805results about "Automatic obstacle detection" patented technology

A system and method are disclosed for training a machine learning model configured to determine a traversability of a real-world environment by a mobile robot based on an image of the real-world environment. The method advantageously generates high quality synthetic training data for training a traversability detection model to discriminate between traversable and untraversable regions in images captured of a real-world environment. The synthetic training data is generated through simulation of a virtual robot in a virtual environment. Once the traversability detection model is trained, it can be deployed to the mobile robot for the purpose of predicting traversability of a real-world environment.

The embodiment of the invention discloses a control method and device of automatic cleaning equipment and related equipment, the automatic cleaning equipment comprises an equipment body and a mechanical arm arranged on the equipment body, the equipment body is provided with a first sensor, and the mechanical arm is provided with a second sensor. The mechanical arm is controlled to extend out of the equipment body, the automatic cleaning equipment is controlled to move in the working environment, and visual information is collected through a first sensor and a second sensor in the moving process; in response to the high-altitude area identified according to the visual information collected by the second sensor, the automatic cleaning equipment is controlled to move to the position below the high-altitude area; the mechanical arm is controlled to enter a high-position scanning posture, and visual information of a high-altitude area is collected through a second sensor; and the visual information collected by the first sensor and the second sensor in the mapping mode is fused to establish a map of the working environment, so that the navigation capability and the obstacle avoidance capability of the automatic cleaning equipment are improved.

The invention relates to the technical field of voice recognition, and discloses a voice instruction recognition and cleaning control method for a sweeping robot, which comprises the following steps: synchronously reading a dust collection motor driving pulse width modulation value and a rolling brush motor load current value of the sweeping robot while acquiring an audio signal acquired by a microphone; retrieving and interpolating in a preset working condition noise spectrum mapping table according to the non-acoustic operation state data, determining a predicted mechanical noise power spectrum at the current moment, taking the predicted mechanical noise power spectrum as a priori reference, calculating an instantaneous signal-to-noise ratio of the predicted mechanical noise power spectrum and the frequency domain feature sequence, and constructing a frequency domain confidence mask; according to the method, a feedforward prediction mechanism is constructed by introducing non-acoustic prior information, and response lag of traditional acoustic posteriori estimation during sudden change of working conditions is avoided.

The embodiment of the invention provides a cleaning robot which comprises a machine body, a cleaning device and a driving device. The driving wheel assembly is arranged on the machine body; the dust collection cleaning device comprises a dust box assembly, a dust collection fan assembly and a dust discharge channel assembly, the first end of the dust discharge channel assembly is connected with the dust box assembly, and the second end of the dust discharge channel assembly extends to the rear end of the machine body; the mopping and washing assembly is arranged on the rear side of the dust box assembly, the mopping and washing device is movably connected with the machine body, and the mopping and washing assembly in the mopping and washing device can extend outwards from one side of the machine body relative to the machine body in the direction of the transverse central axis; the water path system is connected with the mopping assembly; in the direction of the transverse central axis, the dust discharging channel assembly is located on the opposite side of the stretching-out direction of the mopping assembly. According to the technical scheme, the mopping assembly is arranged in the transverse direction of the machine body, and the dust discharging channel is formed in one side of the mopping assembly, so that the mopping assembly with the enough length is arranged on the machine body, and the side space of the mopping assembly can be fully utilized.

An automatic cleaning method and apparatus of a mobile robot, a mobile robot, and a storage medium. The automatic cleaning method comprises collecting a current image of a scene, recognizing an object in the current image, and determining the type of the object (S101); marking the recognized object as a preliminary obstacle on a map of the scene (S102); and on the basis of the type of the preliminary obstacle, performing a corresponding cleaning operation on the basis of a wall-following sensor (S103). By means of image recognition and a wall-following sensor, objects in a scene are intelligently recognized and classified, and accordingly a precise cleaning operation is performed, thereby improving the cleaning efficiency and accuracy of mobile robots.

The invention relates to the technical field of robots, in particular to a robot operation method and system combined with a scene graph. According to the method, a monitoring camera can be utilized to perform semantic segmentation on a shot image (scene graph) of a target ground to identify a stain area and a stain type, the cleaning complexity is distinguished, and the stain cleaning sequence and the cleaning parameters are determined from low to high based on the complexity, so that low-complexity stains can be processed firstly, and the cleaning efficiency is improved. The method has the advantages that high-complexity stains are processed, the sequence is reasonable, secondary pollution is avoided, different cleaning parameters are determined according to different complexities, one-step operation is avoided, cleaning is more reasonable, in addition, a scene graph acquired by the robot is acquired by monitoring, scanning by the robot is not needed, the efficiency is high, and omission is avoided.

