Humanoid robot capable of sweeping floor and control system thereof

Abstract

The invention relates to the technical field of robots, and discloses a humanoid robot capable of sweeping a floor and a control system thereof, and the humanoid robot comprises a cleaning module with a biped walking capability, a multi-degree-of-freedom joint humanoid body, an integrated dust collection device, a rolling brush, a mop and an automatic switching mechanism, a sensing system comprising a laser radar and the like, and a navigation and path planning system based on an SLAM algorithm. And human-computer interaction interfaces such as voice interaction are supported. The control system comprises an embedded processor, a task scheduling unit, a state monitoring unit, a fault processing module and an energy consumption management unit, and can coordinate the work of each module. The robot has the advantages that the problem that an existing sweeping robot is limited in cleaning capacity, and a humanoid robot lacks a cleaning function is solved, organic combination of flexible movement and efficient cleaning is achieved, and the intelligent cleaning requirement under the complex indoor environment is met.

Description

Technical Field

[0001] The present invention relates to the technical field of robots, and in particular to a humanoid robot capable of sweeping the floor and a control system thereof. Background Art

[0002] With the rapid development of artificial intelligence and robotics, service robots are increasingly being used in both home and commercial environments. Traditional sweeping robots are mostly round or square in structure and rely on wheels for floor cleaning. These robots have limited obstacle clearance, making it difficult to navigate obstacles like door thresholds and wires. They also have numerous blind spots and struggle to thoroughly clean areas like corners and under furniture. Furthermore, they offer poor user interaction. Furthermore, the user interface of traditional sweeping robots is simple, limiting users to basic operations. Natural and smooth human-machine interaction and personalized task settings are not possible, leaving room for improvement in user experience.

[0003] Humanoid robots, with their human-like appearance and movement patterns, possess greater environmental adaptability and human-computer interaction capabilities, and are widely used in fields such as nursing and education. However, existing humanoid robots primarily focus on non-cleaning applications and lack designs that integrate efficient cleaning capabilities. Their bipedal walking structure inherently consumes a lot of energy, further impacting their endurance if directly equipped with traditional cleaning equipment. Furthermore, they lack a perception, planning, and execution coordination system optimized for cleaning tasks, making it difficult to automatically adjust cleaning strategies based on floor material and dirt levels. Therefore, a system that combines the flexible movement capabilities of humanoid robots with intelligent cleaning capabilities is urgently needed to improve the efficiency and intelligence of indoor cleaning. Summary of the Invention

[0004] In order to solve the above-mentioned problems, the present invention proposes a sweeping humanoid robot and its control system that solves the problem that the existing sweeping robots have limited cleaning capabilities and humanoid robots lack cleaning functions, realizes the organic combination of flexible movement and efficient cleaning, and meets the intelligent cleaning needs in complex indoor environments.

[0005] In order to solve the above technical problems, the present invention proposes a technical solution: a humanoid robot capable of sweeping the floor, comprising:

[0006] The humanoid robot body has the ability to walk on two legs and is equipped with a humanoid structure with multiple degrees of freedom joints, including a head, torso, arms, and legs. Each joint is driven by a servo motor to achieve flexible movement. The body adopts a modular design and can be equipped with a hand cleaning component according to the usage scenario.

[0007] A cleaning module, integrated into the robot's foot or hand, includes an electric vacuum cleaner, a rotatable roller brush, a wet mop structure, and an automatic switching mechanism for switching cleaning modes according to the floor type;

[0008] A perception system, including lidar, cameras, and infrared sensors, is used for environmental perception, obstacle recognition, and avoidance. The system uses machine learning algorithms to identify floor materials and detect dirt, automatically adjusting cleaning parameters.

[0009] The navigation and path planning system builds environmental maps and plans the optimal cleaning path based on the SLAM algorithm, supporting real-time path optimization to cope with dynamic environmental changes;

[0010] The human-computer interaction interface supports multiple methods such as voice interaction, touch control, and mobile application. Users can set cleaning plans and check cleaning status. The interface also supports personalized voice recognition and natural language dialogue systems.

[0011] A control system for a floor-sweeping humanoid robot, comprising:

[0012] The embedded processor acts as the control core, coordinating the work of the humanoid robot body, cleaning module, perception system, navigation and path planning system, and human-machine interaction interface;

[0013] Task scheduling unit, used to assign cleaning tasks, plan execution order, and realize multi-module collaborative operation;

[0014] The status monitoring unit monitors the operating status of each module of the robot in real time, collects sensor data and feeds it back to the processor;

[0015] The fault handling module has a self-diagnosis function, which can promptly alarm and trigger the fault-tolerant self-recovery mechanism when a fault occurs;

[0016] The energy consumption management unit optimizes the power distribution of each module and extends the robot's endurance.

[0017] Preferably, in the cleaning module, the electric dust suction device is used to collect dust on the ground, the rotatable roller brush is used to remove stubborn stains, and the wet mop structure is used for wet mopping the ground.

