Environment map creating method, self-moving device and readable storage medium

Abstract

The embodiment of the invention provides an environment map creating method, a self-moving device and a readable storage medium, the self-moving device is provided with a front sensor and a rear sensor, the detection range of the front sensor is smaller than that of the rear sensor, and when the self-moving device advances in a target environment to create an environment map for the target environment, the self-moving device carries out environment map creating on the target environment. And repeatedly executing the operation of rotating in situ after advancing for a certain distance so as to create an environment map for the target environment by utilizing detection data acquired by the rear sensor. By adopting the scheme, in the process of exploring and mapping by the self-moving equipment, the self-moving equipment moves for a certain distance to execute the rotation operation once, and then moves for a certain distance to execute the rotation operation until an environment map is created for a target environment; in the rotating process, a rear sensor is used for detecting a larger range in front of the self-moving equipment, so that the detection efficiency of the self-moving equipment is improved to a certain extent, and the purpose of improving the efficiency of creating the environment map is achieved.

Description

Technical Field

[0001] Embodiments of the present application relate to the technical field of robots, and particularly to a method for creating an environmental map, a self-mobile device, and a readable storage medium. Background Art

[0002] With the development of Artificial Intelligence (AI) technology, self-mobile devices have gradually entered people's daily lives, bringing great convenience to people. Common self-mobile devices include floor-sweeping robots, air-purifying robots, etc.

[0003] Currently, sensors such as visual sensors or radar sensors are provided at the front end of self-mobile devices. After first entering an unfamiliar environment, during the forward movement of the self-mobile device, the surrounding environment is continuously detected by the sensors to obtain detection data, and then an environmental map is generated based on the detection data. This method for creating an environmental map is applicable to self-mobile devices with sensors provided at the front end and a relatively large detection range of the sensors.

[0004] When both a front sensor and a rear sensor are provided on the self-mobile device and the detection range of the rear sensor is greater than that of the front sensor, if the above-mentioned method for creating an environmental map is continued to be used, the map creation efficiency is low. Summary of the Invention

[0005] Embodiments of the present application provide a method for creating an environmental map, a self-mobile device, and a readable storage medium. During the movement of the self-mobile device, rotation operations are performed at some positions. During the rotation process, the rear sensor is used to detect a larger range in front of the self-mobile device, so as to achieve the purpose of improving the efficiency of creating an environmental map.

[0006] In a first aspect, an embodiment of the present application provides a method for creating an environmental map, which is applied to a self-mobile device. The self-mobile device has a front sensor and a rear sensor, and the detection range of the front sensor is smaller than that of the rear sensor. The method includes:

[0007] During the process of moving in a target environment to create an environmental map of the target environment, the operation of rotating in place after moving a certain distance is performed multiple times, so as to create an environmental map using the detection data obtained by the rear sensor. The detection data is used to expand the known area, and the known area is the area in the target environment that has been detected.

[0008] In a second aspect, an embodiment of the present application provides a self-mobile device, including:

[0009] A fuselage;

[0010] Drive wheels rotatably provided on the fuselage for driving the self-mobile device to move forward;

[0011] A front sensor, disposed at the front end of the fuselage;

[0012] A rear sensor, disposed at the rear end of the fuselage, and the detection range of the rear sensor is larger than that of the front sensor;

[0013] A control device, electrically connected to the drive wheel, the front sensor, and the rear sensor, and is configured to control the self - moving device to perform an operation of rotating in place after traveling a certain distance multiple times during the process of traveling in a target environment to create an environmental map for the target environment, so as to create an environmental map by using the detection data obtained by the rear sensor, where the detection data is used to expand a known area, and the known area is an area in the target environment that has been detected.

[0014] In a third aspect, an embodiment of the present application provides a self - moving device, including a fuselage, a drive wheel, a processor, and a memory. The memory is used to store a computer program, and the processor is configured to execute the computer program stored in the memory to implement the method described in the first aspect or various possible implementation manners of the first aspect above.

[0015] In a fourth aspect, an embodiment of the present application provides a computer - readable storage medium, in which computer instructions are stored, and when the computer instructions are executed by a processor, they are used to implement the method described in the first aspect or various possible implementation manners of the first aspect above.

