All-terrain intelligent wheelchair system with unmanned holding and lifting capability

Abstract

The invention relates to the field of intelligent medical aids, and discloses an all-terrain intelligent wheelchair system with unmanned holding and lifting capability. Core innovations comprise: 1) an autonomous transfer system based on double-mechanical-arm cooperative control and six-foot dynamic support realizes safe transfer of a human body under a load of 80kg; (2) the intelligent cabin integrates a 64-partition pneumatic bag array and 256-point piezoresistive sensing, dynamic pressure regulation is realized through an LSTM algorithm, and the bedsore occurrence rate is greatly reduced in combination with a phase change gel material; and 3) the multi-mode control interface is compatible with ISO 7176-23 physical control and gesture recognition, and can be widely applied to places such as unmanned auxiliary holding and lifting transfer of wheelchair to personnel, hospital ward transfer, long-term bedridden nursing, outdoor emergency rescue and the like.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical assistive devices and rehabilitation engineering, and specifically to an intelligent wheelchair that integrates a dynamic pressure adjustment cabin, a multi-degree-of-freedom robotic arm, a variable topology support system and a fully automatic driving function. Through biomechanical optimization design and AI collaborative control, it achieves the integrated integration of ergonomic support, active health care and complex terrain mobility functions. Background Art

[0002] The existing technology has the following defects: first, passive support defects: traditional wheelchairs use static sponge / fabric seat cushions, which can easily lead to bedsores after long-term use; second, the human-machine interaction is disconnected: the mechanical control and seat system are designed separately, and the closed-loop adjustment of body position-center of gravity-support cannot be achieved; in addition, there is a lack of emergency response capabilities: there is a lack of active posture correction mechanism in the event of sudden tipping. Of course, the most important point is that all types of wheelchairs currently cannot be lifted unmanned, and they all require human help to complete the movement of people entering and exiting the wheelchair.

[0003] The present invention obtains design inspiration from the following technologies: the method for generating the physical function assessment model of Bangbang Robot (CN111415746A) provides the core logic for the present invention to dynamically adjust the pressure according to the human body state; Toyota Textile's 3D knitted breathable material inspires the design of the cabin surface layer; Panasonic Semiconductor's piezoelectric film sensor array is being studied for application in the sitting posture monitoring system. The functional realization of mature industrial robotic arms in lifting and carrying heavy objects provides a technical foundation for application in the field of health care. The only thing currently lacking is the exploration and development of transplanting mature industrial robotic arm technology into the field of residential home health care and hospital bed care. The present invention is committed to solving this innovation problem. Summary of the invention

[0004] The present invention provides an intelligent wheelchair system with an unmanned autonomous lifting function ( Figure 1 ), its hardware architecture design includes four parts: core module, intelligent cockpit system, auxiliary robotic arm system, safety and energy system ( Figure 2 ).

[0005] As shown in Figure 2 above, the core module of the first part consists of an autonomous driving chassis 100 and a variable number of legs support system 104. Autonomous driving chassis 100: four-wheel independent drive, integrated laser radar 102 (360° scanning), binocular camera 101 (obstacle recognition) and millimeter wave radar 103, supporting L4 autonomous driving; the variable number of legs support system includes basic mode and extended mode: four-wheel drive (mobile state) in basic mode; 3-6 mechanical legs 104 are deployed in extended mode to form triangular / quadrangular / hexagonal support to adapt to complex terrain such as slopes and stairs.

[0006] The second part of the intelligent cockpit system consists of a dynamic pressure adjustment seat and an active health management system. The dynamic pressure adjustment seat 200 adopts a layered structure, with the surface layer: antibacterial silver fiber blended 3D knitted fabric 201; the middle layer: a composite layer of shape memory foam and phase change gel; the bottom layer: a 64-zone pneumatic bag array. The dynamic pressure adjustment seat is equipped with an intelligent monitoring module: a flexible piezoresistive sensor array, an infrared body temperature monitoring and humidity sensor. The active health management system is equipped with an AI sitting posture optimization engine: based on the LSTM neural network to predict the pressure distribution trend, the pneumatic bag is linked for dynamic support; emergency posture correction can be performed: when the IMU detects a risk of falling, the seat side airbags will deploy within 0.1 seconds, and the six-legged support lock will be started simultaneously.

[0007] The third part of the auxiliary robot arm system consists of a foldable continuous robot arm 300 and a modular end effector. The foldable continuous robot arm 300: based on the tensegrity structure design, 5 degrees of freedom flexible joints (rotation / pitch / telescopic / deflection / axial rotation) 302, embedded in the wheelchair armrest 301 after folding, with a load of ≥5kg; modular end effector: supports magnetic quick change of gripper 304 (grasping objects), bionic hand (assisting gripping), bionic lifting plate 303 (assisting lifting and support), and infusion stand (medical care). The foldable continuous robot arm 300 is also integrated with a binocular camera 305, a millimeter wave radar 306, and an infrared sensor 307.

