Obstacle avoidance sensing system and method for a cleaning robot, and cleaning robot

Abstract

The application provides an obstacle avoidance sensing system and method of a cleaning robot and the cleaning robot, comprising a sensor group module; the sensor group module is arranged on the cleaning robot; the sensor group module is used for sensing various material environments in a bathroom to obtain original sensor data; and the cleaning robot moves according to the original sensor data to avoid obstacles. The application uses as few sensors as possible to achieve effective sensing range coverage in the bathroom, uses as few and general sensors as possible to solve most obstacle avoidance sensing problems, and uses prior knowledge to make the multi-sensor anti-interference ability stronger.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of obstacle avoidance sensing, in particular to an obstacle avoidance sensing system, method and cleaning robot of a cleaning robot. BACKGROUND

[0002] Because the environment of a hotel bathroom is complex, high requirements are put forward for the obstacle avoidance sensing system of a hotel robot.

[0003] The Chinese utility model patent document with the publication number CN205121339U discloses a robot indoor obstacle avoidance sensing device, which comprises a group of infrared sensors and two groups of ultrasonic sensors.

[0004] In view of the related art, the inventors believe that the hotel bathroom scene is complex, and a single sensor cannot guarantee safety, and it is difficult to solve most obstacle avoidance sensing problems. SUMMARY

[0005] In view of the defects in the prior art, the purpose of the present application is to provide an obstacle avoidance sensing system, method and cleaning robot of a cleaning robot.

[0006] According to the present application, an obstacle avoidance sensing system of a cleaning robot is provided, which comprises a sensor group module.

[0007] The sensor group module is arranged on the cleaning robot.

[0008] The sensor group module is used to sense the environment of various materials in the bathroom and obtain raw sensor data.

[0009] The cleaning robot moves according to the raw sensor data.

[0010] Preferably, the system further comprises a priori knowledge module and a credibility processing module.

[0011] The a priori knowledge module and the credibility processing module are arranged on the cleaning robot.

[0012] The a priori knowledge module obtains a priori knowledge and transmits the a priori knowledge to the credibility processing module.

[0013] The credibility processing module processes the a priori knowledge and the raw sensor data, judges the credibility of the raw sensor data, and obtains processed sensor data.

[0014] The cleaning robot moves according to the processed sensor data.

[0015] Preferably, the sensor group module comprises the following modules:

[0016] Laser radar module: sensing the two-dimensional profile of the bathroom environment;

[0017] Depth camera module: sensing the bathroom laser radar blind area;

[0018] Ultrasonic module: sensing the bathroom optical interference material;

[0019] The two-dimensional profile data of the bathroom environment, the bathroom laser radar blind area data and the bathroom optical interference material data constitute the original sensor data.

[0020] Preferably, the prior knowledge module comprises sensor characteristic knowledge and human region annotation knowledge, which is used to combine sensor data characteristics and human annotations on the map.

[0021] Preferably, the ultrasonic module, the laser radar module and the depth camera module are sequentially installed on the cleaning robot from the bottom to the top.

[0022] Preferably, the depth camera module is converted by a detachable rotating setting on the cleaning robot.

[0023] Preferably, when the depth camera module is vertically arranged, the vertical field of view angle is 72° and the horizontal field of view angle is 50°.

[0024] Preferably, the ultrasonic module is arranged around the axis of the cleaning robot.

[0025] According to the obstacle avoidance sensing method of the cleaning robot provided by the application, the obstacle avoidance sensing system of the cleaning robot is applied, and the following steps are included:

[0026] Sensing data acquisition step: the sensor group module senses the environment of various materials in the bathroom to obtain original sensor data;

[0027] Robot moving step: the cleaning robot moves according to the original sensor data.

[0028] According to the cleaning robot provided by the application, the obstacle avoidance sensing system of the cleaning robot is included.

[0029] Compared with the prior art, the application has the following beneficial effects:

[0030] 1. The application uses as few sensors as possible to cover the effective sensing range of the bathroom, and uses as few and general sensors as possible to solve most of the obstacle avoidance sensing problems;

[0031] 2、 The present application has stronger anti-interference ability of multi-sensor through prior knowledge presetting;

[0032] 3、 The multi-sensor layout of the present application can effectively perceive the hotel toilet scene, and guarantee the universality and low cost of the sensor. BRIEF DESCRIPTION OF DRAWINGS

[0033] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, read in conjunction with the accompanying drawings:

[0034] Figure 1 It is a schematic diagram of input and output relationship between modules;

[0035] Figure 2 It is a side view schematic diagram of camera FOV;

[0036] Figure 3 It is a top view schematic diagram of camera FOV;

[0037] Figure 4 It is a top view highlighting the ultrasonic module;

[0038] Figure 5 It is a side view highlighting the ultrasonic module. DETAILED DESCRIPTION

[0039] The present application will be described in detail below with specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any form. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of changes and improvements can be made. These all belong to the protection scope of the present application.

