Obstacle avoidance and navigation method and device for sweeping robot based on TOF sensor and storage medium

Abstract

The application discloses a kind of based on TOF sensor's sweeping machine obstacle avoidance and navigation method, device and storage medium, above-mentioned method includes: whether TOF sensor is located at the front end of sweeping machine, if in front end, according to mixed exposure mode, obtain the data of different exposure time length to form the point cloud data for obstacle avoidance and navigation, otherwise, according to automatic exposure mode, obtain the point cloud data for navigation.Therefore only need to process TOF sensor data, obstacle avoidance and navigation can be realized by setting different exposure mode of TOF sensor, reduce the difficulty of research and development, improve data processing speed and sweeping machine work efficiency.

Description

Technical Field

[0001] This invention relates to the field of sweeping machine technology, and in particular to a sweeping machine obstacle avoidance and navigation method, device and storage medium based on TOF sensor. Background Technology

[0002] Currently, consumer-grade mobile robots, especially intelligent robotic vacuum cleaners, not only need to have intelligent obstacle avoidance capabilities, but also need to plan their routes and cover as much indoor area as possible. Therefore, existing robotic vacuum cleaners often use multiple sensors to work together to achieve obstacle avoidance and navigation functions.

[0003] For example, it is common to install LiDAR on the top of a robot vacuum cleaner to achieve indoor SLAM (Simultaneous Localization and Mapping). However, to achieve obstacle avoidance in complex environments, other sensors, such as infrared obstacle avoidance sensors and ultrasonic obstacle avoidance sensors, are also required.

[0004] Different processing methods are required for data from different sensors, which increases the difficulty of developing robotic vacuum cleaners and reduces the data processing speed and work efficiency during use.

[0005] Therefore, existing technologies still need to be improved and enhanced. Summary of the Invention

[0006] The main objective of this invention is to provide a method, device, storage medium, and sweeping machine for obstacle avoidance and navigation based on a TOF sensor, aiming to solve the problems of existing sweeping machines using multiple sensors for obstacle avoidance and navigation, slow data processing speed, and low work efficiency.

[0007] To achieve the above objectives, a first aspect of the present invention provides a method for obstacle avoidance and navigation of a robotic vacuum cleaner based on a Time-of-Flight (TOF) sensor. The method includes: obtaining the orientation of the TOF sensor on the robotic vacuum cleaner based on its direction of travel; when the TOF sensor is located at the front end of the robotic vacuum cleaner, obtaining first exposure data of the TOF sensor at a first exposure duration and second exposure data at a second exposure duration based on a preset hybrid exposure mode, wherein the first exposure duration is shorter than the second exposure duration, and the first and second exposure durations form an exposure cycle; fusing the first and second exposure data to obtain first point cloud data reflecting near and far spatial locations; controlling the movement of the robotic vacuum cleaner based on the first point cloud data to achieve obstacle avoidance and navigation; when the TOF sensor is located at the rear end or side end of the robotic vacuum cleaner, obtaining automatic exposure data of the TOF sensor based on a preset automatic exposure mode; obtaining second point cloud data reflecting far spatial locations based on the automatic exposure data; and controlling the movement of the robotic vacuum cleaner based on the second point cloud data to achieve navigation.

[0008] Optionally, multiple TOF sensors are set to an automatic exposure mode. Obtaining the automatic exposure data of the TOF sensors based on the preset automatic exposure mode includes: asynchronously acquiring the exposure data of the TOF sensors configured in automatic exposure mode; and fusing all the exposure data according to a spatial synchronization method to obtain the automatic exposure data.

[0009] Optionally, the step of obtaining the first exposure data of the TOF sensor under the first exposure time and the second exposure data under the second exposure time based on the preset hybrid exposure mode further includes: judging the second exposure data according to the overexposure identification method; when the second exposure data is overexposed, correcting the second exposure data based on the first exposure data.

[0010] Optionally, correcting the second exposure data based on the first exposure data includes: sequentially obtaining each position of overexposed data in the second exposure data, and replacing the data in the second exposure data corresponding to the position with the data in the first exposure data corresponding to the position.

[0011] Optionally, when the second exposure data is normal, the method further includes: setting the second exposure duration as the next exposure cycle to obtain exposure data; judging the exposure data according to the overexposure identification method; when the exposure data is normal, obtaining the first point cloud data based on the exposure data; when the exposure data is overexposed, setting the first exposure duration and the second exposure duration as the next exposure cycle.