[0016] The environmental map creation method, self - moving device, and readable storage medium provided by the embodiments of the present application. The self - moving device is equipped with a front sensor and a rear sensor, and the detection range of the front sensor is smaller than that of the rear sensor. When the self - moving device travels in a target environment to create an environmental map for the target environment, it performs an operation of rotating in place after traveling a certain distance multiple times, so as to create an environmental map for the target environment by using the detection data obtained by the rear sensor. By adopting this solution, during the process of the self - moving device exploring and mapping, it performs a rotation operation every time it travels a certain distance, and then performs a rotation operation again after traveling another certain distance... until an environmental map for the target environment is created. By rotating at some positions and using the rear sensor to detect a larger range in front of the self - moving device during the rotation process, the detection efficiency of the self - moving device is improved to a certain extent, achieving the purpose of improving the efficiency of creating an environmental map. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic diagram of the process of a robot creating a map of a long corridor using a traditional environmental map creation method;

[0019] Figure 2 It is a schematic diagram of the process of a robot creating a map of a large room using a traditional environmental map creation method;

[0020] Figure 3 It is a schematic diagram of the path of a robot traveling to a relatively distant target point using a traditional environmental map creation method;

[0021] Figure 4 It is a schematic diagram of a robot encountering a wall corner using a traditional environmental map creation method;

[0022] Figure 5 It is a schematic diagram of a self - moving device for executing the environmental map creation method described in the embodiments of the present application;

[0023] Figure 6 is Figure 5 A schematic diagram of the detection range of the self - moving device shown;

[0024] Figure 7 It is a flowchart of the environmental map creation method provided in the embodiments of the present application;

[0025] Figure 8 It is a schematic diagram of determining information gain in the environmental map creation method provided in the embodiments of the present application;

[0026] Figure 9 It is a schematic diagram of the process of creating a map of a long corridor using the environmental map creation method provided in the embodiments of the present application;

[0027] Figure 10 It is a schematic diagram of the process of creating a map of a large room using the environmental map creation method provided in the embodiments of the present application;

[0028] Figure 11A It is a schematic diagram of a mismatched area in the environmental map creation method provided in the embodiments of the present application;

[0029] Figure 11B is Figure 11A A schematic diagram of the expansion of the known area in;

[0030] Figure 12ASchematic diagram of detecting a far - distance obstacle after a rotation in place operation after traveling a certain distance in the previous execution

[0031] Figure 12B Is Figure 12A Corresponding schematic diagram of performing a rotation in place operation after traveling a certain distance in this execution

[0032] Figure 13 Schematic diagram of the process of determining the target angle in the environmental map creation method provided by the embodiments of the present application

[0033] Figure 14 Schematic diagram of the environmental map creation device provided by the embodiments of the present application

[0034] Figure 15 Schematic diagram of the structure of a self - moving device provided by the embodiments of the present application Detailed implementation manners

[0035] Currently, in the field of household mobile cleaning robots, the sensors on the robot are omnidirectional radars or front - mounted vision sensors. The omnidirectional radar is set on the top of the robot and can rotate 360 degrees. When the robot first enters an unfamiliar working environment or when the working environment changes greatly, it creates an environmental map through the mapping exploration function. During the process of creating the environmental map, the robot uses the sensors to detect the working environment, continuously selects target points with larger gains according to the detection data, travels towards the target points and creates an environmental map based on the Simultaneous Localization And Mapping (SLAM) technology. Subsequently, it plans paths, avoids obstacles, and executes tasks based on the environmental map. Among them, the target point refers to a position point in the working environment that can significantly improve the quality of the environmental map or reduce uncertainty.

[0036] The above - mentioned environmental map creation method is applicable to robots with an omnidirectional radar configuration or a front - mounted vision sensor configuration. During the forward movement of these two types of robots, the detection range of the omnidirectional radar or the front - mounted vision sensor is relatively large, which can ensure that the robot continuously selects target points with larger gains in the forward direction. However, for robots with a front - mounted sensor and a rear - mounted sensor configuration, since the detection range of the front - mounted sensor is smaller than that of the rear - mounted sensor, if the environmental map is created based on the detection data of the front - mounted sensor, the following disadvantages exist:

[0037] A. The mapping efficiency is low. The reasons are as follows:

[0038] a1. The rear - mounted sensor with a larger detection range cannot be fully utilized.