A robot cleaner having a shielding module provided under a cleaner body is provided. The shielding module includes a cam and a hiding plate which rotate with the same phase angle about a same rotating shaft. The distance between the center of the hiding plate and the center of the rotating shaft is longer than the distance between a contact point on a cam surface of the cam and the center of the rotating shaft. A guide protrusion of a bumper is in contact with the cam surface at the contact point. Therefore, the rotation movement speed of the hiding plate which rotates to hide a cliff sensor is much faster than the inner movement speed of the bumper, thereby improving sensitivity to the detection of a collision between the bumper and an obstacle by using the cliff sensor.

The embodiment of the invention provides a cleaning robot which comprises a machine body, a cleaning device and a driving device. The driving assembly is arranged on the machine body corresponding to the transverse central axis and used for driving the machine body to advance; the cleaning module is movably connected with the machine body, and the cleaning module can extend outwards from one side of the machine body relative to the machine body in the direction of the transverse central axis; and the obstacle avoidance assembly is located on the machine body on the same side as the extending direction of the cleaning module. According to the technical scheme, the roller obstacle avoidance assembly is arranged on the machine body on the extending side of the mopping assembly and can be combined with the obstacle avoidance sensor assembly on the front side of the machine body for obstacle avoidance, and the roller obstacle avoidance assembly also gives consideration to obstacle avoidance of the cleaning module.

The invention discloses a cleaning robot autonomous operation method and system based on multi-sensor fusion and reinforcement learning. The method comprises the following steps: collecting visual, physical and fuzzy modal information through a multi-sensor system; performing fusion processing on the multi-modal information by using a deep learning model and a hesitant fuzzy set, and constructing a unified environment state representation; a hierarchical reinforcement learning framework is established based on the representation, a cleaning area is divided into a plurality of task units, a high-level strategy selects the task unit, and a bottom-level strategy executes a specific cleaning action; designing a multi-target reward function, comprehensively considering indexes such as cleaning coverage rate and energy consumption efficiency, and introducing a dynamic weight adjustment mechanism; adopting an option-evaluator framework to jointly train a high-level strategy and a bottom-level strategy; and dynamically adjusting the decision model through online learning and a local optimization algorithm in the operation process. According to the invention, closed-loop control from environment perception to task execution is realized, and the autonomy, adaptability and operation efficiency of the cleaning robot in a complex environment are improved.

The application discloses a robot area traversal method and belongs to the field of intelligent robots. The method first judges whether a feedback signal of a laser sensor is received when the robot receives a start signal for starting traversal, and when the feedback signal of the laser sensor is not received, the robot can rely on sensors such as a photosensitive sensor, a gyroscope and an odometer to perform area traversal in a non-laser navigation mode, so as to clean the bottom area of furniture with a relatively low bottom space, thereby improving the application range of the robot and remarkably improving practicability.

The present disclosure relates to the technical field of cleaning. Disclosed are a control method for a cleaning robot, a cleaning robot, a base station, a cleaning system, and a storage medium. The control method comprises: detecting a first obstacle on an initial traveling path of a cleaning robot; analyzing feature information of the first obstacle, wherein the feature information of the first obstacle comprises at least one of the following: type information of the first obstacle, dimensional information in a horizontal plane, contour information, and arrangement density information within a preset range; and controlling the state of a cleaning member and the state of a body on the basis of the feature information of the first obstacle. The control method for the cleaning robot of the present disclosure can use different cleaning modes in different scenarios, thereby improving the cleaning effect of the cleaning robot.

The embodiment of the invention provides a cleaning robot which comprises a machine body, a cleaning device and a driving device. The driving assembly is arranged on the machine body corresponding to the transverse central axis and used for driving the machine body to advance; the outer swinging side brush assembly is located on one side of the front end of the machine body and can swing outwards relative to the machine body; the collision plate assembly is arranged on the front side of the machine body in a surrounding mode, a window is formed in the collision plate assembly and comprises windows, and the windows are asymmetrically distributed in the direction of the vertical central axis. According to the technical scheme, the sensor support is arranged towards one side of the vertical central axis, and the windows are asymmetrically arranged, so that the monitoring range of the sensor module is not affected, and the obstacle avoidance sensor assembly and the outer swing side brush assembly can be arranged at the front end of the machine body.

