A low-power anti-collision method and system for cleaning robots

Abstract

This application discloses a low-power anti-collision method and system for a cleaning robot. The method includes: installing multiple pairs of light intensity sensors and ultrasonic radar on the cleaning robot; using the light intensity sensors to detect ambient light intensity; using the ultrasonic radar to detect obstacles in the forward direction; using a lidar to scan the surrounding environment and dynamically correct the cleaning edge; when the light intensity signal reaches the activation threshold but the ultrasonic radar does not issue an alarm signal, activating the lidar to scan the light source direction; controlling the robot to rotate to the left outside the turning radius, recording the vanishing angle on the left and the vanishing angle on the right; selecting the direction with the smaller angle as the new forward path; if the angles on both sides are equal, planning the shortest forward path through gridded space division and the A* algorithm. This application improves cleaning efficiency and safety, and uses ultrasonic and light intensity sensors for full-process obstacle detection, avoiding the high power consumption of continuously operating the lidar, thus achieving energy saving.

Description

Technical Field

[0001] This application relates to the field of intelligent robot technology, and in particular to a low-power anti-collision method and system for a cleaning robot. Background Technology

[0002] With the rapid development of intelligent cleaning robot technology, its application in complex environments such as underground parking garages and shopping malls is becoming increasingly widespread. However, in environments like underground parking garages, cleaning robots face various collision risks, such as high-value vehicles, pedestrians, and low or black objects (such as license plates, front scrapers, tires, and black bumpers). Traditional cleaning robots typically rely on lidar for environmental perception and path planning, but lidar has the following limitations: lidar is usually mounted on the top of the robot and scans horizontally, creating a natural blind spot for objects below its height (such as license plates and tires); black or matte objects (such as black bumpers and tires) have low reflectivity and weak laser signals, making them easily misjudged as "unobstructed." Furthermore, the power consumption of lidar and signal processing is one to two orders of magnitude higher than that of ultrasonic radar and signal processing. Periodically activating the lidar can achieve both safety and power saving.

[0003] Therefore, there is an urgent need for a collision avoidance method for cleaning robots that is low in power consumption, robust, and can effectively cope with the blind spots of lidar and complex environments. Summary of the Invention

[0004] This application provides a low-power anti-collision method and system for cleaning robots, which effectively avoids obstacles in complex scenarios such as misaligned tires and black obstacles, while optimizing the cleaning edge and forward path, improving cleaning efficiency, enhancing safety, and reducing energy consumption.

[0005] This application provides the following solution:

[0006] According to a first aspect, a collision avoidance method for a cleaning robot is provided, the method comprising: installing multiple pairs of light intensity sensors and ultrasonic radars on the cleaning robot, each pair of light intensity sensors and ultrasonic radars being installed in the same direction; using the light intensity sensors to detect ambient light intensity and obtain a light intensity signal; using the ultrasonic radars to detect obstacles in the forward direction and obtain ultrasonic radar feedback signals; adjusting the movement state of the cleaning robot according to the light intensity signal and the ultrasonic radar feedback signal; installing a lidar on the cleaning robot, the lidar being activated when scanning the surrounding environment is required and / or at preset time intervals; using the lidar to scan the surrounding environment, generating a cleaning edge line, and dynamically correcting the cleaning edge line when approaching a vehicle based on a comparison of the forward and lateral feedback signals of the ultrasonic radar with their respective reference signals; and activating the lidar to scan the light intensity signal when the light intensity signal reaches a trigger threshold but the ultrasonic radar does not issue an alarm signal. If no obstacle is detected in the direction of origin, the robot continues to move forward. If an obstacle is detected in the direction of movement, the robot pauses, rescans to confirm the obstacle's position, and then initiates an obstacle avoidance strategy. The obstacle avoidance strategy includes: controlling the robot to rotate to the left at a fixed angular velocity outside the turning radius, while monitoring the signal strength of the ultrasonic radar and light intensity sensor in the forward and left directions; initiating a laser radar scan at a predetermined angle and recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal; returning to face the obstacle and rotating to the right, while monitoring the signal strength of the ultrasonic radar and light intensity sensor in the forward and right directions; initiating a laser radar scan at a predetermined angle and recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal; comparing the vanishing angles on both sides, selecting the direction with the smaller angle as the new forward path; if the angles on both sides are equal, using the laser radar scan data, through gridded space division and the A* algorithm, planning the shortest trajectory from the current position to the original path as the new forward path.

[0007] According to one achievable method in this application embodiment, adjusting the motion state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal includes: when the signal strength of the light intensity signal reaches a first threshold, controlling the cleaning robot to decelerate; when the signal strength of the light intensity signal reaches a second threshold, controlling the cleaning robot to stop and issuing an alarm signal; when the signal strength of the ultrasonic radar feedback signal reaches a third threshold, controlling the cleaning robot to decelerate; when the signal strength of the ultrasonic radar feedback signal reaches a fourth threshold, controlling the cleaning robot to stop and issuing an alarm signal; when the signal strength of the light intensity signal reaches the first threshold but the signal strength of the ultrasonic radar feedback signal does not reach the fourth threshold, the cleaning robot only decelerates without stopping.

[0008] According to one achievable method in an embodiment of this application, when generating a cleaning edge line by scanning the surrounding environment with a lidar, the method further includes: using the lidar to scan the vehicle outline and identifying the position of protruding parts by ranging, wherein the protruding parts include: rearview mirrors, tires and / or license plates; when the height of the protruding parts above the ground is higher than that of the cleaning robot, the protruding parts are not considered when generating the cleaning edge line.

[0009] According to one achievable method in this application embodiment, the dynamic correction of the cleaning edge line when approaching the vehicle based on the comparison of the forward and lateral feedback signals of the ultrasonic radar with their respective reference signals includes: when the forward and lateral signal strengths of the ultrasonic radar are both lower than 80% of the reference signal, it is determined that the tire is misaligned, the robot approaches the obstacle at a low speed, calculates the offset angle and distance of the protruding part, and extends the cleaning edge line outward by 5 to 10 centimeters to cover the rear area of ​​the tire; when the signal strength is higher than 120% of the reference signal, it is determined to be a black obstacle, the robot stops cleaning and shifts the edge line outward by 10 to 15 centimeters according to the ranging data to avoid collision.

[0010] According to one achievable method in this application embodiment, the step of using LiDAR scanning data to plan the shortest trajectory from the current position to the original path as a new forward route through gridded spatial division and A* algorithm includes: converting the LiDAR scanning data into a gridded map, dividing the environment into grid cells with a side length of 0.4 meters; based on the gridded map, taking the current position as the starting point and the original path after the obstacle as the target point, using the A* algorithm to calculate the shortest trajectory, wherein the A* algorithm combines the actual movement cost from the starting point to the current grid and the estimated cost from the current grid to the target point through an evaluation function, and selects the grid with the lowest cost for expansion; when turning left or right to 60 degrees, the LiDAR is activated to perform a 120-degree angle scan to obtain local environmental data to update the gridded map; based on the shortest trajectory calculated by the A* algorithm, a new forward route is generated and the cleaning robot is controlled to move along the route.