[0039] Traditional environmental map creation methods require the robot to continuously select target points with greater gain and move towards them. Moving towards target points with greater gain will bring greater benefits and higher efficiency to the robot. When the detection range of the front sensor is smaller than that of the rear sensor, the target points with greater gain are often within the detection range of the rear sensor, causing the robot to choose to explore backward rather than forward. However, the robot cannot keep retreating during map building, that is, the robot cannot build a map while retreating. This is because if the robot keeps retreating to explore, from the user's perspective, this behavior is very strange, not in line with common sense and not intelligent enough. Moreover, there is no bumper set at the rear of the robot, and backward detection is very likely to cause collisions, and backward detection is very likely to cause the robot to fall off a cliff. That is to say, long-term backward detection is very unsafe for the robot. Another reason is that the robot moves at a lower speed when retreating, resulting in low detection efficiency.

[0040] Obviously, traditional map building methods require the robot to detect backward, but due to considerations of intelligence and safety, the robot cannot keep retreating, which makes the robot can only use the detection data of the front sensor to build a map during the map building process. However, since the detection range of the front sensor is small, it is impossible to obtain enough detection data at one time for decision-making, so there is a phenomenon that the robot keeps turning back and walking back and forth, resulting in a significant reduction in the map building exploration efficiency. Exemplarily, please refer to Figure 1 and Figure 2 。

[0041] Figure 1 is a schematic diagram of the process of the robot building a map of a long corridor using the traditional environmental map creation method. Please refer to Figure 1 , the rectangle represents the long corridor, the dotted area represents the detection range of the robot's rear sensor, the triangle represents the detection range of the robot's front sensor, and the curve represents the path of the robot during the process of creating the environmental map. Figure 1 In, since the detection range of the robot's front sensor is not sufficient to cover the entire width of the corridor, the robot cannot obtain enough detection data at one time to make an effective navigation decision, so there is a phenomenon of lingering exploration.

[0042] Figure 2 is a schematic diagram of the process of the robot building a map of a large room using the traditional environmental map creation method. Please refer to Figure 2 , the rectangle represents a large room with a large area, that is, a large room, the dotted area represents the detection range of the robot's rear sensor, the triangle represents the detection range of the robot's front sensor, and the curve represents the path of the robot during the process of creating the environmental map. Figure 2In it, due to the small detection range of the robot's front sensor, the robot needs to travel a lot of repetitive distances for exploration. That is to say, when mapping a large room, there will be a phenomenon of constantly wandering and exploring.

[0043] a2. Avoid unknown areas during the mapping process.

[0044] In the working environment, the detected area is the known area, and the undetected area is the unknown area. The robot uses the navigation planning algorithm to perform path planning within the known area, that is, perform path planning based on the known obstacles and safety distances, so as to reduce the collision risk. That is to say, for the known area, the robot can improve safety, efficiency, and reliability by using the navigation planning algorithm to customize emergency measures such as obstacle avoidance strategies in advance. However, for the unknown area, due to more uncertainties, the robot cannot respond to emergencies in a timely manner, and the possibility of collision is very high. When the detection range of the front sensor is very small, if the target point with a large gain is very close in front of the robot, the robot can only take very small steps, resulting in low exploration efficiency. If the target point with a large gain is far from the robot, since the navigation planning only plans paths within the known area, the robot plans a relatively long path, which in turn leads to low exploration efficiency. Exemplarily, please refer to Figure 3 .

[0045] Figure 3 is a schematic diagram of the path of the robot using the traditional environmental map creation method to travel to a farther target point. Please refer to Figure 3 , the black-filled circle represents the target point, and the ideal path is shown by the straight arrow in the figure. However, since the navigation planning only plans paths within the known area, the actual path planned by the robot is shown by the curved arrow in the figure. The actual path is longer than the ideal path, resulting in a significant reduction in exploration efficiency. Moreover, the exploration behavior has a poor visual effect and the exploration route is not smooth.

[0046] a3. At the corners of the wall, the robot needs to use visual perception at close range, resulting in low mapping efficiency.