[0012] Optionally, after obtaining the first point cloud data based on the exposure data, the method further includes: continuously setting the second exposure duration as the next exposure cycle until the exposure data is overexposed.

[0013] Optionally, fusing the first exposure data and the second exposure data to obtain first point cloud data reflecting near-field and far-field spaces includes: performing phase demodulation and depth calculation on the first exposure data and the second exposure data respectively to obtain depth data; and fusing all the depth data according to a time and space synchronization method to obtain the first point cloud data.

[0014] A second aspect of the present invention provides a robot vacuum cleaner obstacle avoidance and navigation device based on a Time-of-Flight (TOF) sensor. The device includes: a position acquisition module for obtaining the position of the TOF sensor on the robot vacuum cleaner based on the robot vacuum cleaner's direction of travel; an exposure data acquisition module for obtaining first exposure data of the TOF sensor at a first exposure duration and second exposure data at a second exposure duration based on a preset hybrid exposure mode; or, obtaining automatic exposure data of the TOF sensor based on a preset automatic exposure mode; a point cloud data acquisition module for fusing the first and second exposure data to obtain first point cloud data reflecting near-field and far-field spaces; or, obtaining second point cloud data reflecting far-field spaces based on the automatic exposure data; and a control module for controlling the robot vacuum cleaner to move based on the first point cloud data to achieve obstacle avoidance and navigation; or, controlling the robot vacuum cleaner to move based on the second point cloud data to achieve navigation.

[0015] A third aspect of the present invention provides a robotic vacuum cleaner, including a memory, a processor, and a plurality of TOF sensors spaced apart on the side of the robotic vacuum cleaner. The memory stores a robotic vacuum cleaner obstacle avoidance and navigation program that can run on the processor. When the robotic vacuum cleaner executes the obstacle avoidance and navigation program, it implements any one of the above-mentioned obstacle avoidance and navigation methods based on TOF sensors.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium storing a robot vacuum cleaner obstacle avoidance and navigation program based on a TOF sensor, wherein the robot vacuum cleaner obstacle avoidance and navigation program based on a TOF sensor, when executed by a processor, implements any of the steps of the above-described robot vacuum cleaner obstacle avoidance and navigation method based on a TOF sensor.

[0017] As can be seen from the above, compared with the prior art, the present invention determines whether the TOF sensor is located at the front of the robot vacuum based on the robot's direction of travel. If it is at the front, it obtains data with different exposure durations using a hybrid exposure mode to form point cloud data for obstacle avoidance and navigation; otherwise, it obtains point cloud data for navigation using an automatic exposure mode. Therefore, only the data from the TOF sensor needs to be processed. By setting different exposure modes for the TOF sensor, obstacle avoidance and navigation can be achieved, reducing development difficulty and improving data processing speed and robot vacuum efficiency. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the obstacle avoidance and navigation method for a robot vacuum cleaner based on a TOF sensor provided in an embodiment of the present invention;

[0020] Figure 2 yes Figure 1 Schematic diagram of hybrid mode exposure in the embodiment;

[0021] Figure 3 yes Figure 1 A schematic diagram of the specific process of step A200 in the embodiment;

[0022] Figure 4 yes Figure 1 A schematic diagram illustrating the specific process when long exposure data is normal in the embodiment;

[0023] Figure 5 This is a schematic diagram of the obstacle avoidance and navigation device for a sweeping robot based on a TOF sensor provided in an embodiment of the present invention. Detailed Implementation

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0025] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0026] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0027] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0028] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrases "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0031] Robotic vacuum cleaners are a type of smart home appliance that uses artificial intelligence to automatically clean floors in a room. Therefore, they need to possess intelligent obstacle avoidance and route planning capabilities. Currently, they often use LiDAR as the primary sensor for positioning and navigation, combined with other sensors such as ultrasonic and infrared sensors for obstacle avoidance.

[0032] The difficulty in developing robotic vacuum cleaners stems from the fact that data from different sensors requires different processing methods and is difficult to integrate. This increases the complexity of data processing and reduces efficiency during use. Furthermore, the characteristics of LiDAR products mean that installing LiDAR inside a robotic vacuum cleaner inevitably leads to a significant reduction in the usable field of view (FOV). The complex internal structure also causes issues such as light path obstruction and multiple reflections of stray light, significantly degrading LiDAR data quality and affecting the accuracy of positioning and navigation algorithms. Therefore, LiDAR is commonly installed on the top of the robotic vacuum cleaner, which increases the overall height and hinders its ability to clean low-ceilinged indoor spaces.