[0011] According to one achievable method in this application embodiment, the method further includes: controlling the cleaning robot through a cloud platform, the control including: reporting the detection signals of the light intensity sensor, the ultrasonic radar and the lidar to the cloud platform; if the alarm information is generated during the cleaning process, reporting the alarm information to the cloud platform; and guiding and planning for potential collisions between multiple cleaning robots.

[0012] According to one achievable method in the embodiments of this application, the outer shell of the cleaning robot is made of polyurethane anti-collision outer shell and coated with a reflective coating.

[0013] According to a second aspect, a low-power anti-collision system for a cleaning robot is provided. The system includes: multiple pairs of light intensity sensors and ultrasonic radars mounted on the cleaning robot, each pair of sensors and radars having the same mounting direction, used to detect ambient light intensity to generate a light intensity signal and detect obstacles in the forward direction to generate an ultrasonic radar feedback signal; a lidar mounted on the cleaning robot, used to scan the surrounding environment to generate cleaning edge lines, the lidar being activated when scanning the surrounding environment is required and / or at preset time intervals; and a control module communicatively connected to the light intensity sensors, ultrasonic radars, and lidar, configured to: adjust the movement state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal. An initial cleaning edge line is generated based on the scanning data from the lidar. As the robot approaches a vehicle, the cleaning edge line is dynamically corrected by comparing the forward and lateral feedback signals from the ultrasonic radar with their respective reference signals. When the light intensity signal reaches a first threshold but the ultrasonic radar does not issue an alarm signal, the lidar is activated to scan the direction of the light source. If no obstacle is detected, the robot continues to move forward. If an obstacle is detected in the forward direction, the robot pauses, rescans to confirm the obstacle's position, and then activates an obstacle avoidance strategy. The obstacle avoidance strategy includes controlling the robot to rotate to the left at a fixed angular velocity outside the turning radius, while simultaneously monitoring the forward and left-side ultrasonic radar and light intensity sensors. Based on the signal strength, the system initiates a lidar scan at a predetermined angle, recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal. After returning to face the obstacle, it rotates to the right, simultaneously monitoring the signal strength of the forward and right ultrasonic radars and light intensity sensors. The system then initiates a lidar scan at a predetermined angle, recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal. Comparing the vanishing angles on both sides, the system selects the direction with the smaller angle as the new forward path. If the angles on both sides are equal, the system uses the lidar scan data to plan the shortest trajectory from the current position to the original path using a gridded space and the A* algorithm, which serves as the new forward path.

[0014] According to a third aspect, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0015] According to a fourth aspect, an electronic device is provided, comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any one of the first aspects.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0017] This application achieves efficient collision avoidance by installing multiple pairs of light intensity sensors and ultrasonic radar on the robot, combined with lidar. The light intensity sensors detect ambient light intensity and prioritize adjusting the robot's movement to effectively handle dynamic light sources such as car headlights. The ultrasonic radar detects obstacles, compensating for the lidar's blind spots for low-lying or dark objects, dynamically correcting the cleaning edge lines to ensure coverage of areas where tires are misaligned and avoiding dark obstacles. By analyzing changes in ultrasonic radar signal strength, the robot intelligently selects a detour direction and generates new routes using gridding and the A* algorithm, optimizing global navigation. The entire process is supported by a cloud platform, enabling multi-robot coordination and path optimization. This method significantly improves the obstacle avoidance capabilities of cleaning robots in complex environments (such as underground parking garages), enhancing cleaning efficiency and safety. Simultaneously, a polyurethane anti-collision outer shell and reflective coating provide additional physical protection, reducing the risk of collisions. Compared to traditional solutions, this method is low-cost, robust, and particularly suitable for scenarios with insufficient lighting or diverse obstacles, demonstrating significant practical value and widespread application potential.

[0018] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a system architecture diagram applicable to the embodiments of this application;

[0021] Figure 2 A flowchart illustrating the low-power anti-collision method for a cleaning robot provided in this application embodiment;

[0022] Figure 3 This is a structural block diagram of the low-power cleaning robot anti-collision system provided in the embodiments of this application;

[0023] Figure 4 A schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0025] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0026] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0027] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0028] In existing technologies, some cleaning robots compensate for the shortcomings of lidar by adding cameras or infrared sensors. However, cameras are ineffective in poorly lit underground parking environments and are costly; infrared sensors have limited ability to detect black objects. In addition, existing solutions lack efficient multi-sensor fusion and anomaly handling mechanisms when dealing with dynamic obstacles (such as protrusions caused by misaligned tires) or planning new paths, resulting in low cleaning efficiency or increased collision risk.

[0029] In view of this, this application provides a new approach. To facilitate understanding of this application, the system architecture on which this application is based will first be described. Figure 1 An exemplary system architecture that can be applied to embodiments of this application is shown, such as Figure 1 As shown, the system architecture may include: a low-power anti-collision system for the cleaning robot and a cloud platform located on the server side.

[0030] The anti-collision system for cleaning robots, according to the method provided in this application, reports the detection signals from light intensity sensors, ultrasonic radar, and lidar to a cloud platform; if the alarm information is generated during the cleaning process, the alarm information is reported to the cloud platform; the cloud platform can store and calculate relevant data, and guide and plan for potential collisions between multiple cleaning robots.

[0031] A cloud platform can be configured as a standalone server, a server cluster, or a cloud server. A cloud server, also known as a cloud computing server or cloud host, is a hosting product within the cloud computing service system, designed to address the management difficulties and weak service scalability inherent in traditional physical hosts and Virtual Private Servers (VPS) services. It should be understood that... Figure 1 The anti-collision system and cloud platform for the cleaning robot shown are merely illustrative. Depending on the implementation requirements, any number of anti-collision systems and cloud platforms for the cleaning robot can be implemented.

[0032] Figure 2 This is a flowchart of a collision avoidance method for a cleaning robot provided in an embodiment of this application. The method can be... Figure 1 The anti-collision system of the cleaning robot in the system shown is being implemented. For example... Figure 2 As shown, the method may include the following steps:

[0033] Step 201: Install multiple pairs of light intensity sensors and ultrasonic radars on the cleaning robot, with each pair of light intensity sensors and ultrasonic radars installed in the same direction.

[0034] Step 202: Detect ambient light intensity using the light intensity sensor to obtain a light intensity signal; detect obstacles in the forward direction using the ultrasonic radar to obtain an ultrasonic radar feedback signal; adjust the movement state of the cleaning robot according to the intensity of the light intensity signal and the ultrasonic radar feedback signal.

[0035] Step 203: Install a lidar on the cleaning robot. The lidar is turned on when it is necessary to scan the surrounding environment and / or at preset time intervals. The lidar is used to scan the surrounding environment, generate cleaning edge lines, and dynamically correct the cleaning edge lines when the robot is close to the vehicle by comparing the positive and lateral feedback signals of the ultrasonic radar with their respective reference signals.