[0008] The fourth part, the safety and energy system, consists of dynamic center of gravity monitoring and high-density battery packs, and has been designed for human-machine control compatibility. Dynamic center of gravity monitoring: six-dimensional force sensor + IMU real-time feedback, triggering the support leg to unfold (automatically start the hexagonal support when the tilt angle is greater than 5°); high-density battery pack: 48V / 30Ah lithium battery, battery life ≥8 hours, support wireless charging and solar energy replenishment. The system adopts human-machine control compatibility design and is equipped with a dual-mode control panel: the physical control area retains the manual rocker / physical buttons that comply with the ISO7176-23 standard; intelligent interaction area: the capacitive touch screen supports gesture sliding selection mode; emergency brake fusion: mechanical emergency stop lever and voice command ("stop" Chinese and English recognition) dual redundancy.

[0009] The global workflow diagram of the present invention is shown in Figure 3 After the system is started, there are three modes to choose from: mobile mode, care mode, and terrain mode. In each mode, the chassis drive and the working status of the robotic arm will change accordingly.

[0010] The primary innovation of this invention is to use multi-modal sensing and high-precision force-controlled robotic arms to coordinate ( Figure 4), the whole process of users entering and exiting the wheelchair autonomously is completed without human intervention. Specifically, the first point is human posture recognition: based on the fusion of ToF camera and millimeter wave radar, a three-dimensional skeleton model of the human body is constructed; machine learning algorithm recognizes the intention of action such as "getting up" and "transferring". The third point is safe lifting execution: dual robotic arms synchronously lift the armpit and knee area; the six-legged support system forms a triangular stable base. The third point is dynamic trajectory optimization: robotic arm path planning based on inverse kinematics calculation; the seat height is automatically aligned with the target plane (such as the bed).

[0011] The specific application scenarios of the present invention will cover hospitals, homes, elderly care, and public transportation.

[0012] Full-process nursing care in the hospital: Preoperative transfer: autonomously complete millimeter-level alignment transfer from the bed to the operating table to avoid secondary injury; Postoperative rehabilitation: seat pressure distribution is dynamically adjusted with the wound healing stage; Night care: automatic triggering of turning assistance through thermal imaging monitoring (axial rotation of 10° every 2 hours).

[0013] Barrier-free life at home. Bathroom assistance: The robotic arm coordinates and supports the transfer of the toilet, and the seat is automatically heated and sterilized; kitchen operation: voice control robotic arm grabs high objects, and activates emergency evacuation mode in case of gas leakage.

[0014] Intensive management of nursing homes. Group services: multi-machine formation enables one caregiver to monitor 10 devices (through 5G edge computing); health warning: seat cushion sensors monitor excrement composition.

[0015] Public transportation hub. Platform transfer: automatically matches the height difference of high-speed rail carriages (adjustment range 0-30cm), and completes platform-carriage transfer in 3 seconds; security inspection assistance: the robotic arm assists passengers through the security gate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 : System architecture diagram.

[0017] Figure 2 : Schematic diagram of the system appearance structure. In the figure: 100: autonomous driving chassis; 101, chassis binocular camera; 102, chassis laser radar; 103, bottom millimeter wave radar; 104, mechanical leg; 200, dynamic pressure adjustment seat; 201, antibacterial silver fiber blended 3D knitted fabric; 300, foldable continuous mechanical arm; 301, wheelchair armrest that can be embedded in the mechanical arm; 302, 5-DOF flexible joint (rotation / pitch / telescopic / yaw / axial rotation); 303, bionic lifting board; 304, gripper; 305, mechanical arm binocular camera; 306, mechanical arm millimeter wave radar; 307, mechanical arm infrared sensor.

[0018] Figure 3: Global workflow diagram.

[0019] Figure 4 ; Safe lifting system execution diagram.

[0020] Figure 5 : Flow chart of unassisted lifting and transfer.

[0021] Figure 6 : Long-term bed rest care workflow diagram.

[0022] Figure 7 : Outdoor emergency rescue workflow diagram. DETAILED DESCRIPTION

[0023] Example 1: Unmanned lifting and transfer ( Figure 5 ). Operation process: 1) User voice command triggers the transfer preparation process; 2) The robotic arm and the six-legged support work together to form a stable working platform; 3) The dual robotic arms perform progressive force-controlled lifting according to ISO 13482 standard; 4) The seat and the target plane are aligned at the millimeter level before sliding; 5) The whole process is automatically terminated and restored to a safe posture if the pressure exceeds the standard.