[0033] In recent years, 3D vision imaging technology has been increasingly applied across various industries, with Time-of-Flight (TOF) technology being one of its key technologies. TOF-based depth cameras use an active imaging method, where the camera system emits a laser towards the target and measures the distance between the target and the image sensor by the time it takes for the light to travel from the light source to the object and back. TOF depth cameras are characterized by their small size, low error, and strong anti-interference capabilities. They can not only acquire high-precision images but also perform functions such as object recognition and obstacle detection. Furthermore, TOF depth calculations are unaffected by the surface grayscale and features of the target object, allowing for highly accurate 3D image detection with minimal computational load, making them suitable for various indoor and outdoor applications.

[0034] This invention provides a method for obstacle avoidance and navigation of a robotic vacuum cleaner based on a Time-of-Flight (TOF) sensor. By setting different exposure modes for the TOF sensors located at different positions on the robotic vacuum cleaner, obstacle avoidance and navigation can be achieved by processing only the data from the TOF sensor, reducing the difficulty of research and development and improving the data processing speed and the working efficiency of the robotic vacuum cleaner.

[0035] Exemplary methods

[0036] like Figure 1 As shown, this embodiment provides a method for obstacle avoidance and navigation of a robotic vacuum cleaner based on a TOF sensor. For ease of description and understanding, the first exposure duration, first exposure data, second exposure duration, and second exposure data in this invention will be referred to as short exposure duration, short exposure data, long exposure duration, and long exposure data, respectively, in the following description.

[0037] Specifically, the above method includes the following steps:

[0038] Step S100: Based on the direction of travel of the sweeping robot, obtain the orientation of the TOF sensor on the sweeping robot;

[0039] Specifically, based on the sweeper's direction of travel, the TOF sensor's position on the sweeper can be categorized as front, rear, left, and right. In this embodiment, the sweeper's direction of travel is first obtained from its control processor, and the light emission direction is obtained from the TOF sensor. If the angle between the direction of travel and the light emission direction is less than a preset angle threshold (e.g., 15 degrees), the TOF sensor is determined to be located at the front of the sweeper. Obviously, the position of the TOF sensor on the sweeper during operation can also be obtained based on the sweeper's direction of travel, rotation angle, and the initial installation position of the TOF sensor.

[0040] Based on the above judgment, if the TOF sensor is located at the front of the robot vacuum:

[0041] Step A200: Based on the preset hybrid exposure mode, obtain the short exposure data of the TOF sensor under short exposure duration and the long exposure data under long exposure duration, wherein the short exposure duration and the long exposure duration form an exposure cycle;

[0042] Specifically, the exposure mode based on the TOF sensor can be set. This invention sets two working modes for the TOF sensor: automatic exposure mode and hybrid exposure mode. The automatic exposure mode allows the TOF sensor to select the exposure time, which is used for long-distance SLAM functionality and can improve frame rate and reduce power consumption. The hybrid exposure mode, as shown in the image, allows the TOF sensor to select the exposure time, which is suitable for long-distance SLAM functions and can improve frame rate and reduce power consumption. Figure 2 As shown, an interval exposure combining set long and short exposure durations is used. One exposure cycle includes a long exposure duration and a short exposure duration, with the short exposure duration being shorter than the long exposure duration. Long exposure data and short exposure data are obtained separately to simultaneously meet the requirements of near-range obstacle avoidance and far-range SLAM functions. This can improve the adaptability of the detection environment in situations where there are different reflectivities in near-range and far-range detection environments.

[0043] Therefore, if the TOF sensor is located at the front of the robot vacuum cleaner, the TOF sensor is set to a hybrid exposure mode, and the TOF sensor data with a short exposure time is obtained as short exposure data and the TOF sensor data with a long exposure time is obtained as long exposure data.

[0044] In the mixed exposure mode, the long and short exposure times can be selected according to specific circumstances, typically 50µs + 300µs or 300µs + 1000µs. 50µs + 300µs is suitable for situations with a small field of view (FOV) and a short working distance; 300µs + 1000µs is suitable for situations with a large FOV and a long working distance.