[0047] For robots with traditional omnidirectional radar or front vision sensor configurations, due to the large forward detection range, they can sense the corners of the wall at a farther distance and do not need to approach the wall to sense, resulting in high mapping efficiency. However, for robots with a small detection range of the front sensor, due to the small detection range of the front sensor, they must approach the wall to sense at close range, resulting in low mapping efficiency. Exemplarily, please refer to Figure 4 .

[0048] Figure 4 is a schematic diagram of the robot using the traditional environmental map creation method encountering the corners of the wall. Please refer to Figure 4Since the detection range of the front sensor is small, the robot must come to the wall so that the wall is within the detection range of the robot's front sensor, so that the robot can detect obstacles such as walls.

[0049] According to a1, a2 and a3 above, it can be seen that for a robot with a small detection range of the front sensor, if the environmental map creation method of the robot with an omnidirectional radar or front visual sensor configuration is used, the mapping efficiency is low because only the detection data of the front sensor can be used.

[0050] B. Missing the map.

[0051] During the exploration process, the robot will draw the detected obstacles on the map in real time. Therefore, when the detection range of the robot's front sensor is small and the detection range of the rear sensor is large, the dynamic obstacles at the door will be written into the map by the rear sensor. During the exploration process, the robot moves forward and the detection range of the front sensor is small. Even if the dynamic obstacle leaves, the robot will not erase the dynamic obstacle, which is prone to missing map construction. For example, the user is blocking the door by looking at the door. The rear sensor detects the dynamic obstacle of the user, and then draws the dynamic obstacle on the map, so that the door is marked as an obstacle and does not need to be explored. When the user walks away, the dynamic obstacle is still retained on the created local map, and the front sensor will not erase the dynamic obstacle previously confirmed by the rear sensor. In other words, even if the front sensor finds that there is no obstacle at the door, the robot still thinks that there is an obstacle at the door, so it will not go in to explore, resulting in missing map construction.

[0052] C. The obstacle point cloud is not fully utilized for decision making, resulting in low mapping efficiency.

[0053] In traditional environmental map creation methods, the robot simply marks the obstacle information detected by sensors such as radar on the map, fails to fully utilize the potential value of the obstacle point cloud, and does not optimize the point selection decisions for exploration planning based on information such as the relative position of the obstacle point cloud, resulting in low mapping efficiency.

[0054] The above-mentioned disadvantage A is caused by the fact that the robot fails to make full use of the detection data of the rear sensor, or even does not use the detection data of the rear sensor. The above-mentioned disadvantage C is that although the robot uses the detection data of the rear sensor, it only simply marks the detected obstacles on the map and does not make decisions based on the obstacle point cloud.

[0055] Based on this, the embodiments of the present application provide an environmental map creation method, a self-mobile device, and a readable storage medium. During the movement of the self-mobile device, rotation operations are performed at some positions to use the rear sensor to detect a larger range in front of the self-mobile device, which improves the detection efficiency of the self-mobile device to a certain extent and achieves the purpose of improving the efficiency of creating an environmental map.

[0056] The self-mobile device described in the embodiments of the present application can be a sweeping robot, a lawn mowing robot, an air purification robot, an automatic spraying device, a nursing robot, a shopping guide robot, a shopping mall service robot, an Automated Guided Vehicle (AGV), etc., which can realize unattended operation of various tasks.

[0057] In the embodiments of the present application, a front sensor is provided at the front end of the self-mobile device, and a rear sensor is provided at the rear end of the self-mobile device. The detection range of the front sensor is smaller than that of the rear sensor. In order to reduce the height of the self-mobile device, the rear sensor usually adopts a sunken design method. In this way, the detection range of the rear sensor is usually much smaller than that of an omnidirectional radar, and the detection range of the omnidirectional radar is 360 degrees.