The technology disclosed includes systems and methods for preparing a segmented occupancy grid map based upon image information of an environment in which a robot moves. The image information is captured by at least one visual spectrum-capable camera and at least one depth measuring camera. The system includes logic to receive image information captured by at least one visual spectrum-capable camera and location information captured by at least one depth measuring camera located on a mobile platform. The system includes logic to extract from the image information, features in the environment. The system includes logic to determine a 3D point cloud of points having 3D information. The system includes logic to determine, from the 3D point cloud, an occupancy map of the environment. The system includes logic to segment the occupancy map into a segmented occupancy map of regions that represent rooms and corridors in the environment.

A cleaning roller mountable on a cleaning robot and an autonomous cleaning robot. The cleaning roller includes a sheath including an outer surface tapering toward a center of the cleaning roller, a core extending along a rotational axis of the cleaning roller from a first end of the core to a second end of the core, the cleaning roller mountable to the cleaning robot to rotate about the rotational axis, and an air gap between an inner surface of the sheath and an outer surface of the core, the air gap extending from a portion of the core between the first end of the core and the second end of the core longitudinally along the rotational axis toward the first end of the core or the second end of the core.

Disclosed is a method and tidying robot system that performs it. A robot of a tidying robot system receives a starting location, a target cleaning area, and obstructions in its path navigating the target cleaning area. It determines a tidying strategy including vacuuming and obstruction handling. It executes the tidying strategy, vacuuming the target cleaning area and moving and / or avoiding obstructions, which are tidyable and / or movable objects. Where the obstruction may be picked up, the robot determines and executes a pickup strategy. Where the obstruction may be relocated but not picked up, it pushes the obstruction to a different location. Where the obstruction cannot be relocated / picked up, it alters its path to avoid the obstruction. The robot determines if the dirt collector is full, and if so, navigates to a base station configured to charge the robot and empty the dirt collector. Otherwise, it continues to execute the tidying strategy.

A cleaning robot may include a body, a driving wheel, and a cleaning member. The driving wheel and the cleaning member may be installed on and rotatably connected with the body. The cleaning robot may have a backward obstacle-crossing mode and a forward obstacle-crossing mode. In the forward obstacle-crossing mode, a head of the cleaning robot may face an obstacle, the driving wheel may rotate forward, and the cleaning member may rotate in the same or in an opposite direction to the rotation of the driving wheel. In the backward obstacle-crossing mode, a tail of the cleaning robot may face the obstacle, the driving wheel may rotate reversely, and a rotation direction of the cleaning member may be the same as or opposite to the rotation direction of the cleaning member in the forward obstacle-crossing mode.

The embodiment of the invention provides a cleaning robot which comprises a machine body, a cleaning device and a driving device. The driving assembly is arranged on the machine body corresponding to the transverse central axis and used for driving the machine body to advance; the cleaning module and the mopping device are movably connected with the machine body, and the mopping device can extend outwards from one side of the machine body relative to the machine body in the direction of the transverse central axis; the water path system comprises a clear water tank which is arranged on the machine body; the clean water tank is located on the opposite side of the extending side of the cleaning module. According to the technical scheme, the clean water tank is arranged on one side of the mopping assembly, and the problem that the trafficability is affected due to the fact that the clean water tank is arranged above the mopping assembly and the machine body is too high can be solved. In addition, the space in the machine body can be fully utilized, so that the capacity of the clear water tank is improved.

The wall-climbing robot comprises a robot main machine, a vertical obstacle crossing platform, a plurality of climbing obstacle crossing units, a control unit and an operation executing mechanism, the vertical obstacle crossing platform is fixedly connected to one side of the robot main machine, and the vertical obstacle crossing platform is perpendicular to the robot main machine. The multiple climbing obstacle crossing units are sequentially arranged at intervals in the length direction of the vertical obstacle crossing platform, all the climbing obstacle crossing units are in sliding fit with the vertical obstacle crossing platform, the climbing obstacle crossing units are arranged on the robot host through the vertical obstacle crossing platform, and the control unit is arranged in the robot host. The control unit is used for controlling the climbing obstacle crossing units to act alternately. The climbing obstacle-crossing unit is in sliding fit with the vertical obstacle-crossing platform, so that stable movement and obstacle-crossing operation of the robot host along the wall surface are realized, the path planning of the operation executing mechanism is greatly optimized, and the robot is particularly suitable for carrying a heavy operation executing mechanism.

