to the right position, the stabilization system maintains the stability of the robot, thereby improving the execution reliability; (3) During the process of the robot navigating to the vicinity of the fallen person, the robotic arm has been extended and the bed board has been tilted, and the multi-axis collaborative motion system makes the rescue action more efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a structural schematic diagram of the present invention;

[0015] Figure 2 It is a front view of the present invention;

[0016] Figure 3 Schematic diagram of the driving mechanism of the present invention;

[0017] Figure 4 A bottom view of the present invention;

[0018] Figure 5 This is a flowchart of the rescue process of the home care robot in the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0020] refer to Figures 1 to 4 An embodiment of the present invention provides a home care robot, comprising: a bed 1, a drive mechanism 2, a robotic arm 3, a bed tilting device 4, and a stabilizing mechanism. The bed 1 includes a bed board 101, an outer frame 102, a simple controller 103, and a support frame 104. The drive mechanism includes a hip frame 21, a rotating hip 22, a leg assembly 23, a foot assembly 24, and a motor module 25; the leg assembly 23 includes an upper arm 231, a lower arm 232, and a first rotating shaft 233; the foot assembly 24 includes a hub 241, a foot motor module 242, and a hub rotating shaft 243. The robotic arm includes an end effector 301. The bed tilting device includes a connecting frame 401, a slide rail 402, a slider 403, and a slider motor 404. The stabilizing mechanism includes a three-axis gyroscope 501 and a retractable auxiliary arm 502.

[0021] The bed 1's bed deck 101 is segmented. The middle section can be raised, lowered, and tilted by the bed tilting mechanism 4. The two end decks are fixed and house simple controllers 103. A support frame 104 is located between the bed 1 and the drive mechanism 2. The simple controller is connected to a device that monitors vital signs. Once a person is lifted and placed on the bed, while the person is still conscious, the device can be operated to monitor vital signs and provide more meaningful rescue information. An external frame surrounds the bed deck, except for the side where the robotic arm 3 is mounted, to prevent the person from accidentally falling off the bed. There are two robotic arms 3, which are installed on the back of the bed at both ends. The length of the robotic arm can be changed. The other end of the robotic arm is the end effector 301, which is made of flexible material. The robotic arm extends until it reaches the hips of the fallen person. The end effector bends and hugs the fallen person, which can avoid secondary injuries to the fallen person caused by excessive mechanical rigidity, and can adapt to people of different body shapes or postures and successfully pick up the fallen person.

[0022] The middle section of the bed board is raised, lowered, and tilted by a bed tilting device 4. The bed tilting device 4 consists of a connecting frame 401, a slide rail 402, a slider 403, and a slider motor 404. One end of the connecting frame 401 is connected to the lower end of the outer frame, and the other end is connected to the support frame 104. The connecting frames are preferably two symmetrical connecting frames. When receiving a command, the connecting frames drive the bed to tilt downward. There are also two slide rails 402, which are arranged parallel to the lower ends of the bed board. One end of the slide rail is fixedly connected to the back of the bed board, and the other end is arranged along the direction of the bed board's tilt. The slider 403 is used in conjunction with the slide rails. The top of the slider 403 is connected to the middle section of the bed board. A slider motor 404 is located at the fixed end of the slide rail to control the slider's movement speed. When the connecting frame drives the part above the support frame to tilt downward, the slider motor drives the slider to slide down the slide rail, simultaneously driving the middle section of the bed board above the slider downward. When the robotic arm is extended, the middle section of the bed board tilts toward the side of the robotic arm and slowly descends. When the robotic arm embraces the fallen person, it moves the fallen person to the middle section of the bed board until the fallen person is transferred to the middle section of the bed board. The slider motor drives the slider to retract at a fixed speed, and the middle section of the bed board slowly rises back to the right until it is completely horizontally docked with the bed boards at both ends. During this process, the robotic arm still embraces the rescued person, providing stable restraint to prevent the rescued person from subsequently breaking away and falling unconsciously.