[0036] Step 204: When the light intensity signal reaches the activation threshold but the ultrasonic radar does not issue an alarm signal, the laser radar is activated to scan the direction of the light source. If no obstacle is detected, the robot continues to move forward. If an obstacle is detected in the forward direction, the robot pauses and rescans to confirm the obstacle's position before activating the obstacle avoidance strategy. The obstacle avoidance strategy includes: controlling the robot to rotate to the left at a fixed angular velocity outside the turning radius, while monitoring the signal strength of the ultrasonic radar and light intensity sensor in the forward and left directions, activating the laser radar scan at a predetermined angle, and recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal; returning to face the obstacle and rotating to the right, while monitoring the signal strength of the ultrasonic radar and light intensity sensor in the forward and right directions, activating the laser radar scan at a predetermined angle, and recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal; comparing the vanishing angles on both sides, selecting the direction with the smaller angle as the new forward path; if the angles on both sides are equal, using the laser radar scan data, the robot plans the shortest trajectory from the current position to the original path using a gridded space and the A* algorithm as the new forward path.

[0037] As can be seen from the above process, this application achieves efficient collision avoidance by installing multiple pairs of light intensity sensors and ultrasonic radar on the robot, combined with lidar. The light intensity sensors detect ambient light intensity and prioritize adjusting the robot's movement state to effectively cope with dynamic light sources such as car headlights; the ultrasonic radar detects obstacles, compensating for the blind spots of lidar on low or dark objects, dynamically correcting the cleaning edge lines to ensure coverage of areas where tires are not aligned and avoiding dark obstacles; based on changes in ultrasonic radar signal intensity, the robot intelligently selects a detour direction and generates a new forward path through gridding and the A* algorithm, optimizing global navigation. The entire process, combined with cloud platform support, enables multi-robot coordination and path optimization. This method significantly improves the obstacle avoidance ability of cleaning robots in complex environments (such as underground parking garages), enhancing cleaning efficiency and safety, while providing additional physical protection through a polyurethane anti-collision outer shell and reflective coating, reducing the risk of collisions. Compared to traditional solutions, this method is low-cost, robust, and particularly suitable for scenarios with insufficient lighting or diverse obstacles, possessing significant practical value and promotion potential.

[0038] The following describes in detail each step of the above process and the effects that can be further produced, with reference to the embodiments. It should be noted that the terms "first" and "second" involved in this disclosure do not have limitations in terms of size, order, or quantity, but are only used to distinguish them in name. For example, "first threshold" and "second threshold" are used to distinguish thresholds for different purposes.

[0039] First, the above step 201, namely "installing multiple pairs of light intensity sensors and ultrasonic radars on the cleaning robot, with each pair of light intensity sensors and ultrasonic radars installed in the same direction," will be described in detail with reference to the embodiments.

[0040] First, this application installs multiple pairs of light intensity sensors and ultrasonic radars on the cleaning robot, enhancing the robot's perception capabilities in complex environments through the collaborative work of these sensors and radars. This design means the robot is equipped with multiple light intensity sensor and ultrasonic radar units, each pair consisting of one light intensity sensor and one ultrasonic radar, working together to perform environmental perception tasks. The light intensity sensor detects the light intensity of the surrounding environment, such as the strong light from vehicle headlights, to determine the presence of approaching dynamic objects. The ultrasonic radar, by emitting ultrasonic waves and receiving reflected signals, detects the distance and position of obstacles ahead, suitable for detecting low-lying or dark objects that are difficult for lidar to identify. The setup of multiple sensor pairs ensures comprehensive coverage of the surrounding environment, enabling the robot to acquire light intensity and obstacle information from different angles, thereby improving the accuracy and robustness of perception.

[0041] Secondly, each pair of light intensity sensors and ultrasonic radars is installed in the same direction. This feature means that in each pair of sensors, the light intensity sensor and ultrasonic radar are aligned, pointing in the same direction. This unidirectional design allows the two types of sensors to work collaboratively in the same area, forming a complementary perception mechanism. For example, when the robot approaches a vehicle, the light intensity sensor can detect changes in the light intensity of the vehicle's headlights, indicating that a vehicle may be approaching; while the ultrasonic radar simultaneously detects the distance to obstacles in that direction, providing accurate ranging data. This collaborative perception can more accurately determine the environmental state, such as distinguishing between dynamic light sources and static obstacles, thus providing a reliable basis for subsequent motion state adjustments and path planning.

[0042] This design is typically implemented by evenly distributing sensor pairs around the robot, for example, installing a pair at regular intervals to achieve 360-degree omnidirectional perception. In practical applications, the number of sensor pairs and their distribution angles can be optimized based on the robot's size and the application scenario. For example, in complex environments like underground parking garages, six pairs of sensors distributed at 60-degree intervals can effectively cover all directions, ensuring no blind spots in detection. The sensitivity of light intensity sensors is usually designed to detect a wide range of changes from low light to strong light, while the detection range of ultrasonic radar is generally controlled within 5 meters to ensure high-precision short-range obstacle avoidance capabilities.

[0043] The following describes in detail step 202, namely, "using the light intensity sensor to detect the ambient light intensity and obtain a light intensity signal; using the ultrasonic radar to detect obstacles in the forward direction and obtain an ultrasonic radar feedback signal; adjusting the motion state of the cleaning robot according to the light intensity signal and the ultrasonic radar feedback signal", with reference to the embodiments.

[0044] First, a light intensity sensor detects the light intensity of the surrounding environment, such as changes in headlight illumination in a vehicle's parking garage, and generates a corresponding light intensity signal. This signal reflects the strength of light sources in the environment. Especially when a vehicle approaches, the light intensity sensor can quickly detect the strong glare from the headlights, alerting to potential dynamic collision risks. Ultrasonic radar, on the other hand, emits ultrasonic waves and receives reflected signals to detect obstacles in the robot's path, generating feedback signals. These feedback signals contain information about the distance and position of obstacles, making them particularly useful for low-lying or dark objects that are difficult for lidar to detect, such as tires or black bumpers. The combination of these two types of sensors allows the robot to perceive its environment from both lighting and physical obstacle perspectives, providing multi-layered collision avoidance capabilities. Light intensity sensors typically have a wide sensitivity range, capable of detecting changes from weak ambient light to strong light, with a response time generally less than 50 milliseconds, ensuring real-time performance. Ultrasonic radar typically has a detection range of less than 5 meters, analyzing the position and shape of obstacles through forward and lateral signals.

[0045] Secondly, the core control logic of this feature is adjusting the robot's motion state based on the intensity of light and ultrasonic radar feedback signals. Light intensity signals typically take precedence over ultrasonic radar signals because changes in light intensity (such as headlights illuminating) are often early warning signals of dynamic obstacles (such as approaching vehicles). For example, when the light sensor detects strong light, the robot may slow down to observe its environment; if the light intensity continues to increase, indicating that a vehicle may be very close, the robot will stop moving and issue an alarm. The ultrasonic radar feedback signal provides precise obstacle distance information; when an object is detected ahead, the robot adjusts its speed or stops based on the signal strength to avoid a collision. This tiered response mechanism ensures the robot reacts appropriately at different risk levels, improving both safety and cleaning efficiency.