[0045] Preferably, in this embodiment, two iTOF (indirect TOF technology) 3D sensors are mounted opposite each other on the side of the sweeping machine. When the sweeping machine is running, the iTOF 3D sensor located at the front of the sweeping machine operates in a mixed exposure mode, while the iTOF 3D sensor located at the rear of the sweeping machine operates in an automatic exposure mode.

[0046] Furthermore, when using two iTOF 3D sensors, the operating modes of the two sensors can be set: simultaneous / synchronous operation or time-sharing / asynchronous operation. Synchronous operation improves the real-time performance of obstacle avoidance and SLAM navigation; asynchronous operation reduces the requirements on the overall depth calculation system of the robot vacuum, thus reducing the development difficulty and cost of the robot vacuum system.

[0047] Step A300: Fuse the short exposure data and long exposure data to obtain the first point cloud data reflecting near-field and far-field space;

[0048] Step A400: Based on the first point cloud data, control the robot vacuum cleaner to move to achieve obstacle avoidance and navigation.

[0049] Specifically, the TOF sensor directly outputs the 3D data of the object being measured. This means that the short-exposure and long-exposure data obtained are 3D data, specifically 3D image data. The short-exposure data represents the 3D data of nearby objects, while the long-exposure data represents the 3D data of distant objects. Phase demodulation and depth calculation are performed on the short-exposure and long-exposure data respectively to obtain depth data. Then, all the depth data are fused using a time-space synchronization method to obtain the first point cloud data. Based on this point cloud data, the distances between various objects in the indoor space and the robot vacuum are calculated for localization and map construction, thereby controlling the robot vacuum's movement to achieve obstacle avoidance and navigation.

[0050] Phase demodulation and depth calculation are commonly used optical signal processing methods. They are mainly used to calculate the depth information of the target after the TOF sensor emits a light wave of a specific wavelength, which is reflected on the surface of the target and received by the TOF sensor. The depth information of the target is calculated based on the time difference or phase difference between the emitted and received light waves.

[0051] Time synchronization involves adding timestamps to the independently collected data from the TOF sensors to ensure that all TOF sensors have synchronized timestamps. Spatial synchronization involves transforming data from different TOF sensor coordinate systems to the same coordinate system (such as the coordinate system of a robot vacuum cleaner). After synchronizing all depth data in both time and space, the depth data is fused according to multi-sensor fusion theory to obtain the first point cloud data.

[0052] This embodiment uses a long and short exposure interval method, which can improve the phase signal quality of each frame of image data acquired, and at the same time, the signal-to-noise ratio of the data is the highest.

[0053] Otherwise, if the acquired TOF sensor is located at the rear or side of the robot vacuum, the TOF sensor is controlled to use automatic exposure mode to acquire point cloud data for SLAM construction. The specific steps are as follows:

[0054] Step B200: Obtain automatic exposure data from the TOF sensor based on the preset automatic exposure mode;

[0055] Specifically, if the TOF sensor is located on the left, right, or rear of the robot vacuum, the data from the TOF sensor is mainly used to build SLAM for the robot vacuum's navigation. Therefore, these TOF sensors are set to automatic exposure mode.

[0056] Furthermore, if multiple Time-of-Flight (TOF) sensors are installed on the robot vacuum, all TOF sensors except the one located at the front are set to automatic exposure mode. To improve the robot vacuum's processor computing power and efficiency, the exposure data of these TOF sensors configured for automatic exposure mode are acquired asynchronously; then, by fusing all the exposure data using a spatial synchronization method, automatic exposure data can be obtained.

[0057] Step B300: Based on the automatic exposure data, obtain second point cloud data reflecting distant space;

[0058] Step B400: Based on the second point cloud data, control the robot vacuum cleaner to move to achieve navigation.

[0059] Specifically, the automatic exposure data is the 3D data of distant objects. Then, phase demodulation and depth calculation are performed on the automatic exposure data to obtain depth data. Next, based on the depth data and the pose data from the TOF sensor, coordinate transformation is performed according to the intrinsic and extrinsic parameter matrix transformation formula to obtain the second point cloud data. The distances between each object in the indoor space and the robot vacuum cleaner are calculated based on the point cloud data, and a map is built to control the robot vacuum cleaner's movement for navigation. For details on phase demodulation and depth calculation, please refer to the relevant descriptions in step A400.

