Intelligent ground garbage cleaning robot

Abstract

The invention relates to an intelligent ground garbage cleaning robot, and belongs to the technical field of service robots and intelligent homes. The ground garbage cleaning robot is of a double-wheel differential driving structure, and universal wheels are arranged at the tail of the ground garbage cleaning robot and used for auxiliary supporting and stable operation. According to the invention, an STM32 single-chip microcomputer is used as a control core, garbage identification is carried out by fusing a Siped MaixCAM visual module, and real-time detection and classification of garbage are realized by combining a lightweight neural network model. A multi-task scheduling architecture and a PID control algorithm are integrated in the robot, dynamic cooperative control over modules such as a left wheel, a right wheel, a dust collection fan and the like is achieved, and a recognition-planning-cleaning closed-loop system control process is formed. Structurally, the streamline dust collection air duct is innovatively designed, the dust collection efficiency can be effectively improved, noise can be effectively reduced, and the streamline dust collection air duct is suitable for efficient recycling of different kinds of garbage. The device is high in recognition precision, compact in structure and fast in control response, and has high environmental adaptability and popularization value.

Description

Technical Field

[0001] The present invention relates to the field of intelligent cleaning equipment, and in particular to an intelligent ground garbage cleaning robot with garbage recognition, path planning and autonomous cleaning capabilities, belonging to the field of service robots and smart home automation control technology. Background Art

[0002] With the increasing popularity of smart home systems, floor-cleaning robots, as typical intelligent terminal devices, have been widely used in indoor scenarios such as homes, offices, and education. Existing floor-cleaning robots mostly rely on technologies such as infrared obstacle avoidance and gyroscopic navigation to achieve autonomous movement and basic cleaning operations. Some products integrate image recognition modules to improve their adaptability to complex floor environments.

[0003] However, in actual application scenarios with diverse garbage types and complex environments, traditional ground garbage cleaning robots still have the following problems: First, the garbage recognition ability is limited, and it is difficult to accurately distinguish common garbage types such as cigarette butts and paper balls. The recognition robustness is unstable under different lighting and background conditions; second, the path planning and cleaning strategies lack coordinated optimization, which can easily lead to path redundancy, high energy consumption or missed cleaning; third, the control system architecture generally adopts a single-threaded or low real-time processing structure, which makes it difficult to achieve efficient linkage control between multiple modules.

[0004] The invention patent application, "Robot Sweeping System and Robot Sweeping," with publication number CN105640442B, designs a robot sweeping system with atmospheric pressure detection and notification capabilities. This system can monitor environmental changes in real time through an atmospheric pressure sensor and issue health reminders to users via voice or mobile devices, enhancing the device's human-computer interaction capabilities. However, this type of technology focuses on environmental information feedback and user emotional intervention, without substantially optimizing core aspects such as waste identification, path planning, and cleaning execution. Its cleaning function still relies on fixed strategies and lacks intelligent identification and response mechanisms, making it difficult to meet the actual needs for efficient cleaning, waste sorting, and autonomous path adjustment in complex environments.

[0005] In addition, the dust suction duct structure design in some existing products is unreasonable, resulting in restricted airflow path, low suction utilization rate, and poor performance in noise control and anti-clogging performance, further limiting the actual cleaning effect and scope of application of ground garbage cleaning robots.

[0006] Therefore, there is an urgent need for an intelligent ground garbage cleaning robot system with high recognition accuracy, adjustable motion path, efficient execution response, and reasonable structural integration to improve cleaning efficiency and environmental adaptability and meet the actual usage needs in complex application scenarios. Summary of the Invention

[0007] The purpose of the present invention is to overcome the shortcomings of existing ground garbage cleaning robots, such as low garbage recognition accuracy, poor cleaning control coupling, and low dust suction structure efficiency, and to provide an intelligent ground garbage cleaning robot system with high recognition performance, path adaptive control and efficient dust suction capabilities, so as to improve its cleaning capabilities in multi-type garbage recognition and dynamic environments.