[0046] In terms of implementation, the signal processing of the light intensity sensor and ultrasonic radar is completed by the robot's control module. The control module judges the signal strength through preset thresholds. Specifically, when the signal strength of the light intensity signal reaches the first threshold, the cleaning robot is controlled to decelerate; when the signal strength of the light intensity signal reaches the second threshold, the cleaning robot is controlled to stop and issue an alarm signal. When the signal strength of the ultrasonic radar feedback signal reaches the third threshold, the cleaning robot is controlled to decelerate; when the signal strength of the ultrasonic radar feedback signal reaches the fourth threshold, the cleaning robot is controlled to stop and issue an alarm signal. When the signal strength of the light intensity signal reaches the first threshold but the signal strength of the ultrasonic radar feedback signal does not reach the fourth threshold, the cleaning robot only decelerates without stopping.

[0047] For example, deceleration is triggered when the light intensity reaches 50% of the ambient light reference and stops when it reaches 80%; similarly, deceleration is triggered when the ultrasonic radar signal strength reaches 100% of the reference and stops when it reaches 120%. When the light intensity reaches a first threshold, such as 50% of the ambient light reference, it indicates that a vehicle may be approaching, but the ultrasonic radar has not detected a sufficiently close obstacle (i.e., the signal strength has not reached the alarm threshold, such as 120% of the reference). In this case, the robot only decelerates, for example, reducing its speed from the normal value to 0.1 m / s, to increase observation time and allow for possible subsequent stopping or obstacle avoidance maneuvers. This strategy avoids the inefficiency caused by frequent stops due to slight changes in light intensity, while improving alertness to potential dynamic obstacles through deceleration. These thresholds are based on the reference signal set during initialization in an obstacle-free environment, ensuring the accuracy of the judgment. Signal processing may also involve weighted fusion algorithms, such as Kalman filtering, to integrate the light intensity and ultrasonic signals and reduce noise interference.

[0048] This application employs multiple pairs of light intensity sensors and ultrasonic radars. Each pair of sensors independently detects data from one direction. The control module integrates the signals from multiple directions using a fusion algorithm (such as Kalman filtering) to generate a complete environmental perception map. For example, the left-side sensor pair might detect light intensity and obstacles approaching the vehicle, while the right-side sensor pair might detect unobstructed areas. The control module combines this data to determine whether to steer to the right to avoid obstacles. The redundant design of multiple sensor pairs also addresses the failure or interference of a single sensor. For instance, if a pair of ultrasonic radars is interfered with by metal reflections, sensors in other directions can still provide valid data, maintaining system stability.

[0049] The following describes in detail step 203, namely, "installing a lidar on the cleaning robot, wherein the lidar is turned on when it is necessary to scan the surrounding environment and / or at a preset time interval; using the lidar to scan the surrounding environment, generating a cleaning edge line, and dynamically correcting the cleaning edge line by comparing the positive and lateral feedback signals of the ultrasonic radar with their respective reference signals when approaching the vehicle."

[0050] A lidar sensor mounted on top of the robot scans the surrounding environment to generate a high-precision 3D point cloud map by emitting laser pulses and measuring reflection times. This point cloud map accurately depicts the outlines of objects in the environment, such as vehicles, walls, or other obstacles in an underground parking garage. Based on the point cloud data, the robot generates cleaning edges—virtual boundaries around the target area, such as a vehicle—to guide the cleaning path and ensure coverage of the target area. The generation of cleaning edges relies on the lidar's high resolution and ranging accuracy, typically identifying the positions of protruding components such as rearview mirrors, tires, and license plates, providing a basic path planning for subsequent cleaning tasks.

[0051] LiDAR requires significantly more power for operation and data processing than ultrasonic sensors, and its lifespan is directly related to operating time. To reduce system power consumption and extend LiDAR lifespan, it is activated at preset time intervals when scanning the surrounding environment is required. For example, it can be activated every 3 seconds during normal operation or as needed, then enter power-saving mode. Normal operation primarily utilizes ultrasonic sensors and light intensity sensors for obstacle avoidance. A cleaning robot typically moves very slowly at 0.8 meters per second, while ultrasonic radar can detect obstacles within a 5-meter range. Therefore, activating LiDAR every 3 seconds keeps it within the ultrasonic radar's detection radius. Furthermore, LiDAR primarily analyzes objects at greater distances, so it doesn't affect the robot's obstacle analysis and judgment. This significantly reduces the robot's overall power consumption.

[0052] Preferably, when using the LiDAR to scan the surrounding environment and generate the cleaning edge line, the LiDAR is used to scan the vehicle outline and the position of protruding parts is identified by ranging. The protruding parts include: rearview mirrors, tires and / or license plates; when the height of the protruding parts above the ground is higher than that of the cleaning robot, the protruding parts are not considered when generating the cleaning edge line.

[0053] Robotic cleaning systems are typically designed with a low profile, generally between 0.3 and 0.5 meters in height, suitable for cleaning areas near the ground. Certain protruding components, such as rearview mirrors, are usually located higher on the vehicle, for example, more than 1 meter above the ground, significantly taller than the robot. These components do not affect the robot's cleaning path and do not need to be considered obstacle avoidance objects. Therefore, when generating cleaning edges, the robot uses LiDAR ranging data to determine the height of components. If a component is higher than the top of the robot, it ignores it and generates edges only based on lower components (such as tires and license plates) or the vehicle's outline. This processing logic optimizes the generation of cleaning edges, reduces unnecessary path adjustments, and improves cleaning efficiency. In terms of implementation, the LiDAR scanning and data processing are coordinated by the robot's control module. LiDAR typically has 360-degree horizontal scanning capability, a scanning frequency of 10 to 20 times per second, a resolution of 0.1 degrees, and a ranging accuracy of ±2 centimeters. When scanning the vehicle's outline, the LiDAR generates point cloud data, and the control module analyzes the point cloud using algorithms to identify the coordinates and height of protruding components such as rearview mirrors, tires, and license plates. For example, the control module can determine whether the rearview mirror is higher than the top of the robot by using the height coordinates of the point cloud. If it is confirmed that the rearview mirror is higher than the robot, the system discards the coordinates of these points when generating the cleaning edge line, retaining only the data of components close to the ground to generate an edge line that fits the bottom of the vehicle. The edge line data is stored in the local path planning module for use in subsequent cleaning tasks.

[0054] When approaching a vehicle, the limitations of lidar become apparent, particularly due to blind spots when detecting low or dark objects. To address this, ultrasonic radar is introduced with forward and lateral feedback signals, dynamically correcting the cleaning edge by comparing them with their respective reference signals. The ultrasonic radar detects the distance and position of obstacles in front and to the sides by emitting ultrasonic waves and receiving reflected signals, generating forward and lateral feedback signals. The reference signal is the intensity of the ultrasonic wave reflection signal recorded in an unobstructed environment during initialization, representing the reflection characteristics of a normal cleaning edge. When the robot approaches a vehicle, if the forward and lateral signal intensities deviate from the reference signal—for example, below 80% or above 120%—it indicates an anomaly such as misaligned tires or a dark obstacle. Based on these signal differences, the control module adjusts the cleaning edge, for example, by extending or shifting it outward to ensure coverage of the cleaning area or to avoid obstacles.