[0023] The driving mechanism 2 is used to accurately reach the fallen person before rescue, and to smoothly move the patient to the door after rescue to wait for rescue. The driving mechanism 2 is arranged below the support frame and consists of a hip frame 21, a rotating hip 22, a leg assembly 23, a foot assembly 24, and a motor module 25. The hip frame 21 plays a supporting role and accommodates the integrated control system. The rotating hip 22 is respectively arranged on both sides of the hip frame, with two rotating hips on one side. The rotating hip is connected to the leg assembly 23. The leg assembly includes an upper arm 231 rotatably connected to the rotating hip, a lower arm 232 rotatably connected below the upper arm, and a first rotating shaft 233 for connecting the upper arm and the lower arm. The upper end of the upper arm 231 is fixedly connected to the rotating hip and rotates as the rotating hip rotates. The lower arm can rotate around the first rotating shaft 233. The foot assembly 24 includes a wheel hub 241 connected to the lower end of the lower arm and used to drive the robot's movement. A foot motor module 242 is located within the wheel hub to control its movement. The wheel hub is connected to the lower arm via a wheel hub rotation shaft 243. The foot motor module 242 can control the state of the wheel hub. The upper end of the retractable auxiliary arm 502 is fixedly connected to the hip frame 21, and the lower end is fixedly connected to the middle part of the lower arm, which functions to define the lower arm's movement trajectory.

[0024] The present invention also provides a home care robot system, which is applied to the home care robot mentioned above. The system includes an identification and positioning system for detecting whether a patient has fallen and determining the location of the fall, a multi-axis collaborative motion system that can simultaneously control the drive mechanism and extend the robotic arm to pick up the fallen person, and a stabilization system for maintaining smooth movement of the robot.

[0025] The robot is equipped with binocular cameras for detecting human falls. It uses these cameras to simultaneously capture RGB images (an image format that represents and displays color using red, green, and blue color channels) and a depth point cloud. Using a human pose estimation model (YOLOv7-HumanPose), which combines real-time detection with keypoint localization, the robot detects 17 biomechanical key points on the human body, including core support points such as the cervical spine and sacrum. The model outputs the coordinates of these 17 key points, uses the least squares method to fit the trunk principal axis vector (the line connecting the cervical spine and sacrum), and calculates the spatial angle θ between this axis and the ground normal vector based on binocular disparity data. The ground normal vector is extracted from the depth point cloud using a random sampling consensus algorithm. If θ exceeds 75° and the abnormal posture is confirmed to persist for more than 5 seconds, the system triggers a fall detection and simultaneously calls a pre-set emergency number. Based on RGB color and depth image data, the system uses an improved OctoMap architecture to construct a 3D occupancy grid map with a voxel resolution of 5cm×5cm×5cm. Bayesian filtering is used to update the probability distribution of obstacles in real time, identifying dangerous areas such as broken glass and spilled liquids and effectively filtering out low-risk interference objects. A hemispherical safe operating space with a radius of 1.5m is generated, centered on the center of mass of the fallen person. The robot path is planned using an improved RRT* algorithm. A collision-free path is generated in the configuration space, and a goal-oriented node strategy and an adaptive probabilistic strategy are used to optimize search efficiency, reducing planning time compared to traditional RRT algorithms.

[0026] The multi-axis collaborative motion system uses a two-layer path planning algorithm combining global and local planning to navigate to the fallen patient. It then integrates the local dynamic obstacle avoidance (DWA) algorithm to construct a velocity-space dynamic window and calculate the trajectory evaluation function in real time, balancing path orientation, obstacle avoidance distance, and motion efficiency. When a sudden obstacle, such as a moving pet, is detected in the navigation map, a repulsive field function is used to drive the trajectory correction path. Path data is transmitted to the control system via a bus to ensure the synchronization accuracy of the robotic arm and drive mechanism, ensuring collision-free navigation to the location where the elderly person has fallen. During navigation, the drive mechanism and robotic arm are controlled in a coordinated manner, triggering the robotic arm to pre-extend when the robot approaches 1.2 meters in front of the fallen patient, aligning the robotic arm's end effector with the human body's normal vector. The bed tilting device cooperates with the robotic arm to lift the patient onto the bed.

[0027] When the bed is tilted, the foot motor module controls the wheel hub to detect the ground slip rate. When it is detected that the torque fluctuation of the wheel hub exceeds the preset threshold, the foot motor module controls the wheel hub to get stuck. When the bed tilt device is retracting to a horizontal position, when the three-axis gyroscope detects that the pitch angle deviation of the bed is greater than the preset threshold, the retractable auxiliary arm compensates to ensure the stability of the robot. The connecting frame 401 tilts the bed downward, and the motor module drives the rotating hip section, pushing the upper arm downward. Simultaneously, the retractable auxiliary arm 502 contracts, raising the lower arm around the first rotation axis 233. At this point, the foot motor module controls the wheel hub to touch the ground at 80 rpm to detect the slip rate; when the torque fluctuation exceeds the threshold of ±15%, the foot motor module drives the wheel hub to lock and secure it. Because the rescued person is on the bed board, the bed is prone to instability during the return process, causing it to tip over. At this time, the three-axis gyroscope 501 located in the hip section frame comes into action, setting a pitch angle for the robot. When the three-axis gyroscope detects that the pitch angle deviation of the robot body is greater than the preset value, the retractable auxiliary arm 502 immediately compensates, pressing down against the lower arm 232 to ensure stability.