[0008] To achieve the above-mentioned objectives, the present invention provides an intelligent ground garbage cleaning robot, comprising: a robot body, a differential drive wheel group, a universal wheel, a control module, a visual recognition module, a dust suction actuator and a dust suction duct structure, wherein: the differential drive wheel group is used to realize the robot's autonomous steering and straight-line movement, and combined with the universal wheel to provide support and posture stability; the visual recognition module is arranged at the front end of the robot, for acquiring ground image information, and realizing garbage type and location recognition through an embedded target detection algorithm; the control module is used to receive the recognition result, and realize dynamic coordination of drive and dust suction execution based on the path planning algorithm and PID control strategy; the dust suction actuator includes a high-speed fan and an integrated dust suction duct, and the duct adopts a closed streamlined structure to improve suction efficiency and reduce noise; the control module completes the parallel processing of motion control, dust suction control and recognition communication through a multi-task scheduling mechanism, forming a closed-loop control system of recognition-planning-execution.

[0009] Furthermore, the visual recognition module is an edge computing platform that supports neural network reasoning and runs a lightweight target detection algorithm to achieve high frame rate recognition; the control module is designed based on a 32-bit microcontroller and supports UART serial port and PWM control signal output.

[0010] Furthermore, the inner wall of the dust suction duct is a low-roughness streamline structure with a surface roughness Ra of no more than 0.8 μm, so as to improve the air flow efficiency and avoid wind resistance loss; the dust suction module supports dynamic adjustment of suction force to adapt to the cleaning needs of different types of garbage.

[0011] The intelligent floor garbage cleaning robot system proposed by the present invention has the following beneficial effects:

[0012] 1. By combining an edge computing platform with a lightweight deep learning model, we can achieve highly robust garbage target recognition, improving recognition accuracy and speed;

[0013] 2. The control system adopts an embedded multi-task scheduling mechanism and PID closed-loop control strategy to achieve coordinated control of the two-wheel differential drive and dust collection module;

[0014] 3. Adopt a new dust collection duct design to improve airflow efficiency, reduce energy consumption and noise, and enhance the recycling capacity of various types of garbage (such as cigarette butts, paper balls, etc.);

[0015] 4. The system has the advantages of high modularity, scalability and cost control, and is suitable for automatic cleaning tasks in various indoor environments such as homes and offices.

[0016] In summary, the present invention significantly improves the intelligent cleaning level of the ground garbage cleaning robot in complex environments through coordinated optimization at the three levels of structure, control and recognition, and has good prospects for promotion and application and engineering feasibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a structural view of the overall appearance of an intelligent ground garbage cleaning robot in an embodiment of the present invention.

[0019] Figure 2 This is a cross-sectional view of an intelligent floor garbage cleaning robot in an embodiment of the present invention.

[0020] Among them, 1 is the visual recognition module, 2 is the main body shell of the floor cleaning robot, 3 is the power battery compartment, 4 is the dust suction fan, 5 is the outer cover of the garbage collection bin, 6 is the power battery, 7 is the driving wheel, 8 is the driving motor, 9 is the visual recognition module fixing bracket, 10 is the garbage cleaning air duct, 11 is the dust suction fan protective cover, 12 is the garbage collection bin, and 13 is the universal wheel. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] The present invention provides Figure 1 and Figure 2 As shown, an embodiment of the present invention provides an intelligent floor garbage cleaning robot, which includes:

[0024] Visual recognition module 1, floor cleaning robot main body shell 2, power battery compartment 3, dust suction fan 4, garbage collection bin shell cover 5, power battery 6, drive wheel 7, drive motor 8, visual recognition module fixing frame 9, garbage cleaning air duct 10, dust suction fan protective cover 11, garbage collection bin 12 and universal wheel 13.

[0025] The main body housing 2 forms the robot's main structure, integrating various functional modules within it and providing support, layout, and protection. The power battery 6, installed in the power battery compartment 3, powers the entire system and provides a stable DC voltage output. The power supply provides voltage adaptation for the control module, vision module, fan, and motor through a voltage conversion circuit.

[0026] The drive wheels 7 are symmetrically arranged on both sides and connected by drive motors 8, forming a differential drive system that enables the robot's forward, reverse, and on-the-spot steering functions. Universal wheels 13 are located at the rear of the robot chassis to provide auxiliary support and dynamic balance, effectively reducing the robot's posture deviation during movement. These universal wheels utilize a spherical rolling structure, offering strong adaptability and flexible rotation, making them suitable for multi-directional movement environments.