[0058] Next, taking the self-mobile device as a sweeping robot as an example, the self-mobile device described in the embodiments of the present application will be described in detail. Exemplarily, please refer to Figure 5 and Figure 6 。

[0059] Figure 5 is a schematic diagram of the self-mobile device for executing the environmental map creation method described in the embodiments of the present application, Figure 6 is Figure 5 a schematic diagram of the detection range of the self-mobile device shown. Please refer to Figure 5, in the embodiment of the present application, the self - moving device 500 includes a body 51. A front - end sensor and a rear - end sensor 52 are provided on the body 51. The front - end sensor is, for example, a vision sensor, including but not limited to a monocular camera and a binocular camera. The rear - end sensor 52 is, for example, a lidar, and usually adopts a sunken design method to reduce the thickness of the body 51. The detection data obtained by the rear - end sensor 52 is also called point cloud data, obstacle point cloud, etc. A side brush 53, a rotary brush, a mop, drive wheels, omnidirectional wheels, a dust box, etc. are also provided on the body 51. The side brush 53 is used to clean the ground, and the rotary brush continuously brushes the ground, and can sweep dust and garbage into the dust box. The self - moving device can perform the floor - sweeping task or the mopping task alone, or can perform the floor - sweeping task and the mopping task at the same time. The side brush 53, the rotary brush, and the mop are all provided at the bottom of the body 51, and relative to the forward direction of the body 51, the side brush 53 and the rotary brush are provided in front of the mop. Based on this structure, when the self - moving device 500 performs the floor - sweeping task and the mopping task at the same time, it can achieve the effect of sweeping first and then mopping. In addition, the self - moving device 500 may not be provided with a mop, that is, the self - moving device 500 only has the floor - sweeping function and does not have the mopping function.

[0060] Please refer to Figure 6 , because the rear - end sensor 52 adopts a sunken design, the detection range of the rear - end sensor 52 is not a circular area centered on the self - moving device 500, but a certain area behind the self - moving device 500, such as a 180 - degree range. After the self - moving device exits the base station (station), the detection range of the rear - end sensor 52 is much larger than that of the front - end sensor. For example, the detection radius of the front - end sensor is 0.75 meters, and the detection radius of the rear - end sensor is 3.5 meters.

[0061] It should be noted that the detection radius of the rear - end sensor being 3.5 meters does not mean that the rear - end sensor can only detect a range of 3.5 meters, but refers to an effective range. The detection data for this effective range is used to create an environmental map. In fact, the range that the rear - end sensor can sense is larger than the effective range. For example, in fact, the sensing radius of the rear - end sensor is 5 meters, but only the 3.5 - meter range is taken as the effective range.

[0062] Next, based on Figure 5 and Figure 6 's description, the environmental map creation method described in the embodiment of the present application will be described in detail. Exemplarily, please refer to Figure 7 .

[0063] Figure 7 is a flowchart of the environmental map creation method provided by the embodiment of the present application. The execution subject of this embodiment is the self - moving device. The self - moving device has a front - end sensor and a rear - end sensor, and the detection range of the front - end sensor is smaller than that of the rear - end sensor. This embodiment includes:

[0064] 701. During the process of moving in the target environment to create an environmental map of the target environment, the operation of rotating in place after moving a certain distance is performed for the nth time, where n ≥ 1 and n is an integer.

[0065] When the self - moving device first enters an unfamiliar target environment or the target environment changes significantly, it is necessary to move in the target environment to create an environmental map of the target environment. The target environment can be a house with three bedrooms and two living rooms, a house with two bedrooms and one living room, etc., and the embodiments of the present application do not limit this.

[0066] During the process of the self - moving device moving in the target environment, the operation of rotating in place after moving a certain distance is performed multiple times. During the rotation process, the surrounding environment is detected by the rear - mounted sensor to obtain detection data, and an environmental map is created based on algorithms such as SLAM in combination with the detection data obtained by the rear - mounted sensor. Herein, a certain distance refers to the length of the trajectory of the self - moving device between two adjacent rotation operations. For example, the self - moving device moves from position A to position B, and the map coordinates of position A in the target environment are different from those of position B in the target environment, that is, position A and position B are two different positions in the target environment. Moving a certain distance can be a straight - line advance or a curved advance, and the embodiments of the present application do not limit this. In one implementation, the distances of any two adjacent advances are the same, that is, every time a certain distance is advanced, the self - moving device performs a rotation operation. For example, a rotation operation is performed every 3 meters of travel; in another implementation, the distances of two adjacent advances are different. For example, a rotation operation is performed after traveling 3 meters last time, and a rotation operation is performed after traveling 1.5 meters this time.