[0055] In terms of implementation, the coordinated operation of LiDAR and ultrasonic radar is achieved through a control module. The LiDAR typically scans 10-20 times per second, with a resolution of 0.1 degrees and a ranging accuracy of ±2 cm. After generating the initial cleaning edge line, it is stored in the local path planning module. The ultrasonic radar generally has a detection range of less than 5 meters, covering both frontal and lateral areas, with a signal processing delay of less than 50 milliseconds. The control module calculates the edge line offset by comparing the ultrasonic radar feedback signal with a reference signal. For example, if both the frontal and lateral signal strengths are below 80% of the reference, it indicates the tire is misaligned. The robot approaches at low speed, calculates the offset angle and distance of the protruding part using triangulation, and extends the edge line outward by 5-10 cm. If the signal strength is above 120% of the reference, it indicates a black obstacle, the robot stops cleaning, and shifts the edge line by 10-15 cm. The corrected edge line is verified through continuous scanning to ensure accuracy.

[0056] The following describes step 204, namely, "When the light intensity signal reaches the activation threshold but the ultrasonic radar does not issue an alarm signal, the laser radar is activated to scan the direction of the light source. If no obstacle is detected, the robot continues to move forward. If an obstacle is detected in the forward direction, the robot pauses and rescans to confirm the obstacle's position before activating an obstacle avoidance strategy. The obstacle avoidance strategy includes: controlling the robot to rotate to the left at a fixed angular velocity outside the turning radius, while monitoring the signal strength of the forward and left-side ultrasonic radar and light intensity sensors, activating the laser radar scan at a predetermined angle, and recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal; returning to face the obstacle and rotating to the right, while monitoring the signal strength of the forward and right-side ultrasonic radar and light intensity sensors, activating the laser radar scan at a predetermined angle, and recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal; comparing the vanishing angles on both sides, selecting the direction with the smaller angle as the new forward path; if the angles on both sides are equal, using the laser radar scan data, the shortest trajectory from the current position to the original path is planned using a gridded space division and the A* algorithm as the new forward path."

[0057] A light intensity sensor detects ambient light intensity, such as changes in vehicle headlight illumination, and generates a light intensity signal. When the light intensity signal reaches a trigger threshold, such as 50% of the ambient light baseline, it indicates the presence of a potential dynamic obstacle, such as an approaching vehicle. The ultrasonic radar simultaneously detects the distance and position of the obstacle ahead, generating positive and lateral feedback signals. If the ultrasonic radar does not issue an alarm signal, for example, if the signal strength is below 120% of the baseline, it indicates that the obstacle is far away or does not yet pose a direct threat. In this case, the robot does not stop immediately but instead activates the lidar to scan the direction of the light source to further confirm the presence of an obstacle. This tiered response mechanism utilizes the early warning characteristics of light intensity signals, combined with the high-precision perception of lidar, to avoid unnecessary pauses and improve cleaning efficiency.

[0058] When the cleaning robot detects a new obstacle, such as a black object that is difficult for lidar to detect, the ultrasonic radar emits ultrasonic waves and receives the reflected signals to generate a feedback signal to determine the obstacle's presence and location. The strength of the feedback signal reflects the obstacle's distance and reflection characteristics; for example, a black bumper might cause the signal strength to be higher than the baseline value. Based on the change in signal strength, the robot determines that it needs to adjust its path to bypass the obstacle. The steering design outside the original turning radius ensures that the robot maintains a safe distance and avoids collisions when approaching obstacles. The steering process compares the signal changes on the left and right sides to select a better detour direction, thereby generating a new forward path.

[0059] Specifically, the robot first begins turning left or right from a safe distance of 2 meters outside the original turning radius. During the turn, the ultrasonic radar monitors the feedback signal strength on the left or right side in real time. When the signal strength drops below the reference signal, it indicates that the obstacle is no longer within the detection range in that direction. The robot records the turning angle at this time using a high-precision angle encoder, called the "vanishing angle." The robot first turns left, records the left vanishing angle, then returns to the initial position and turns right, recording the right vanishing angle. During the turn, this application initiates a laser radar scan at a predetermined angle, which can further improve the lifespan of the laser radar. This preset angle can be 60 degrees, and the laser radar is activated to perform a 120-degree scan when turning 60 degrees to the left or right (solid-state laser radar generally takes 1 second to start and scans 120 degrees). In addition, this preset angle can also be set and modified according to actual needs.

[0060] By comparing the vanishing angles on the left and right sides, the robot selects the direction with the smaller angle as the new forward path, because a smaller turning angle usually means a shorter detour, thus improving cleaning efficiency. If the angles on both sides are equal, the robot selects the direction with the smallest deviation from the original path to maintain path continuity.

[0061] In terms of implementation, the signal processing and steering control of the ultrasonic radar are handled by the robot's control module. The ultrasonic radar is typically installed in all directions (front, back, left, right), covering 360 degrees, with a detection distance of less than 5 meters and a signal processing delay of less than 50 milliseconds. The reference signal is the intensity of the ultrasonic reflected signal recorded in an unobstructed environment during initialization, used to compare changes in the current signal. For example, when the signal intensity exceeds 120% of the reference, path adjustment is triggered. A fixed angular velocity, such as 5 degrees per second, is used during steering to ensure smoothness and controllability. The angle encoder has an accuracy of 0.1 degrees, accurately recording the vanishing angle. The control module compares the left and right vanishing angles using an algorithm and combines this with the original path data to generate a new forward route. The new route is stored in the local path planning module and can be uploaded to a cloud platform to optimize the global path.

[0062] If the vanishing angles on both sides are equal, it indicates that the obstacle is symmetrical on both sides of the robot's forward direction, and a better detour direction cannot be determined solely based on the angle. In this case, the system further analyzes the environment using LiDAR scanning data, and calculates the shortest trajectory from the current position to the obstacle using a gridded spatial division and the A* algorithm. This mechanism avoids blindly selecting a detour direction, ensuring the efficiency of the path and the continuity of the cleaning task.

[0063] Mesh spatial partitioning is a fundamental step in path planning. LiDAR generates 3D point cloud data of the environment through high-precision scanning. The control module converts this data into a mesh map, dividing the environment into equidistant grid cells, such as those with sides of 0.4 meters. Each grid cell represents a movable location, marked as either passable or an obstacle area. This discretization simplifies the representation of complex environments, enabling path planning algorithms to operate efficiently with limited computational resources. The mesh map is based on 120-degree scan data acquired by the LiDAR when turning left or right to a predetermined angle (e.g., 60 degrees), ensuring the map reflects the latest environmental information around obstacles, thus providing accurate input for the A* algorithm.