[0060] As described above, by determining whether the TOF sensor is located at the front of the robot vacuum, if it is, a hybrid exposure mode is used to obtain data with different exposure durations to form point cloud data for obstacle avoidance and navigation. Otherwise, an automatic exposure mode is used to obtain point cloud data for navigation. Therefore, only TOF sensor data needs to be processed. By setting different exposure modes for the TOF sensor, obstacle avoidance and navigation can be achieved, reducing development difficulty and improving data processing speed and robot vacuum efficiency.

[0061] Because of the relatively low height of the robotic vacuum cleaner, the TOF sensor preferably uses an asymmetric light source. This means reducing the laser energy near the ground in the vertical direction and increasing the laser energy further away from the ground. This not only helps to reduce the impact of ground scattering signals on the TOF sensor's detection depth but also improves the signal-to-noise ratio for long-range detection, extending the detection distance of the TOF sensor and thus improving the reliability of long-range SLAM.

[0062] Due to the characteristics of the imaging principle of Time-of-Flight (TOF) cameras and the complexity of the measurement environment, the depth information acquired by TOF cameras will be somewhat incomplete, which will ultimately affect the application of TOF cameras. Specifically, TOF cameras may experience overexposure, leading to data anomalies. For example, objects close to the camera, highly reflective objects, or sunlight on the ground are all prone to overexposure by TOF cameras.

[0063] Therefore, in some embodiments, such as Figure 3 As shown, step A200 further includes the following steps:

[0064] Step A210: Judge the long exposure data according to the overexposure identification method;

[0065] Specifically, since overexposure is prone to occur during long exposure times, it is necessary to use overexposure identification methods to judge long exposure data.

[0066] Since the exposure data obtained by the TOF sensor is a frame of three-dimensional image data, the overexposure identification method can adopt conventional image overexposure identification methods in this field, such as converting the image into a grayscale image and dividing the grayscale image into multiple metering units; determining whether the number of pixels with grayscale values ​​between a preset value and 255 in each metering unit is greater than a preset threshold; when it is greater than the preset threshold, determining whether the metering unit is located in an infrared supplementary lighting area and whether there is a moving object in the infrared supplementary lighting area; when the above two conditions are met, it is determined that the metering unit has an overexposure phenomenon.

[0067] Step A220: If the long exposure data is overexposed, correct the long exposure data based on the short exposure data.

[0068] Specifically, when overexposure is detected in long exposure data, the long exposure data is corrected based on short exposure data within the same exposure cycle.

[0069] In this embodiment, a hybrid exposure mode of 50µs and 300µs is used. If the long exposure data is normal, the 300µs phase data and the 50µs phase data are directly fused for calculation. If overexposure is detected in the long exposure data, based on the overexposure judgment result, all positions of overexposed data in the long exposure data are obtained to form a position set; then, for each position in the position set, the data in the short exposure data corresponding to that position is used to replace the data in the long exposure data corresponding to that position. That is, the overexposed pixel data in the 300µs phase data is replaced by the 50µs phase data.

[0070] As described above, by simply replacing data, overexposed data can be corrected, the quality of the phase signal acquired in each frame can be improved, and the fused depth data can be made more accurate.

[0071] While sequentially spaced hybrid exposure modes can maximize the quality of phase data, they also result in high power consumption and low frame rates. When applications require low power consumption and high frame rates, if the exposure data obtained from long exposures is normal, it is unnecessary to acquire exposure data from short exposures. Therefore, in some embodiments, such as... Figure 4 As shown, if the long exposure data is normal, the following specific steps are also included:

[0072] Step A230: Set the long exposure duration as the next exposure cycle and obtain the exposure data;

[0073] Step A240: Determine the exposure data according to the overexposure identification method;

[0074] Step A250: If the exposure data is normal, obtain the first point cloud data based on the exposure data;

[0075] Otherwise, proceed to step A260: Set the short exposure duration and long exposure duration as the next exposure cycle.

[0076] Specifically, in this embodiment, after obtaining data for one exposure cycle using a hybrid exposure mode, if it is determined that there are no overexposed pixels in the 300µs phase data, the next exposure cycle can simply use a long exposure duration of 300µs. If the obtained exposure data is still normal, the first point cloud data can be directly obtained based on this exposure data, thus enabling obstacle avoidance and navigation functions. In other words, the next exposure cycle reduces the acquisition of short exposure data, lowers the data processing load, and improves the working efficiency of the TOF sensor.