[0067] Please refer to Figure 6 , during the process of the self - moving device performing a rotation operation, taking a 360 - degree rotation as an example, the detection range of the rear - mounted sensor of the self - moving device is continuously updated, so as to detect a 360 - degree range centered on the self - moving device. This 360 - degree range includes the front range before the self - moving device rotates, and this front range covers the detection range of the front - mounted sensor. When the self - moving device expands the known area according to the detection range of the rear - mounted sensor, the front range becomes the known area. For example, when the detection radius of the front - mounted sensor is 0.75 meters and the detection range of the rear - mounted sensor is 3.5 meters, after the rotation operation, a 3.5 - meter range in front of the self - moving device is the known area, and within this range, the self - moving device can plan a navigation path to move forward quickly.

[0068] 702. Determine whether the creation of the environmental map is completed. If the creation of the environmental map is completed, end; if the creation of the environmental map is not completed, execute step 703.

[0069] 703. Let n = n + 1 and return to step 701.

[0070] The process of creating an environmental map is essentially a process of continuously expanding the known area. When there are still unexplored unknown areas in the target environment, it means that the environmental map has not been created yet, and it is necessary to continue moving forward and obtain detection data. During the movement process, the self-mobile device continuously performs the operation of moving forward a certain distance and then rotating in place. Each time the rotation operation is performed, the self-mobile device rotates one week or a certain angle to obtain detection data using the rear sensor. The size of the rotation angle can be a preset angle or determined in real time, and the embodiments of the present application do not limit it. The preset angle is, for example, 360 degrees, 180 degrees, etc.

[0071] In the embodiments of the present application, during the process of the self-mobile device moving forward a certain distance each time, a position where a rotation operation needs to be performed is determined according to the current position. The position where the rotation operation needs to be performed is the current position or other positions. After that, the self-mobile device moves to the position where the rotation operation needs to be performed, performs the rotation operation, and then continues to move forward, and re-determines the position where the rotation operation needs to be performed. In this way, from the perspective of the user, during the exploration and mapping process of the self-mobile device, it moves a certain distance and performs a rotation operation once, and then moves another distance and performs the rotation operation again... until the known area completely covers the target environment, that is, until the environmental map is created.

[0072] Please refer to Figure 3 , when the self-mobile device performs the rotation operation, since the unknown area is within the detection range of the rear sensor, it is transformed into a known area. When planning the path, the self-mobile device plans the actual path between the current position and the target point as shown by the straight arrow in the figure. The path shown by the straight arrow is the shortest and the exploration behavior is beautiful compared to the path shown by the curved arrow, achieving the purpose of improving the mapping efficiency.

[0073] Please refer to Figure 4 , when the distance between the self-mobile device and the wall corner is relatively far, the rotation operation is performed so that the detection range of the rear sensor covers the wall corner, and the area where the wall corner is located is transformed into a known area. In this way, the self-mobile device does not need to move forward to the wall, achieving the purpose of improving the mapping efficiency.

[0074] The environmental map creation method provided by the embodiments of the present application is such that the self - moving device is equipped with a front - mounted sensor and a rear - mounted sensor. The detection range of the front - mounted sensor is smaller than that of the rear - mounted sensor. When the self - moving device travels in the target environment to create an environmental map of the target environment, it repeatedly performs the operation of rotating in place after traveling a certain distance, so as to create an environmental map of the target environment using the detection data obtained by the rear - mounted sensor. With this solution, during the process of the self - moving device exploring and mapping, it rotates once after walking a certain distance, and then rotates again after walking another certain distance... until the environmental map of the target environment is created. By rotating at some positions, the rear - mounted sensor is used to detect a larger range in front of the self - moving device, which improves the detection efficiency of the self - moving device to a certain extent and achieves the purpose of improving the efficiency of creating the environmental map.

[0075] In step 701 of the above - mentioned embodiment, the operation of the self - moving device repeatedly performing the operation of rotating in place after traveling a certain distance means that during each process of the self - moving device traveling a certain distance, it determines the position where the rotation operation needs to be performed according to the current position. This position where the rotation operation needs to be performed can be the current position or other positions. Next, a detailed description will be given on how the self - moving device determines the position where the rotation operation needs to be performed according to the current position.