[0018] Preferably, the navigation and path planning system realizes map construction and real-time path optimization based on SLAM algorithm, and is suitable for home office and medical indoor environments.

[0019] Preferably, the bipedal design of the humanoid robot body has the ability to overcome obstacles with a height not exceeding 5 cm.

[0020] Preferably, the laser radar of the perception system is installed on the head, and the camera and infrared sensor are distributed on the torso and feet, forming a 360-degree environmental perception range.

[0021] Preferably, the mobile application of the human-computer interaction interface supports remote real-time monitoring, and users can view the robot's location and cleaning progress through the mobile phone App.

[0022] Preferably, the task scheduling unit can automatically coordinate the working sequence of the cleaning module and the navigation system according to the priority of the cleaning task.

[0023] Preferably, the energy consumption management unit dynamically adjusts the servo motor power so that the robot's endurance time in standard cleaning mode is not less than 90 minutes.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] It realizes the integration of humanoid robots and cleaning functions. The bipedal walking structure is combined with multi-degree-of-freedom joints to improve obstacle crossing ability and environmental adaptability. The modular cleaning components can be flexibly switched according to needs, significantly improving cleaning efficiency and flexibility.

[0026] It has good human-computer interaction capabilities, multiple interaction methods and natural language dialogue systems to improve user convenience and usage experience.

[0027] The multimodal perception system and intelligent navigation algorithm enable it to adapt to a variety of complex environments, expanding the application scenarios of service robots. DETAILED DESCRIPTION

[0028] Example 1

[0029] In this embodiment, the humanoid robot body adopts a humanoid structure design with an overall moderate height and reasonable weight. It is composed of an aluminum alloy frame and a polymer composite shell, which combines lightness and structural strength. The head is integrated with a laser radar and a depth camera for environmental scanning and visual perception. An embedded processor and battery pack are set inside the torso to provide the control core and power source for each module, and hand cleaning component interfaces are reserved on both sides of the torso. The arms have multiple degrees of freedom and are driven by servo motors to achieve flexible movement. The ends are equipped with magnetic interfaces for quick replacement of hand tools such as cleaning brushes and rags. The legs adopt a bipedal bionic ankle joint design. Each leg contains multiple degrees of freedom. Pressure sensors and infrared obstacle avoidance modules are installed on the soles of the feet. The contact area of ​​the single foot is reasonable, which can provide stable support and ensure balance during walking.

[0030] The robot's joints are driven by servo motors, and its inertial measurement unit (IMU) adjusts its center of gravity in real time, enabling smooth movement. In actual testing, it easily crossed thresholds of a certain height and maintained stability on slopes of a certain gradient. Its modular design allows users to choose whether to install a hand cleaning module based on the needs of different cleaning scenarios. When high-altitude cleaning is required, installing the hand cleaning module allows the robot to perform tasks such as wiping desktops and windowsills, effectively expanding its cleaning range.

[0031] The control system adopts a hierarchical control architecture. The bottom-level motion control layer implements closed-loop control of joint torque based on relevant algorithms to ensure a smooth and fluent walking gait; the middle-level path planning layer receives instructions from the navigation system and generates real-time walking paths and joint motion trajectories, enabling the robot to move accurately according to the planned path; the high-level task coordination layer dynamically allocates motion energy consumption and cleaning module power according to the priority of the cleaning task, realizing efficient collaborative execution of multiple tasks.

[0032] Example 2

[0033] In this embodiment, the cleaning module is integrated into the soles of the feet and adopts a drawer-type modular structure. The size of a single module is appropriate and easy to install and replace. The module contains an electric vacuum device, a rotatable roller brush, a wet mop structure and an automatic switching mechanism. The electric vacuum device has a certain power and can generate sufficient suction to effectively collect dust on the ground in conjunction with the filtering device. The rotatable roller brush adopts a mixed soft and hard bristle design and is suitable for removing stubborn stains on floors such as carpets. The wet mop structure has a built-in water tank, and the mop is made of high-quality material. The water output is controlled by a peristaltic pump, which can achieve constant pressure mopping of the floor. The automatic switching mechanism is driven by a micro-stepping motor to drive the gear set, which can complete the switching of different cleaning modes in a short time according to the ground material signal feedback from the sensing system.

[0034] When the perception system detects a hard surface, the control system automatically activates vacuuming and wet mopping mode: the vacuum unit performs a preliminary clean to remove surface dust, followed by a wet mop at an appropriate speed. The mop's pressure sensor adjusts the downward pressure in real time to ensure a thorough cleaning. If carpet is detected, the roller brush module activates, working in conjunction with the vacuum unit for a deep clean. Simultaneously, the navigation system automatically adjusts the cleaning path density, increasing the frequency of cleaning in carpeted areas to ensure no residual dirt.

[0035] In mixed-surface testing, the robot efficiently completed full-area cleaning tasks, achieving significantly better cleaning efficiency than traditional wheeled sweepers. Its high corner cleaning coverage effectively addresses the inadequate cleaning of corners by traditional robots. It is suitable for a variety of floor materials and complex indoor environments, demonstrating excellent environmental adaptability and cleaning capabilities.