[0077] If the exposure data obtained by using a long exposure time of 300us is overexposed, then the mixed exposure mode is still used, and the short exposure time and the long exposure time are set as the next exposure cycle.

[0078] Furthermore, if the exposure data obtained by setting the long exposure duration to the next exposure cycle is still normal, the next exposure cycle can continue to be set to a long exposure duration until the obtained exposure data results in overexposure. In other words, if the exposure data obtained by the long exposure duration is always normal, the long exposure duration will always be used as the next exposure cycle; otherwise, a hybrid exposure mode will be adopted to further reduce the power consumption of the robot vacuum cleaner and improve the processing speed.

[0079] It should be noted that, although Figure 2 The short exposure time is displayed before the long exposure time, but the processing order of long and short exposure data is not limited. Long exposure data can be processed first; if the long exposure data is normal, the short exposure data can be discarded, saving processing time. Furthermore, if the TOF sensor uses two laser emitters, the laser emitter corresponding to the short exposure time can be paused.

[0080] As described above, the system determines whether there are overexposed pixels based on the 300µs phase data of each frame. If so, a hybrid exposure mode is used; otherwise, the next exposure cycle continues with a long exposure of 300µs. This approach minimizes power consumption, increases frame rate, and reduces motion artifacts.

[0081] Exemplary device

[0082] like Figure 5 As shown, corresponding to the above-mentioned obstacle avoidance and navigation method for robotic vacuum cleaners based on TOF sensors, this embodiment of the invention also provides an obstacle avoidance and navigation device for robotic vacuum cleaners based on TOF sensors, the device comprising:

[0083] The orientation acquisition module 600 is used to obtain the orientation of the TOF sensor on the sweeper based on the sweeper's direction of travel;

[0084] The exposure data acquisition module 610 is used to obtain first exposure data of the TOF sensor under a first exposure duration and second exposure data under a second exposure duration based on a preset hybrid exposure mode; or, to obtain automatic exposure data of the TOF sensor based on a preset automatic exposure mode.

[0085] The point cloud data acquisition module 620 is used to fuse the first exposure data and the second exposure data to obtain first point cloud data reflecting near-field space and far-field space; or, based on the automatic exposure data, to obtain second point cloud data reflecting far-field space.

[0086] The control module 630 is used to control the robot vacuum cleaner to move based on the first point cloud data to achieve obstacle avoidance and navigation; or, based on the second point cloud data, to control the robot vacuum cleaner to move to achieve navigation.

[0087] Specifically, in this embodiment, the specific functions of each module of the above-mentioned obstacle avoidance and navigation device for robot vacuums based on TOF sensors can be referred to the corresponding description in the above-mentioned obstacle avoidance and navigation method for robot vacuums based on TOF sensors, and will not be repeated here.

[0088] In one embodiment, a robotic vacuum cleaner is provided, comprising a memory, a processor, and a plurality of Time-of-Flight (TOF) sensors spaced apart on the side of the vacuum cleaner. The memory stores an obstacle avoidance and navigation program for the robotic vacuum cleaner that can run on the processor. When the robotic vacuum cleaner executes the obstacle avoidance and navigation program, it performs the following instructions:

[0089] Based on the direction of travel of the sweeping robot, the position of the TOF sensor on the sweeping robot is obtained;

[0090] When the TOF sensor is located at the front of the robot vacuum cleaner,

[0091] Based on a preset hybrid exposure mode, the first exposure data of the TOF sensor under the first exposure duration and the second exposure data under the second exposure duration are obtained respectively. The first exposure duration is shorter than the second exposure duration, and the first exposure duration and the second exposure duration form an exposure cycle.

[0092] By fusing the first exposure data and the second exposure data, a first point cloud data reflecting near-field and far-field space is obtained;

[0093] Based on the first point cloud data, the robot vacuum cleaner is controlled to move to achieve obstacle avoidance and navigation;

[0094] When the TOF sensor is located at the rear or side of the robot vacuum cleaner,

[0095] Based on a preset automatic exposure mode, the automatic exposure data of the TOF sensor is obtained;

[0096] Based on the automatic exposure data, second point cloud data reflecting distant space is obtained;

[0097] Based on the second point of cloud data, the robot vacuum cleaner is controlled to move in order to achieve navigation.