[0064] The A* algorithm is used to calculate the shortest trajectory. It is a heuristic search algorithm that selects the optimal path using an evaluation function f(n) = g(n) + h(n), where g(n) is the actual cost of movement from the current position to the current grid, such as distance or time; and h(n) is the estimated cost from the current grid to the target point (the original path after the obstacle), typically based on Euclidean distance. For example, starting from the current position and targeting the original path after the obstacle, the A* algorithm prioritizes expanding the grid with the smallest f(n), gradually constructing the shortest trajectory. The algorithm manages the grids to be explored and those already explored using open and closed lists, ensuring that the optimal path is maintained while reducing computational cost. The generated trajectory serves as a new forward route, guiding the robot around obstacles and back to the original path.

[0065] This application also controls the cleaning robot through a cloud platform. The control includes: reporting the detection signals of the light intensity sensor, the ultrasonic radar, and the lidar to the cloud platform; if the alarm information is generated during the cleaning process, reporting the alarm information to the cloud platform; and guiding and planning for potential collisions between multiple cleaning robots.

[0066] The core role of the cloud platform in controlling a robotic vacuum cleaner is to receive and process signals from multiple sensors. The robot is equipped with a light intensity sensor, ultrasonic radar, and lidar, used to detect ambient light intensity, obstacle distance, and environmental contours, respectively. The detection signals generated by these sensors, including light intensity signals, forward and lateral feedback signals from the ultrasonic radar, and point cloud data from the lidar, are uploaded to the cloud platform in real time via a communication module. As a data processing center, the cloud platform can store and analyze these signals, providing global environmental perception capabilities. For example, the light intensity sensor detects changes in headlight intensity, the ultrasonic radar identifies black obstacles, and the lidar generates the vehicle's outline. This integrated information helps the cloud platform build a real-time environmental map, supporting path planning and obstacle avoidance decisions.

[0067] During the cleaning process, when the light intensity sensor detects that the light intensity reaches the alarm threshold (e.g., 80% of the ambient light baseline), or the ultrasonic radar feedback signal strength reaches 120% of the baseline, the robot will trigger an alarm, such as an audible and visual alarm, and generate alarm information including obstacle location, signal strength, and timestamp. This alarm information is uploaded to the cloud platform via the communication module for real-time monitoring by administrators or the back-end system. The cloud platform can analyze potential risks based on the alarm information, such as identifying areas where alarms are frequently triggered, optimizing cleaning strategies, or adjusting robot deployment. Furthermore, centralized management of alarm information facilitates post-event analysis and system optimization, such as improving sensor thresholds or path algorithms by statistically analyzing alarm frequency.

[0068] Furthermore, this feature emphasizes the cloud platform's guidance and planning capabilities in multi-robot collaboration. In scenarios such as underground parking garages, multiple cleaning robots may operate simultaneously, posing a potential collision risk. The cloud platform receives the position, path, and sensor data of each robot, analyzes their movement trajectories in real time, and detects potential intersections or proximity. For example, when two robots are less than 1 meter apart, the cloud platform adjusts path priorities and plans coordinated routes to ensure the robots do not interfere with each other. This multi-robot coordination function relies on the cloud platform's high-performance computing and real-time communication capabilities, enabling dynamic optimization of global cleaning task allocation, improving overall efficiency, and reducing collision risks.

[0069] In terms of implementation, communication between the cloud platform and the cleaning robot is typically achieved via Wi-Fi or 4G networks, with data upload rates exceeding 10Mbps to ensure real-time performance. Sensor signals are preprocessed using algorithms such as Kalman filtering before uploading, fusing light intensity, ultrasonic, and lidar data to improve data quality. The cloud platform employs a distributed computing architecture to process data from multiple robots, running path planning algorithms such as A* or dynamic programming to generate collision-free coordinated routes. Alarm information processing involves database storage and a real-time notification system, allowing administrators to receive alarms and intervene remotely via a management application. The cloud platform's response latency is typically controlled within 100 milliseconds to meet the demands of dynamic environments.

[0070] Preferably, the cleaning robot of this application has a polyurethane anti-collision outer shell coated with a reflective coating. Polyurethane is a high-strength, highly elastic polymer material with excellent impact resistance and energy absorption properties. In environments such as underground parking garages, cleaning robots may experience minor collisions with vehicles, walls, or other objects due to path errors or dynamic obstacles. The polyurethane anti-collision outer shell can absorb the energy generated by the collision, reducing damage to the robot's internal components and external objects. For example, compared to hard metal or plastic shells, the polyurethane outer shell can disperse the impact force through elastic deformation upon collision, protecting the robot's sensors and drive system while reducing the potential risk of scratches to high-value objects such as vehicles. The reflective coating on the outer shell surface further enhances the robot's safety in complex environments. The reflective coating is typically made of highly reflective materials, such as coatings containing microglass beads, which can reflect light back to the direction of the light source under illumination. In poorly lit scenarios such as underground parking garages, when vehicle headlights or other light sources shine on the robot's outer shell, the reflective coating produces a bright reflective effect, providing a clear visual cues to drivers or pedestrians. This cues significantly improve the robot's visibility, especially at night or in low-light environments, reducing the risk of collisions due to limited visibility. The reflective coating is also designed for durability, resisting wear and tear during cleaning and environmental corrosion.

[0071] The method provided in this application can be applied to various scenarios, including but not limited to: In underground parking garages, where vehicles are densely packed and lighting is insufficient, obstacles such as misaligned tires and black bumpers, which are difficult for LiDAR to detect, exist. This method uses a light intensity sensor to detect changes in headlight intensity, ultrasonic radar to detect low-lying or black objects, and LiDAR to generate and dynamically correct cleaning edges, ensuring the robot bypasses obstacles and covers the cleaning area, such as the tire area under vehicles. A cloud platform coordinates multiple robots in real time to avoid collisions and uploads alarm information for monitoring by management personnel, significantly improving cleaning efficiency and safety. In large shopping mall parking lots, where vehicles move frequently, a reflective coating enhances robot visibility, and a polyurethane anti-collision coating reduces collision damage. The robot quickly plans detour routes by selecting the steering angle using ultrasonic radar, adapting to environments with frequent vehicle entry and exit. In these scenarios, this method combines multi-sensor collaboration, dynamic path planning, and cloud platform control, balancing cleaning coverage and collision avoidance capabilities. It is particularly suitable for complex locations with dense high-value objects and insufficient lighting, demonstrating an efficient and economical cleaning solution.

[0072] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0073] According to another embodiment, a collision avoidance system for a cleaning robot is provided. Figure 3 A schematic block diagram of a collision avoidance system for a cleaning robot according to one embodiment is shown. Figure 3 As shown, the device 300 includes:

[0074] Multiple light intensity sensors and ultrasonic radar pairs 301 are installed on the cleaning robot. Each pair of light intensity sensors and ultrasonic radars is installed in the same direction and is used to detect ambient light intensity to generate light intensity signals and detect obstacles in the forward direction to generate ultrasonic radar feedback signals.