[0036] Example 3

[0037] The perception system utilizes a distributed sensor layout, with a lidar sensor mounted on the head. This creates a high-precision point cloud map of the surroundings, enabling a comprehensive scan of the surroundings. A depth camera, also mounted on the head, combines this with a machine learning model to accurately identify obstacles such as tables, chairs, and electrical wires, enabling dynamic obstacle avoidance. Infrared sensors located on the torso and feet detect close-range obstacles, triggering an emergency stop mechanism to ensure safe operation. A floor material recognition algorithm uses a camera to capture floor images. Trained through a neural network, it accurately distinguishes between various floor materials and automatically adjusts cleaning parameters based on the recognition results.

[0038] The navigation system builds an environmental map based on relevant algorithms and supports real-time updates. During cleaning tasks, the robot first performs global planning, generating a reasonable main cleaning path based on the room layout, prioritizing open areas. When narrow areas such as under furniture are detected, local optimization mode is activated, allowing the robot to perform detailed cleaning of corners through small foot movements and trunk rotations. If a sudden obstacle is detected during cleaning, the path planning system can quickly regenerate a detour, achieving efficient dynamic obstacle avoidance.

[0039] The control system uses an embedded processor to enable coordinated operation of various modules. The task scheduling unit automatically coordinates the operating sequence of cleaning modules and the navigation system based on cleaning task priority, ensuring effective cleaning of key areas. The energy management unit optimizes power distribution among modules and extends the robot's battery life by dynamically adjusting servo motor power. The human-machine interface is rich in functions. Users can use voice commands on the touch screen or remotely control the robot via a mobile app, viewing real-time status information such as the robot's position and cleaning progress. The interface's natural language dialogue function meets users' needs for task queries and status feedback, providing a convenient and user-friendly interactive experience.

[0040] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A humanoid robot capable of sweeping the floor, characterized in that: include: The humanoid robot body has the ability to walk on two legs and is equipped with a humanoid structure with multiple degrees of freedom joints, including a head, torso, arms, and legs. Each joint is driven by a servo motor to achieve flexible movement. The body adopts a modular design and can be equipped with a hand cleaning component according to the usage scenario. A cleaning module, integrated into the robot's foot or hand, includes an electric vacuum cleaner, a rotatable roller brush, a wet mop structure, and an automatic switching mechanism for switching cleaning modes according to the floor type; A perception system, including lidar, cameras, and infrared sensors, is used for environmental perception, obstacle recognition, and avoidance. The system uses machine learning algorithms to identify floor materials and detect dirt, automatically adjusting cleaning parameters. The navigation and path planning system builds environmental maps and plans the optimal cleaning path based on the SLAM algorithm, supporting real-time path optimization to cope with dynamic environmental changes; The human-computer interaction interface supports multiple methods such as voice interaction, touch control, and mobile application. Users can set cleaning plans and check cleaning status. The interface also supports personalized voice recognition and natural language dialogue systems.

2. A control system for a floor-sweeping humanoid robot, characterized in that: The system is used to control the humanoid robot according to claim 1, comprising: The embedded processor acts as the control core, coordinating the work of the humanoid robot body, cleaning module, perception system, navigation and path planning system, and human-machine interaction interface; Task scheduling unit, used to assign cleaning tasks, plan execution order, and realize multi-module collaborative operation; The status monitoring unit monitors the operating status of each module of the robot in real time, collects sensor data and feeds it back to the processor; The fault handling module has a self-diagnosis function, which can promptly alarm and trigger the fault-tolerant self-recovery mechanism when a fault occurs; The energy consumption management unit optimizes the power distribution of each module and extends the robot's endurance.

3. The floor-sweeping humanoid robot according to claim 1, characterized in that: In the cleaning module, the electric dust suction device is used to collect dust on the ground, the rotatable roller brush is used to remove stubborn stains, and the wet mop structure is used for wet mopping the ground.

4. The floor-sweeping humanoid robot according to claim 1, wherein: The navigation and path planning system realizes map construction and real-time path optimization based on SLAM algorithm, and is suitable for home, office and medical indoor environments.

5. The floor-sweeping humanoid robot according to claim 1, characterized in that: The bipedal design of the humanoid robot body has the ability to overcome obstacles with a height of no more than 5 cm.

6. The floor-sweeping humanoid robot according to claim 1, characterized in that: The laser radar of the perception system is installed on the head, and the camera and infrared sensors are distributed on the torso and feet to form a 360-degree environmental perception range.

7. The floor-sweeping humanoid robot according to claim 1, characterized in that: The mobile application of the human-machine interaction interface supports remote real-time monitoring, and users can check the robot's location and cleaning progress through the mobile phone app.

8. The control system according to claim 2, characterized in that: The task scheduling unit can automatically coordinate the working sequence of the cleaning module and the navigation system according to the cleaning task priority.

9. The control system according to claim 2, characterized in that: The energy consumption management unit dynamically adjusts the servo motor power so that the robot can last no less than 90 minutes in standard cleaning mode.