[0040] The embodiment of the present application discloses an obstacle avoidance sensing system of a three-dimensional cleaning robot, which realizes effective perception of the toilet scene with as few and most universal parameters as possible, and further improves the data reliability by combining the fusion of prior knowledge module and reliability processing module. As shown in Figure 1 and Figure 2 The sensor group module, the prior knowledge module and the reliability processing module are arranged on the cleaning robot.

[0041] Sensor group module: used for perceiving the environment of various materials in the toilet to obtain original sensor data. The sensor group module is mainly used for obtaining the original sensor data of the environment.

[0042] The sensor group module includes a laser radar module (model: Starsecond Pavo, Wanji WLR716), a depth camera module (model: Intel realsense), and an ultrasonic module (model: Youcee JSN-SR20-Y1). The ultrasonic module, the laser radar module, and the depth camera module are sequentially installed in the direction from the bottom to the top of the cleaning robot, and all of them can sense the human body to avoid obstacles.

[0043] The laser radar module (2D laser radar module): sensing the two-dimensional profile (horizontal two-dimensional profile) of the bathroom environment, and the installation height on the cleaning robot is about 450 mm.

[0044] As shown in Figure 2 and Figure 3 , the depth camera module (3D depth camera module) changes the field of view angle by a detachable rotation setting. When the depth camera module is vertically arranged, the vertical field of view angle is 72°, and the horizontal field of view angle is 50°. The depth camera module: sensing the laser radar blind area of the bathroom, mainly covering the basin and the toilet.

[0045] Specifically, the 3D depth camera module is used to sense the laser radar blind area height of the bathroom basin and the toilet, and the installation height on the cleaning robot is about 800 mm from the ground, and the position is located in the middle of the robot (in the middle of the robot in the width direction), as shown in Figure 2 , Figure 3 ; the depth camera adopts the FOV parameter of the general camera on the market, which can ensure the replaceability and universality of the sensor. The field of view angle of a general camera is 72° in the horizontal direction and 50° in the vertical direction, which is difficult to effectively sense the basin and the toilet. The vertical installation of the camera can replace the horizontal and vertical field of view angles, and meet the FOV requirement of the bathroom scene in the vertical direction. FOV represents the field of view angle.

[0046] As shown in Figure 4 and Figure 5 , the ultrasonic module is arranged around the axial direction of the cleaning robot. The ultrasonic module: sensing the optical interference material in the bathroom. The two-dimensional profile data of the bathroom environment, the laser radar blind area data of the bathroom, and the optical interference material data of the bathroom constitute the original sensor data.

[0047] Specifically, the ultrasonic module is used to sense the optical interference material, mainly used to sense the glass in the shower room and the debris on the ground. There are 3 ultrasonic modules installed on each face of the robot, a total of 12 ultrasonic modules, respectively located at the two sides and the middle position of each face (in the length and width directions), and the installation height from the ground is 300 mm, as shown in Figure 4 and Figure 5 .

[0048] The prior knowledge module obtains prior knowledge and transmits the prior knowledge to the credibility processing module. The prior knowledge module includes sensor characteristic knowledge and human region labeling knowledge, which are used to combine sensor data characteristics and human labeling on a map. The prior knowledge module is used to accept prior knowledge and improve the reliability of the credibility processing module.

[0049] Specifically, the prior knowledge module mainly includes sensor characteristic knowledge and human region labeling knowledge, which are used to combine sensor data characteristics (such as laser optical intensity, data stability, etc.) and human labeling on a map (such as a bathroom glass strong area labeled on a map during deployment, and the ultrasonic sensor detects this place to take more trusted ultrasonic data).

[0050] The credibility processing module processes the prior knowledge and the original sensor data, judges the credibility of the original sensor data, and obtains processed sensor data.

[0051] Specifically, the credibility processing module is used to process the interference of part of sensor noise. The credibility processing module is mainly used to accept real-time sensor data and load prior knowledge, and judge the credibility of sensor data. Specifically, the method can be realized by designing a filter, a cost function, etc.