[0075] The lidar 302 is installed on the cleaning robot and is used to scan the surrounding environment to generate cleaning edge lines. The lidar is turned on when it is necessary to scan the surrounding environment and / or at preset time intervals.

[0076] Control module 303, communicatively connected to the light intensity sensor, ultrasonic radar, and lidar, is configured to: adjust the movement state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal; generate an initial cleaning edge line based on the lidar scanning data, and dynamically correct the cleaning edge line when approaching a vehicle by comparing the forward and lateral feedback signals of the ultrasonic radar with their respective reference signals; when the light intensity signal reaches a first threshold but the ultrasonic radar does not issue an alarm signal, activate the lidar to scan the light source direction; if no obstacle is detected, continue moving forward; if an obstacle is detected in the forward direction, pause and rescan to confirm the obstacle's position before activating an obstacle avoidance strategy; the obstacle avoidance strategy includes: controlling the robot to rotate... The device rotates to the left at a fixed angular velocity outside the curve radius, while simultaneously monitoring the signal strength of the ultrasonic radar and light intensity sensors in the forward and left directions. At a predetermined angle, a lidar scan is initiated, recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal. After returning to face the obstacle, the device rotates to the right, simultaneously monitoring the signal strength of the ultrasonic radar and light intensity sensors in the forward and right directions. At a predetermined angle, a lidar scan is initiated, recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal. The vanishing angles on both sides are compared, and the direction with the smaller angle is selected as the new forward path. If the angles on both sides are equal, the lidar scan data is used to plan the shortest trajectory from the current position to the original path using a gridded space and the A* algorithm, which is then used as the new forward path.

[0077] As one possible implementation, the control module 303, when adjusting the motion state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal, can be configured to: control the cleaning robot to decelerate when the signal strength of the light intensity signal reaches a first threshold; control the cleaning robot to stop and issue an alarm signal when the signal strength of the light intensity signal reaches a second threshold; control the cleaning robot to slow down when the signal strength of the ultrasonic radar feedback signal reaches a third threshold; control the cleaning robot to stop and issue an alarm signal when the signal strength of the ultrasonic radar feedback signal reaches a fourth threshold; and control the cleaning robot to only decelerate without stopping when the signal strength of the light intensity signal reaches the first threshold but the signal strength of the ultrasonic radar feedback signal does not reach the fourth threshold.

[0078] As an implementable approach, when the control module 303 uses the lidar to scan the surrounding environment and generate cleaning edge lines, it can also be configured to: use the lidar to scan the vehicle outline and identify the position of protruding parts by ranging, the protruding parts including: rearview mirrors, tires and / or license plates; when the height of the protruding parts above the ground is higher than that of the cleaning robot, the protruding parts are not considered when generating cleaning edge lines.

[0079] As an implementable approach, when the control module 303 dynamically corrects the cleaning edge line by comparing the forward and lateral feedback signals of the ultrasonic radar with their respective reference signals as it approaches the vehicle, it can be configured as follows: when the forward and lateral signal strengths of the ultrasonic radar are both lower than 80% of the reference signal, it is determined that the tire is misaligned, and the robot approaches the obstacle at a low speed, calculates the offset angle and distance of the protruding part, and extends the cleaning edge line outward by 5 to 10 centimeters to cover the rear area of ​​the tire; when the signal strength is higher than 120% of the reference signal, it is determined to be a black obstacle, and the robot stops cleaning and shifts the edge line outward by 10 to 15 centimeters according to the ranging data to avoid collision.

[0080] As an feasible approach, the method of using LiDAR scanning data to plan the shortest trajectory from the current position to the original path as a new forward route through gridded spatial division and the A* algorithm includes: converting the LiDAR scanning data into a gridded map, dividing the environment into grid cells with a side length of 0.4 meters; based on the gridded map, taking the current position as the starting point and the original path behind the obstacle as the target point, using the A* algorithm to calculate the shortest trajectory, wherein the A* algorithm combines the actual movement cost from the starting point to the current grid and the estimated cost from the current grid to the target point through an evaluation function, and selects the grid with the lowest cost for expansion; when turning left or right to 60 degrees, activating the LiDAR to perform a 120-degree angle scan to acquire local environmental data to update the gridded map; generating a new forward route based on the shortest trajectory calculated by the A* algorithm and controlling the cleaning robot to move along the route.

[0081] As an implementable approach, the cleaning robot anti-collision system also controls the cleaning robot through a cloud platform. The control includes: reporting the detection signals from the light intensity sensor, the ultrasonic radar, and the lidar to the cloud platform; reporting the alarm information to the cloud platform if the alarm information is generated during the cleaning process; and guiding and planning for potential collisions between multiple cleaning robots.

[0082] As an feasible approach, the cleaning robot's shell is made of polyurethane anti-collision material and coated with a reflective coating.

[0083] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. Components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0084] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0085] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0086] And an electronic device comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0087] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0088] in, Figure 4 An exemplary architecture of an electronic device is shown, which may include a processor 410, a video display adapter 411, a disk drive 412, an input / output interface 413, a network interface 414, and a memory 420. The processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420 can communicate with each other via a communication bus 430.

[0089] The processor 410 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

[0090] The memory 420 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 420 can store the operating system 421 for controlling the operation of the electronic device 400, and the basic input / output system (BIOS) 422 for controlling the low-level operations of the electronic device 400. Additionally, it can store a web browser 423, a data storage management system 424, and a robot vacuum anti-collision system 425, etc. The aforementioned robot vacuum anti-collision system 425 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 420 and executed by the processor 410.

[0091] Input / output interface 413 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0092] Network interface 414 is used to connect a communication module (not shown in the figure) to enable communication and interaction between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0093] Bus 430 includes a pathway for transmitting information between various components of the device, such as processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420.