[0098] Optionally, multiple TOF sensors are set to an automatic exposure mode, and obtaining automatic exposure data from the TOF sensors based on the preset automatic exposure mode includes:

[0099] Exposure data from a TOF sensor configured for automatic exposure mode is acquired asynchronously.

[0100] The automatic exposure data is obtained by fusing all the exposure data according to the spatial synchronization method.

[0101] Optionally, the step of obtaining the first exposure data of the TOF sensor at the first exposure duration and the second exposure data at the second exposure duration based on a preset hybrid exposure mode further includes:

[0102] The second exposure data is judged according to the overexposure identification method;

[0103] If the second exposure data is overexposed, the second exposure data is corrected based on the first exposure data.

[0104] Optionally, correcting the second exposure data based on the first exposure data includes:

[0105] Sequentially obtain each position of the overexposed data in the second exposure data, and replace the data in the second exposure data corresponding to the position with the data in the first exposure data corresponding to the position.

[0106] Optionally, when the second exposure data is normal, it also includes:

[0107] Set the second exposure duration as the next exposure cycle to obtain exposure data;

[0108] The exposure data is judged according to the overexposure identification method;

[0109] When the exposure data is normal, the first point cloud data is obtained based on the exposure data;

[0110] When the exposure data is overexposed, the first exposure duration and the second exposure duration are set as the next exposure cycle.

[0111] Optionally, after obtaining the first point cloud data based on the exposure data, the method further includes:

[0112] The second exposure duration is continuously set as the next exposure cycle until the exposure data is overexposed.

[0113] Optionally, the fusion of the first exposure data and the second exposure data to obtain first point cloud data reflecting near-field and far-field spaces includes:

[0114] Phase demodulation and depth calculation are performed on the first exposure data and the second exposure data respectively to obtain depth data;

[0115] The first point cloud data is obtained by fusing all the depth data using a time and space synchronization method.

[0116] In some embodiments, to reduce costs, a single iTOF 3D sensor can be used on the robot vacuum cleaner, mounted at the front end. Since the iTOF 3D sensor is fixed inside the robot vacuum cleaner, 360° indoor data can be collected using the robot vacuum cleaner's own rotation mechanism, thus meeting SLAM requirements.

[0117] In some embodiments, three iTOF 3D sensors can be installed circumferentially at intervals on the floor, with each iTOF 3D sensor having a horizontal FOV ≥ 120°, one of which is mounted on the robot vacuum facing the direction of travel. By stitching together three iTOF 3D sensors with a horizontal FOV ≥ 120°, real-time 360° spatial point cloud data acquisition is achieved. Even with three iTOF 3D sensors installed, the overall height remains low compared to existing robot vacuums, resulting in a smaller and more compact design.

[0118] Furthermore, to reduce laser energy near the ground, the optical axis of the iTOF 3D sensor can be tilted upwards during installation. The specific tilt angle is determined by the robot vacuum's closest working distance, the field of view (FOV), and the installation height of the sensor center. The maximum upward tilt angle must meet the robot vacuum's closest working distance to avoid "blind spots" at close range, preventing the inability to detect nearby obstacles.

[0119] This invention also provides a computer-readable storage medium storing a robot vacuum cleaner obstacle avoidance and navigation program based on a TOF sensor. When the TOF sensor-based robot vacuum cleaner obstacle avoidance and navigation program is executed by a processor, it implements the steps of any of the TOF sensor-based robot vacuum cleaner obstacle avoidance and navigation methods provided in this invention.