Claims

1. A home care robot comprising a bed (1), a driving mechanism (2) disposed below the bed, and a mechanical arm (3) connected to the bed for lifting a patient, characterized in that: The invention also includes a bed tilting device (4) for tilting the bed (1) toward one side of the robot arm for cooperating with the robot arm to lift the patient, and a stabilizing mechanism for adjusting the robot posture to prevent the robot from tipping over. The robot arm (3) has an end effector (301), which is made of a flexible material. The bed tilting device (4) is arranged below the bed (1), and includes a connecting frame (401) with one end connected to the outer frame (102) and the other end connected to the bed, a slide rail (402) arranged along the tilting direction of the bed board, a slider (403) used in conjunction with the slide rail and fixedly connected to the back of the bed board at the top, and a slider motor (404) for driving the slider to move. The connecting frame (401) can drive the bed toward one side of the robot arm. The driving mechanism (2) comprises a crotch frame (21) provided below the bed (1), a rotating crotch (22) symmetrically provided on both sides of the crotch frame, a leg assembly (23) connected below the rotating crotch, a foot assembly (24) connected below the leg assembly, and a motor module (25) provided inside the crotch frame (21), wherein the motor module (25) provides power for the movement of the rotating crotch and the leg assembly; the stabilizing mechanism comprises a three-axis gyroscope (501) provided on the crotch frame and used for monitoring the tilt angle of the bed, and a retractable auxiliary arm (502) whose upper end is fixedly connected to the crotch frame and whose lower end is fixedly connected to the lower arm, wherein the retractable auxiliary arm (502) keeps the robot stable by controlling the lower arm.

2. The home care robot according to claim 1, characterized in that The bed body (1) is composed of a bed board (101) and an outer frame (102) arranged at the edge of the bed board, and the bed board (101) is segmented.

3. The home care robot according to claim 1, characterized in that: The leg assembly (23) comprises an upper arm (231) connected to the rotating hip, a lower arm (232) rotatably connected below the upper arm, and a first rotating shaft (233) for connecting the upper arm and the lower arm.

4. The home care robot according to claim 1, characterized in that: The foot assembly (24) includes a wheel hub (241) connected to the lower end of the lower arm and used to drive the robot to walk, and a foot motor module (242) arranged inside the wheel hub and used to control the movement of the wheel hub. The wheel hub is connected to the lower arm via a wheel hub rotating shaft (243), and the foot motor module (242) can control the state of the wheel hub.

5. A home care robot system, applied to the home care robot according to claim 1, characterized in that: The system includes an identification and positioning system for detecting whether a patient has fallen and determining the location of the fall, a multi-axis collaborative motion system that can simultaneously control the drive mechanism and extend the robotic arm to pick up the fallen person, and a stabilization system for maintaining the smooth movement of the robot.

6. The home care robot system according to claim 5, characterized in that: The identification and positioning system constructs a 3D occupancy grid map based on the RGB color image and depth image data, and marks obstacles within the map; Detect the distribution of key points on the human body, determine the location where the patient fell, and call for help at the same time; extract the feasible area within a 1.5m radius centered on the patient's fallen location to generate the safe operating space boundary.

7. The home care robot system according to claim 5, characterized in that: The multi-axis collaborative motion system uses a two-layer path planning algorithm of global planning and local planning to navigate to the fallen person. During the path navigation, the drive mechanism and the robotic arm are linked and controlled to trigger the robotic arm to extend before the robot reaches the falling position. The end effector of the robotic arm is aligned with the normal vector of the human body, and the bed tilting device cooperates with the robotic arm to carry the patient to the bed board.

8. The home care robot system according to claim 5, characterized in that: When the bed is tilted, the foot motor module controls the wheel hub to detect the ground slip rate. When it is detected that the torque fluctuation of the wheel hub exceeds the preset threshold, the foot motor module controls the wheel hub to be stuck. When the bed tilting device is retracted to a horizontal position, when the three-axis gyroscope detects that the pitch angle deviation of the bed is greater than the preset threshold, the retractable auxiliary arm compensates to ensure the stability of the robot.

Images