A system and method of software control and autonomy of a disinfection module for an autonomous or semi-autonomous cleaning device. The disinfection module control software resides on a computer with processor and memory. The disinfection module is connected to peripherals for sensing and localizing in the environment and controlling the actuators for the motion of the cleaning device. The disinfection module control software processes the sprayer's status, follows and plans paths to move the cleaning device to the spray targets, generates appropriate motion commands, and controls the disinfection module's pump, fan, LED, and electrostatic generator states.

The invention provides a cleaning method and device of a cleaning robot, the cleaning robot, equipment and a medium, and relates to the technical field of cleaning equipment. Comprising the steps that in the process that a cleaning robot cleans a to-be-cleaned area, under the condition that it is detected that a heavily-polluted area exists, the smudginess area of the heavily-polluted area is determined; if the dirty area is larger than a preset threshold value, the robot is controlled to advance in an advancing mode based on a preset path so as to clean at least part of the heavily-dirty area through a mop assembly; if the smudginess area is smaller than or equal to the preset threshold value, the robot is controlled to advance in a backward mode based on the preset path so as to clean at least part of the heavily-polluted area through the mop assembly, and the robot can push smudginess out of the heavily-polluted area and pollute the adjacent area due to backward cleaning, so that the smudginess can be pushed out of the heavily-polluted area and pollute the adjacent area for the heavily-polluted areas with different areas. The cleaning robot can make intelligent cleaning decisions, and the cleaning effect is improved.

The invention provides a control method and device of a cleaning robot and the cleaning robot, and relates to the technical field of cleaning equipment.The cleaning robot comprises a mop assembly, the mop assembly comprises a mop, and the mop is provided with a first position, a second position and a third position; the first position and the second position are two end point positions of the mop in the reciprocating motion process, the walking direction of the cleaning robot is the longitudinal direction, the first position, the second position and the third position are arranged at intervals in the transverse direction perpendicular to the longitudinal direction, and the third position is the maximum limit position extending out relative to a cleaning robot body in the transverse direction; the spacing distance between the first position and the second position is smaller than that between the first position and the third position; in the process that the cleaning robot cleans the to-be-cleaned area, the mop is controlled to vibrate back and forth between the transverse first position and the transverse second position, so that the friction cleaning effect of the heavily polluted area is enhanced, and meanwhile, the problem that stains are dragged by the mop to form secondary traces is effectively avoided.

The embodiment of the invention provides a cleaning robot. The cleaning robot comprises a chassis; the shell is arranged above the chassis to form a machine body of the cleaning robot; the driving assembly is used for driving the chassis to advance; the cleaning module is arranged on the chassis; the mainboard assembly is arranged in the machine body; the obstacle avoidance sensor assembly comprises a sensor module and a sensor support, the obstacle avoidance sensor assembly is arranged on the chassis, and the sensor module is arranged on the sensor support; a window is arranged on the shell corresponding to the position of the sensor module; wherein the sensor support is provided with a heat dissipation hole, the heat dissipation hole is communicated with the interior and the exterior of the machine body through a window, and the mainboard assembly is arranged near the heat dissipation hole to dissipate heat of the mainboard assembly. According to the technical scheme, the heat dissipation holes are formed in the sensor support, external airflow can enter the machine body through the heat dissipation holes to dissipate heat of the mainboard assembly, the heat dissipation effect is good, and the mainboard assembly is not prone to overheating.

The invention relates to a self-moving robot and a system thereof. The method comprises the following steps: acquiring a depth picture acquired by a laser ranging unit and binocular RGB pictures acquired by a left eye RGB camera and a right eye RGB camera; based on a pre-trained target deep learning algorithm, determining a target depth map according to a depth picture acquired by the laser ranging unit and a binocular RGB picture acquired by the left eye RGB camera and the right eye RGB camera together; determining a 3D map according to a left eye RGB picture collected by a left eye RGB camera and the target depth map; determining an obstacle avoidance strategy according to the 3D map and RGB pictures collected by the left eye RGB camera; and controlling the self-moving robot to move according to the obstacle avoidance strategy. At least the overall height of the machine body can be reduced, a 3D map is established in real time, an obstacle avoidance strategy is determined, and the cleaning coverage rate and the accuracy of the obstacle avoidance strategy are improved.