[0052] The steps of processing the prior knowledge and the original sensor data include the following: step S0: loading prior knowledge when starting, including human region labeling knowledge and sensor data characteristic knowledge, and initializing the confidence of received sensor data. Step S1: obtaining robot position information from the positioning module, judging whether there is a human labeled area near the robot, if yes, going to step S2, and if no, going to step S3. Step S2: judging which data is in the human labeled area, and updating the confidence of the sensor according to the human labeled area category data. Step S3: accepting the original sensor data of step S2 or step S0, calculating the final sensor confidence through existing algorithms such as filtering fusion, and removing sensor data with low confidence. Step S4: outputting the final sensor data. The credibility of the original sensor data is judged, and the processed sensor data is obtained. The processed data can be output in two forms according to the needs: 1, fused three-dimensional point cloud; 2, data form corresponding to each type of sensor (laser radar 2D point cloud, depth camera 3D point cloud, ultrasonic wave: ranging beam).

[0053] The cleaning robot moves according to the original sensor data. Specifically, the cleaning robot moves according to the processed sensor data.

[0054] The embodiment of the application also discloses a cleaning robot obstacle avoidance sensing method, which is used for Figure 1 and Figure 2As shown, the obstacle avoidance sensing system of the cleaning robot comprises the following steps: a sensing data acquisition step: the sensor group module senses the environment of various materials in the bathroom to obtain original sensor data.

[0055] A priori knowledge transmission step: the priori knowledge module obtains priori knowledge and transmits the priori knowledge to the credibility processing module.

[0056] A credibility processing step: the credibility processing module processes the priori knowledge and the original sensor data to determine the credibility of the original sensor data and obtains processed sensing data.

[0057] A robot moving step: the cleaning robot moves according to the original sensor data. Specifically, the cleaning robot moves according to the processed sensing data.

[0058] The embodiment of the present application also discloses a cleaning robot, as shown in Figure 1 and Figure 2 The obstacle avoidance sensing system of the cleaning robot is shown.

[0059] The present application proposes a multi-sensor layout and a credibility determination method based on sensor characteristics, which can effectively sense the hotel bathroom scene and ensure the universality and low cost of the sensor.

[0060] In the description of the present application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0061] The specific embodiments of the present application are described above. It should be understood that the present application is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essential content of the present application. In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A barrier sensing system for a cleaning robot, characterized in that The sensor group module comprises: The sensor group module is arranged on the cleaning robot. The sensor group module is used for sensing the environment of various materials in the bathroom to obtain original sensor data. The cleaning robot moves according to the original sensor data. The system further comprises a priori knowledge module and a credibility processing module. The priori knowledge module and the credibility processing module are arranged on the cleaning robot. The priori knowledge module obtains priori knowledge and transmits the priori knowledge to the credibility processing module. The credibility processing module processes the priori knowledge and the original sensor data to determine the credibility of the original sensor data and obtains processed sensor data. The cleaning robot moves according to the processed sensor data. The sensor group module comprises the following modules: The laser radar module is used for sensing the two-dimensional profile of the bathroom environment. The depth camera module is used for sensing the laser radar blind area of the bathroom. The ultrasonic module is used for sensing the optical interference material in the bathroom. The two-dimensional profile data of the bathroom environment, the laser radar blind area data of the bathroom and the optical interference material data in the bathroom constitute the original sensor data. The depth camera module is arranged on the cleaning robot and is rotatable to change the field of view angle. When the depth camera module is arranged vertically, the vertical field of view angle is 72° and the horizontal field of view angle is 50°. 2.The obstacle avoidance sensor system of the cleaning robot according to claim 1, characterized in that, The priori knowledge module comprises sensor characteristic knowledge and artificial region annotation knowledge, which are used to combine the sensor data characteristics and the artificial annotation on the map. 3.The obstacle avoidance sensor system of the cleaning robot according to claim 1, wherein, The ultrasonic module, the laser radar module and the depth camera module are sequentially arranged on the cleaning robot from the bottom to the top. 4.The obstacle avoidance sensor system of the cleaning robot according to claim 1, wherein, The ultrasonic module is arranged around the axis of the cleaning robot.

5. A method for obstacle avoidance sensing of a cleaning robot, characterized in that, The obstacle avoidance sensing system of the cleaning robot according to any one of claims 1-4 comprises the following steps: The sensor group module senses the environment of various materials in the bathroom to obtain original sensor data. The cleaning robot moves according to the original sensor data.

6. A cleaning robot characterized by, The obstacle avoidance sensing system of the cleaning robot according to any one of claims 1-4.

Images