[0094] It should be noted that although the above-described device only shows the processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, memory 420, bus 430, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0095] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer program product. This computer program product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0096] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A low-power anti-collision method for a cleaning robot, characterized in that, The method includes: Multiple pairs of light intensity sensors and ultrasonic radars are installed on the cleaning robot, with each pair of light intensity sensors and ultrasonic radars installed in the same direction. The ambient light intensity is detected by the light intensity sensor to obtain a light intensity signal; obstacles in the forward direction are detected by the ultrasonic radar to obtain an ultrasonic radar feedback signal; the movement state of the cleaning robot is adjusted according to the light intensity signal and the ultrasonic radar feedback signal; wherein, each pair of sensors independently detects data in one direction, and the multi-directional signals are integrated by a fusion algorithm to generate a complete environmental perception map; A lidar is installed on the cleaning robot. The lidar is activated when it is necessary to scan the surrounding environment and / or at preset time intervals. The lidar is used to scan the surrounding environment and generate cleaning edge lines. The lidar is also used to scan the vehicle outline and identify the position of protruding parts by ranging. The protruding parts include: rearview mirrors, tires and / or license plates. When the height of the protruding parts above the ground is higher than that of the cleaning robot, the protruding parts are not considered when generating the cleaning edge lines. When the forward and lateral signal strengths of the ultrasonic radar are both lower than 80% of the reference signal, it is determined that the tire is misaligned. The robot approaches the obstacle at a low speed, calculates the offset angle and distance of the protruding part, and extends the cleaning edge 5 to 10 centimeters outward to cover the rear area of ​​the tire. When the signal strength is higher than 120% of the reference signal, it is identified as a black obstacle. The robot stops cleaning and shifts outward by 10 to 15 centimeters according to the ranging data to avoid collision. When the light intensity signal reaches the activation threshold but the ultrasonic radar does not issue an alarm signal, the laser radar is activated to scan the direction of the light source. If no obstacle is detected, the system continues to move forward. If an obstacle is detected in the forward direction, the system pauses and rescans to confirm the obstacle's position before activating the obstacle avoidance strategy. The obstacle avoidance strategy includes: controlling the robot to rotate to the left at a fixed angular velocity outside the turning radius, while monitoring the signal strength of the forward and left-side ultrasonic radar and light intensity sensors; initiating a laser radar scan at a predetermined angle and recording the left vanishing angle when the left ultrasonic radar signal strength drops below the reference signal; returning to face the obstacle and rotating to the right, while monitoring the signal strength of the forward and right-side ultrasonic radar and light intensity sensors; initiating a laser radar scan at a predetermined angle and recording the right vanishing angle when the right ultrasonic radar signal strength drops below the reference signal; comparing the vanishing angles on both sides, selecting the direction with the smaller angle as the new forward path; if the angles on both sides are equal, converting the laser radar scan data into a gridded map, dividing the environment into grid cells with a side length of 0.4 meters; Based on the gridded map, with the current position as the starting point and the original path after the obstacle as the target point, the shortest trajectory is calculated using the A* algorithm. The A* algorithm combines the actual movement cost from the starting point to the current grid and the estimated cost from the current grid to the target point through an evaluation function, and selects the grid with the minimum cost for expansion. When the left and right turns reach 60 degrees, the lidar is activated to perform a 120-degree angle scan to acquire local environmental data and update the gridded map. Based on the shortest trajectory calculated by the A* algorithm, a new forward path is generated and the cleaning robot is controlled to move along that path.

2. The method according to claim 1, characterized in that, The step of adjusting the motion state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal includes: When the signal strength of the light intensity signal reaches the first threshold, the cleaning robot is controlled to decelerate; when the signal strength of the light intensity signal reaches the second threshold, the cleaning robot is controlled to stop and an alarm signal is issued. When the signal strength of the ultrasonic radar feedback signal reaches the third threshold, the cleaning robot is controlled to slow down. When the signal strength of the ultrasonic radar feedback signal reaches the fourth threshold, the cleaning robot is controlled to stop and issue an alarm signal. When the signal strength of the light intensity signal reaches the first threshold but the signal strength of the ultrasonic radar feedback signal does not reach the fourth threshold, the cleaning robot only slows down and does not stop.

3. The method according to claim 1, characterized in that, The method further includes: controlling the cleaning robot via a cloud platform, wherein the control includes: The detection signals from the light intensity sensor, the ultrasonic radar, and the lidar are reported to the cloud platform; If the alarm information is generated during the cleaning process, the alarm information will be reported to the cloud platform; Guide and plan for potential collisions between multiple cleaning robots.

4. The anti-collision method for a cleaning robot according to any one of claims 1-3, characterized in that, The cleaning robot's shell is made of polyurethane anti-collision material and coated with a reflective coating.

5. A low-power anti-collision system for a cleaning robot, characterized in that, The system includes: Multiple light intensity sensors and ultrasonic radar pairs are installed on the cleaning robot. Each pair of light intensity sensors and ultrasonic radars is installed in the same direction and is used to detect ambient light intensity to generate light intensity signals and detect obstacles in the direction of movement to generate ultrasonic radar feedback signals. A lidar, installed on the cleaning robot, is used to scan the surrounding environment to generate cleaning edge lines. The lidar is turned on when it is necessary to scan the surrounding environment and / or at preset time intervals. The control module, communicatively connected to the light intensity sensor, ultrasonic radar, and lidar, is configured to: adjust the motion state of the cleaning robot based on the light intensity signal and the ultrasonic radar feedback signal; wherein each pair of sensors independently detects data in one direction, and integrates multi-directional signals through a fusion algorithm to generate a complete environmental perception map; generate an initial cleaning edge line based on the lidar scanning data, scan the vehicle outline using the lidar, and identify the position of protruding components through ranging, the protruding components including: rearview mirrors, tires, and / or license plates; when the height of the protruding component above the ground is higher than that of the cleaning robot, the protruding component is not considered when generating the cleaning edge line; when the forward and lateral signal strengths of the ultrasonic radar are both lower than 80% of the reference signal... If the robot detects a misaligned tire, it approaches the obstacle at low speed, calculates the offset angle and distance of the protruding part, and extends its cleaning edge by 5 to 10 centimeters to cover the rear area of ​​the tire. When the signal strength is higher than 120% of the reference signal, it is identified as a black obstacle. The robot stops cleaning and shifts its cleaning edge by 10 to 15 centimeters based on the ranging data to avoid collision. When the light intensity signal reaches a first threshold but the ultrasonic radar does not issue an alarm signal, the laser radar is activated to scan the direction of the light source. If no obstacle is detected, it continues to move forward. If an obstacle is detected in the forward direction, it pauses and rescans to confirm the obstacle's position before activating the obstacle avoidance strategy. The obstacle avoidance strategy includes... The robot is controlled to rotate to the left at a fixed angular velocity outside the turning radius, while simultaneously monitoring the signal strength of the ultrasonic radar and light intensity sensors on both the front and left sides. A laser radar scan is initiated at a predetermined angle, recording the left vanishing angle when the left ultrasonic radar signal strength drops below a reference signal. After returning to face the obstacle, the robot rotates to the right, simultaneously monitoring the signal strength of the ultrasonic radar and light intensity sensors on both the front and right sides. A laser radar scan is initiated at a predetermined angle, recording the right vanishing angle when the right ultrasonic radar signal strength drops below a reference signal. The vanishing angles on both sides are compared, and the direction with the smaller angle is selected as the new forward path. If the angles on both sides are equal, the laser radar... The scanned data is converted into a gridded map, dividing the environment into grid cells with a side length of 0.4 meters. Based on the gridded map, with the current position as the starting point and the original path after the obstacle as the target point, the shortest trajectory is calculated using the A* algorithm. The A* algorithm combines the actual movement cost from the starting point to the current grid and the estimated cost from the current grid to the target point through an evaluation function, and selects the grid with the lowest cost for expansion. When turning left or right to 60 degrees, the LiDAR is activated to perform a 120-degree angle scan to acquire local environmental data to update the gridded map. According to the shortest trajectory calculated by the A* algorithm, a new forward route is generated and the cleaning robot is controlled to move along the route.

6. An electronic device, characterized in that, include: One or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 4.

Images