[0024] Example 2: Hospital ward transfer. The workflow is as follows: 1) The nurse sends a transfer command through a tablet computer, and the wheelchair automatically drives to the bedside; 2) The six legs are unfolded to form a stable platform, and the mechanical arms support the patient's armpits and knees; 3) The chassis height is synchronously raised and lowered to the same level as the bed, and the patient is moved to the wheelchair seat; 4) When encountering a threshold on the way, the four corner supports are automatically switched to ensure smooth passage.

[0025] Example 3: Carrying heavy objects at home. The workflow is as follows: 1) The user voice commands "take the milk from the top layer of the refrigerator", the robotic arm unfolds and switches the gripper; 2) The laser radar builds a 3D map of the kitchen and plans an obstacle-avoiding path to the refrigerator; 3) The four legs are locked, and the robotic arm accurately grabs the object and returns it to the user.

[0026] Example 4: Long-term bed rest care ( Figure 6 ). The working process is as follows: 1) The system recognizes that the user has been riding for 2 hours continuously and starts the automatic decompression mode; 2) The pneumatic bag rises and falls in a wave sequence to promote blood circulation; 3) The robotic arm delivers a water cup and assists in adjusting the head tilt angle; 4) When the pressure sensor detects that the pressure of the right ischial tuberosity is greater than 28kPa, the airbag on the left side of the seat inflates to slightly tilt the body 5°, and the stiffness of the supporting leg is adjusted synchronously.

[0027] Example 5: Outdoor emergency rescue ( Figure 7The workflow is as follows: 1) The rescue mode is automatically activated when the user falls into a ravine; 2) The six legs support the ground (grip force ≥ 500N), the robotic arm throws out the first aid kit and fixes the user's torso; 3) The location is sent to the emergency center through the 5G network, and vital signs monitoring (heart rate, blood oxygen) is started simultaneously.

Claims

1. An intelligent wheelchair system with unmanned autonomous lifting function, characterized in that include: Dual-arm collaborative control module: includes a symmetrically arranged 5-DOF robotic arm equipped with a tactile feedback end effector; Hexapod dynamic support system: 3-6 hydraulically driven mechanical legs can be deployed to form a stable base with an anti-overturning moment ≥1500N·m; Intelligent perception system: A three-dimensional environmental modeling module that integrates ToF camera and millimeter-wave radar to achieve skeletal tracking of 17 key points of the human body; Progressive force control algorithm: The contact pressure of the robotic arm is loaded at a controllable rate of 5-10N / s, and the emergency stop protection is activated within 0.1 second when the pressure fluctuation is >±15%.

2. The system according to claim 1, characterized in that: The hexapod support system adopts a topological adaptive structure, which switches to an alternating triangular gait (step length 30 cm ± 2 cm) when climbing stairs, and retracts to a four-wheel drive mode when moving on flat ground.

3. The system according to claim 1, characterized in that: The end effector of the robotic arm integrates: a tactile feedback layer: a piezoresistive film sensor and a vibration motor; and a medical disinfection module: an ultraviolet lamp with a wavelength of 265 nm.

4. The system according to claim 1, characterized in that: The system supports multi-machine formation collaboration, realizes dual wheelchair stretcher transport mode through 5G communication, and adopts federated learning algorithm to optimize group path planning.

5. The system according to claim 1, characterized in that: The smart cockpit system includes: a dynamic pressure regulation layer: a 64-partition pneumatic bag array and a phase change gel composite structure; a health monitoring module: a flexible piezoresistive sensor and an infrared body temperature detection unit.

6. The system according to claim 5, characterized in that: The cockpit is integrated with an energy recovery device, which uses seat vibration to drive the piezoelectric film to generate electricity, preferentially supplying power to the heating pad.

7. The system according to claim 1, characterized in that: The human-computer interaction interface is compatible with: physical control end: emergency stop lever and joystick that comply with ISO 7176-23 standard; intelligent interaction end: supports gesture recognition and brain wave control.

8. The system according to claim 1, characterized in that: The lifting and transfer process includes an automatic alignment function of the seat-target plane, which uses: a laser ranging unit: measuring plane height difference; visual feature matching: ORB feature point extraction algorithm.

9. The system according to claim 8, characterized in that: The alignment process synchronously adjusts the surface hardness of the seat and the target plane to match each other, and controls the inflation amount of the pneumatic bag to make the hardness difference ≤5%, using closed-loop PID control.

Images