[0076] In one way, the self - moving device determines whether the current position is the position where the rotation operation needs to be performed according to the information gain.

[0077] In this way, during each process of the self - moving device traveling a certain distance, it determines the information gain of the current position. When the information gain is greater than the preset threshold, it performs a rotation operation at the current position to create an environmental map using the detection data obtained by the rear - mounted sensor. Among them, the information gain is used to indicate the amount of information of the unknown area obtained by the self - moving device at the current position, and the unknown area is the area in the target environment that has not been detected.

[0078] In the embodiments of the present application, during the map - building exploration process of the self - moving device, the information gain is detected at a fixed frequency. This information gain refers to the amount of information of the unknown area that the self - moving device can obtain at the current position. The greater the information gain, the greater the amount of information of the unknown area obtained by performing a rotation operation at the current position. The greater the amount of information, the more conducive it is to improving the map - building efficiency and quality. When the information gain of the current position is greater than the preset threshold, it indicates that the area near the current position is an unknown area, an information - rich area, a high - uncertainty area, etc. An information - rich area refers to an area containing a large number of features, such as a corner area, an area where furniture is placed, etc. A high - uncertainty area refers to an area with a relatively high uncertainty. Obtaining the detection data of this area and mapping it can reduce the uncertainty of the environmental map and improve the map - building quality. When the information gain of the current position is greater than the preset threshold, the self - moving device takes the current position as a position where the rotation operation needs to be performed, that is, it performs a rotation operation at the current position.

[0079] The embodiments of the present application do not limit the method for determining the information gain of the current position. For example, a global topological skeleton map is drawn at the current position, and the global topological skeleton map is used to simply illustrate the surrounding environment, hereinafter referred to as the topological skeleton map. The topological skeleton map can help the self-mobile device better understand the environment it is in, which is crucial for long-term exploration and environmental map creation. The self-mobile device determines the information gain of the current position according to the topological skeleton map.

[0080] For another example, the self-mobile device determines the information gain of the current position according to the boundary line. Exemplarily, please refer to Figure 8 , Figure 8 is a schematic diagram for determining the information gain in the environmental map creation method provided by the embodiments of the present application. Please refer to Figure 8 , there is an area in the target environment, and this area is an open area formed by the wall 80. The self-mobile device 81 explores within this area. When located at the current position, this area includes a known area 82, unknown areas 84, 85, and 86. The boundary lines between the known area 82 and the unknown areas include boundary lines 87, 88, and 89. The self-mobile device determines a detection area 83 centered on the current position, which is a square area shown by the dotted line in the figure. This detection area contains boundary lines 87 and 88. The self-mobile device determines a plurality of boundary points from boundary lines 87 and 88, and each boundary point represents a grid in the environmental map. Then, the self-mobile device determines the sum of the number of boundary points on boundary lines 87 and 88, the distance 1 from the self-mobile device to the center of boundary line 87, the distance 2 from the self-mobile device to the center of boundary line 88, and determines the sum of distance 1 and distance 2 to obtain the distance sum. The more the number of boundary points, the larger the unknown area; the smaller the distance sum, the smaller the distance between the self-mobile device and the boundary line, and thus the greater the information gain of the current position; the smaller the number of boundary points, the smaller the unknown area; the larger the distance sum, the larger the distance between the self-mobile device and the boundary line, and thus the smaller the information gain of the current position.

[0081] Next, two actual scenarios are used to elaborate in detail on the environmental map creation method described in the embodiments of the present application. Exemplarily, please refer to Figure 9 and Figure 10 .

[0082] Figure 9 is a schematic diagram of the process of creating a map for a long corridor using the environmental map creation method provided by the embodiments of the present application. Please refer to Figure 9, since the detection range of the rear sensor of the self - moving device is relatively large, the rotation of the self - moving device enables the detection range of the rear sensor to cover the entire width of the corridor. By rotating in place, the self - moving device can expand the known area, thus achieving more effective exploration and navigation in narrow spaces such as long corridors. Therefore, compared with Figure 1 , the self - moving device does not need to turn back for exploration, but rotates in place once after walking a certain distance, that is, moves straight forward to map the entire long corridor, with high mapping efficiency.