[0027] The visual recognition module 1 is mounted on the front of the robot and connected to the main body housing 2 via a visual recognition module mounting bracket 9, ensuring a stable viewing angle and rigid mounting. This module is used to capture ground images and detect and classify garbage objects in real time using a pre-configured embedded neural network model. Data exchange between the visual recognition module and the main control chip is achieved via serial communication, with recognition results transmitted to the control system in real time.

[0028] The dust suction fan 4 is located in the center of the robot and connects to the garbage collection air duct 10, providing suction to collect ground trash. A dust suction fan protective cover 11 is installed outside the fan to prevent foreign matter from being drawn into the fan impeller, enhancing the stability and safety of system operation. The dust suction fan's operating state is regulated by the control module, which automatically adjusts the suction strength based on the recognition results.

[0029] The garbage cleaning duct 10, located between the dust suction inlet and the garbage collection bin 12, features a closed, streamlined design with an inner wall surface roughness Ra no greater than 0.8 μm. This reduces wind resistance and improves suction efficiency, ensuring that larger particles of garbage (such as cigarette butts and paper balls) can pass through and be effectively collected. The duct structure runs through the main unit and, together with the dust suction blower and collection bin, forms a complete cleaning channel.

[0030] The garbage collection bin 12 is used to centrally store the sucked-in garbage and is equipped with a garbage collection bin outer shell cover 5 to achieve sealing and external protection of the bin body, making it convenient for subsequent disassembly and dumping and cleaning.

[0031] This invention achieves a closed-loop control logic of identification, path planning, and cleaning execution through the collaboration between the visual recognition module and the control module. Based on an embedded multi-tasking scheduling system and PID control strategy, the control system adjusts the drive system and dust collection system in real time, improving garbage recognition accuracy and cleaning efficiency.

[0032] To sum up, the intelligent sweeping robot provided in this embodiment has the integrated characteristics of modular structure, intelligent recognition, differential drive and efficient dust collection. It can adapt to a variety of indoor environments, complete efficient and stable ground garbage cleaning tasks, and has good practicality and promotion prospects.

Claims

1. An intelligent ground garbage cleaning robot, characterized by: The robot body, differential drive wheel set, universal wheels for support, control module, visual recognition module, dust collection actuator and dust collection air duct system; the differential drive wheel set is used to enable the robot to move and turn autonomously on the indoor floor; the universal wheels are set at the rear of the robot to provide auxiliary support and balance; The control module is used to receive information transmitted by the recognition module and perform path planning, differential control and dust collection execution actions; the visual recognition module is used to collect ground images and identify the location and type of garbage targets; the dust collection actuator includes a dust collection fan and an integrated dust collection duct, the dust collection duct is connected to the front suction port of the robot, and the rear end is connected to the garbage collection bin; the control module realizes dynamic coordination of the differential drive wheel set and the dust collection actuator through a closed-loop control strategy, forming a closed-loop control system of recognition-planning-execution.

2. The intelligent floor garbage cleaning robot according to claim 1, characterized in that: The control module adopts a hardware platform based on a 32-bit microcontroller and runs a multi-tasking scheduling system and a PID control algorithm to achieve motion control, dust collection control and data synchronization.

3. The intelligent floor garbage cleaning robot according to claim 1 or 2, characterized in that: The visual recognition module adopts an embedded image processing platform that supports edge computing and is configured with a deep learning algorithm for the detection and classification of garbage targets.

4. The intelligent floor garbage cleaning robot according to any one of claims 1 to 3, characterized in that: The dust collection air duct is a closed streamline structure, and the roughness Ra of the inner wall is not greater than 0.8 μm, so as to reduce wind resistance loss and improve dust collection efficiency.

5. The control module according to claim 2, wherein: The microcontroller is an STM32 series chip with a main frequency of not less than 72MHz and has at least two serial port resources for data interaction between the visual recognition module and the drive control module.

6. The visual recognition module according to claim 3, wherein: The recognition module is MaixCAM or an edge AI vision platform with equivalent functions, which has the ability to deploy deep learning models and supports real-time reasoning of the target detection algorithm YOLOv5-s.

Images