[0120] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0121] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the above device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0122] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0124] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of the above modules or units is merely a logical functional division, and in actual implementation, it can be divided in other ways. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0125] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not mean that the essence of the corresponding technical solutions deviates from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for obstacle avoidance and navigation of a robot vacuum cleaner based on TOF sensors, the robot vacuum cleaner comprising at least two TOF sensors, characterized in that, The method comprises the following steps: obtaining the position of a TOF sensor on the sweeper based on the traveling direction of the sweeper; when the TOF sensor is located at the front end of the sweeper, obtaining first exposure data and second exposure data of the TOF sensor under a first exposure time and a second exposure time respectively based on a preset mixed exposure mode, the first exposure time being shorter than the second exposure time, and the first exposure time and the second exposure time forming an exposure cycle; fusing the first exposure data and the second exposure data to obtain first point cloud data reflecting near and far spaces; controlling the movement of the sweeper based on the first point cloud data to achieve obstacle avoidance and navigation; when the TOF sensor is located at the rear end or side end of the sweeper, obtaining automatic exposure data of the TOF sensor based on a preset automatic exposure mode; obtaining second point cloud data reflecting far space based on the automatic exposure data; controlling the movement of the sweeper based on the second point cloud data to achieve navigation. 2.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner as claimed in claim 1, wherein, The method of setting multiple TOF sensors to an automatic exposure mode and obtaining automatic exposure data of the TOF sensor based on a preset automatic exposure mode comprises the following steps: acquiring exposure data of the TOF sensor configured to the automatic exposure mode in an asynchronous manner; fusing all the exposure data to obtain the automatic exposure data according to a space synchronization method. 3.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner according to claim 1, wherein, The method of obtaining first exposure data and second exposure data of the TOF sensor under a first exposure time and a second exposure time respectively based on a preset mixed exposure mode further comprises the following steps: judging the second exposure data according to an overexposure identification method; when the second exposure data is overexposed, correcting the second exposure data based on the first exposure data. 4.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner according to claim 3, wherein, The method of correcting the second exposure data based on the first exposure data comprises the following steps: sequentially acquiring each position of overexposed data in the second exposure data, and replacing the data corresponding to the position in the second exposure data with the data corresponding to the position in the first exposure data. 5.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner as claimed in claim 3, wherein, When the second exposure data is normal, the method further comprises the following steps: setting the second exposure time as the next exposure cycle to obtain exposure data; judging the exposure data according to an overexposure identification method; when the exposure data is normal, obtaining the first point cloud data based on the exposure data; when the exposure data is overexposed, setting the first exposure time and the second exposure time as the next exposure cycle. 6.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner as claimed in claim 5, wherein, After the method of obtaining the first point cloud data based on the exposure data, the method further comprises the following steps: continuously setting the second exposure time as the next exposure cycle until the exposure data is overexposed. 7.The TOF sensor-based obstacle avoidance and navigation method for a robot vacuum cleaner according to claim 1, wherein, The method of fusing the first exposure data and the second exposure data to obtain first point cloud data reflecting near and far spaces comprises the following steps: performing phase demodulation and depth calculation on the first exposure data and the second exposure data respectively to obtain depth data; fusing all the depth data to obtain the first point cloud data according to a time and space synchronization method.

8. A TOF sensor based obstacle avoidance and navigation device for a robotic vacuum cleaner, characterized in that, The device comprises: a position acquisition module configured to obtain the position of a TOF sensor on the sweeper based on the traveling direction of the sweeper; The exposure data acquisition module is configured to, when the TOF sensor is located at the front end of the robot cleaner, acquire first exposure data of the TOF sensor under a first exposure duration and second exposure data of the TOF sensor under a second exposure duration based on a preset mixed exposure mode; and when the TOF sensor is located at the rear end or side end of the robot cleaner, acquire automatic exposure data of the TOF sensor based on a preset automatic exposure mode. The point cloud data acquisition module is configured to, when the TOF sensor is located at the front end of the robot cleaner, fuse the first exposure data and the second exposure data to acquire first point cloud data reflecting a near distance space and a far distance space; and when the TOF sensor is located at the rear end or side end of the robot cleaner, acquire second point cloud data reflecting the far distance space based on the automatic exposure data. The control module is configured to, when the TOF sensor is located at the front end of the robot cleaner, control the robot cleaner to move based on the first point cloud data to realize obstacle avoidance and navigation; and when the TOF sensor is located at the rear end or side end of the robot cleaner, control the robot cleaner to move based on the second point cloud data to realize navigation.

9. A robot vacuum cleaner characterised in that The robot cleaner comprises a memory, a processor and a plurality of TOF sensors arranged at intervals on the side of the robot cleaner, and the memory stores a robot cleaner obstacle avoidance and navigation program executable on the processor. When the robot cleaner executes the obstacle avoidance and navigation program, the method for obstacle avoidance and navigation based on the TOF sensor according to any one of claims 1 to 7 is realized.

10. A computer readable storage medium, characterized in that, The computer readable storage medium stores a robot cleaner obstacle avoidance and navigation program based on the TOF sensor, and when the processor executes the robot cleaner obstacle avoidance and navigation program based on the TOF sensor, the steps of the method for robot cleaner obstacle avoidance and navigation based on the TOF sensor according to any one of claims 1 to 7 are realized.

Images