The invention provides a cleaning base station and a control method, and relates to the technical field of cleaning equipment, the cleaning base station comprises a liquid storage tank, the liquid storage tank comprises a first liquid storage cavity and a second liquid storage cavity, the first liquid storage cavity is used for storing cleaning liquid, and the second liquid storage cavity is used for storing deodorant liquid. The liquid storage tank is further provided with a first liquid outlet and a second liquid outlet, and the base station body is provided with a first connector and a second connector. The first connector is connected with the first liquid outlet, and the second connector is connected with the second liquid outlet. The cleaning liquid is conveyed to the wiping and mopping piece through the first liquid outlet and the first connector at a time, and the deodorization liquid is conveyed to the wiping and mopping piece through the second liquid outlet and the second connector in sequence. The first liquid outlet and the second liquid outlet are located in the outer surface of the liquid storage tank. According to the cleaning base station provided by the invention, the removal effect of the cleaning equipment on odor and stains can be effectively improved, and the comprehensive cleaning capacity of the cleaning equipment is improved.

Assembly structure mounted on a base of a self-driving robot, the assembly structure comprising: a protective shell comprising a protective chamber for accommodating a positioning module; and a cover plate placed on top of the protective case to form a receiving space between the cover plate and the protective case, which is used to receive a voice receiving module; wherein the self-driving robot also includes a top cover which is placed on the base and wherein the top cover is provided with a through-hole to allow space for the protective cover of the mounting structure.

The invention provides a robot control method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring an image acquired in real time; according to the image acquired in real time, determining whether a current scene meets conditions of a specific scene; and under the condition that the current scene meets the conditions of a specific scene, controlling a robot to execute a cleaning task corresponding to the specific scene. According to the method, scene recognition and cleaning task triggering functions in a specific scene are added for the robot, a user does not need to manually set and issue a cleaning instruction to the robot, the convenience of user operation can be enhanced, and the use experience of the user is improved.

The application provides a self-moving robot, which comprises a chassis, a laser radar and a front wheel assembly. The laser radar is fixedly connected to one side of the chassis, and the front wheel assembly is rotatably supported on the side of the chassis away from the laser radar. The front wheel assembly is at least partially overlapped with the laser radar in the projection on the chassis. The front wheel assembly is at least partially overlapped with the laser radar in the projection on the chassis, so that the laser radar can be sufficiently supported and stabilized by the chassis and the front wheel assembly, avoiding shaking or damage of the laser radar caused by excessive external pressure.

This specification discloses a robot obstacle detection method, apparatus, medium, and electronic device, wherein the method includes the steps of acquiring three-dimensional point cloud data of an obstacle in a first coordinate system, the first coordinate system being a coordinate system established by the robot at its current position with the robot as its origin; fusing the three-dimensional point cloud data into a robot-centered grid map to determine obstacle grid points in the grid map; clustering the obstacle grid points to obtain at least one obstacle grid point set; and generating obstacle objects in the grid map based on the obstacle grid point set.

A docking station including a docking unit providing a traveling path for a robotic vacuum cleaner, having a mounted mopping cloth, that enters the docking unit; a mopping cloth separator disposed on the docking unit and configured to automatically separate the mounted mopping cloth from the robotic vacuum cleaner while the robotic vacuum cleaner is at a first position along the traveling path; and a mopping cloth coupler disposed on the docking unit and configured so that, after the mounted mopping cloth is separated from the robotic vacuum cleaner, and the robotic vacuum cleaner thereafter moves to a second position along the traveling path, the mopping cloth coupler supplies a mopping cloth to be mounted to the robotic vacuum cleaner from below the traveling path while the robotic vacuum cleaner is at the second position.

Embodiments of the disclosure provide a robot cleaner for cleaning a stair. The robot cleaner includes a main body configured to rotate around a first shaft. The robot cleaner further includes at least one leg each configured to rotate around a second shaft. The robot cleaner also includes at least one connector each configured to connect each leg to the main body. Each connector comprises a first end attached to the first shaft and a second end attached to the second shaft. The robot cleaner is configured to climb the stair by first lifting the main body onto a tread of the stair through rotations of the main body and the at least one connector and then lifting the at least one leg onto the tread through rotations of the at least one connector and the at least one leg.