[0083] Figure 10 is a schematic diagram of the process of mapping a large room using the environment map creation method provided in the embodiments of the present application. Please refer to Figure 10 , since the detection range of the rear sensor of the self - moving device is relatively large, the rotation of the self - moving device enables the detection range of the rear sensor to cover a larger area and detect a wider area in the room. By performing a rotation operation in place, the self - moving device can collect more environmental information without displacement, reducing the need for physical movement. Based on the more comprehensive environmental information, the robot can more accurately evaluate the surrounding environment, thereby optimizing the path planning and avoiding unnecessary detours and repeated explorations. Therefore, compared with Figure 2 , the self - moving device does not need to frequently turn back to the middle area of the large room and does not need to walk a lot of repeated distances for exploration, but moves straight along the boundary of the large room, with high mapping efficiency.

[0084] It should be noted that although Figure 9 and Figure 10 in, the self - moving device performs a rotation operation every time it moves straight forward a certain distance. However, the embodiments of the present application do not limit this. In other feasible implementation manners, the self - moving device can also move forward in a curve and perform a rotation operation after moving forward a certain distance.

[0085] Adopting this solution, when the information gain at the current position is relatively large, a rotation operation is performed to ensure that the detected data obtained contains more information about the unknown area, and there is no need to frequently perform rotation operations, achieving the purpose of improving the mapping efficiency.

[0086] Optionally, in the above - mentioned embodiments, when the information gain is less than or equal to the preset threshold, the self - moving device continues to move forward from the current position.

[0087] Exemplarily, when the amount of information about the unknown area obtained at the current position is small, it indicates that there are more known areas around the self - moving device, and there is no need to perform a rotation operation. The self - moving device continues to move forward. For example, the self - moving device continues to move forward from the current position; or, the self - moving device turns left or right from the current position, etc. The embodiments of the present application do not limit this.

[0088] With this solution, when the information gain at the current position is relatively small, there is no need to perform a rotation operation but continue to move forward, thus avoiding frequent rotation operations and achieving the purpose of improving the mapping efficiency.

[0089] In another way, during each movement of a certain distance, the mobile device moves from the current position to the reference position and performs a rotation operation at the reference position.

[0090] In the embodiments of the present application, the mobile device continuously expands the known area by using the detection data of the rear sensor to create an environmental map. The mobile device performs a rotation operation to obtain the detection data of the surrounding environment by using the rear sensor. When the rear sensor detects an obstacle, the mobile device draws the obstacle on the environmental map. After the rotation is completed, the mobile device continues to move forward. If the obstacle is drawn on the environmental map once the rear sensor detects it, it is very likely to cause missing mapping. For example, the obstacle may be a dynamic obstacle located at the door, corridor entrance, etc. If the location of the dynamic obstacle is not further explored, it will lead to missing mapping.

[0091] To avoid missing mapping, after each time the mobile device obtains the detection data based on the rotation operation, it uses the front sensor to sense the target environment at the current position or during the movement to obtain a sensing result, and uses this sensing result to verify the detection result of the rear sensor to verify whether there is a dynamic obstacle that has left. When the sensing result is inconsistent with the detection result of the latest detection data, it indicates that the obstacle detected by the rear sensor in the environmental map is a dynamic obstacle and the dynamic obstacle has left the initial position. Then, the mobile device determines the initial position according to the detection data of the rear sensor, that is, the position where the dynamic obstacle is detected.

[0092] Exemplarily, after the mobile device completes the rotation, it determines whether there is a dynamic obstacle that has left the initial position in the target environment according to the sensing result of the front sensor and the detection result of the rear sensor at the current position. When the sensing result of the front sensor is consistent with the detection result of the rear sensor, it is considered that there is no dynamic obstacle or the dynamic obstacle has not left the initial position in the target environment. When the sensing result of the front sensor is inconsistent with the detection result of the rear sensor, it is considered that there is a dynamic obstacle that has left the initial position in the target environment, and further determines the initial position according to the detection data of the rear sensor.

[0093] In the embodiments of the present application, when the sensing result of the front sensor is inconsistent with the detection result of the rear sensor, there is an area in the target environment. The detection result of the rear sensor believes that there is an obstacle in this area, and the sensing result of the front sensor believes that there is no obstacle in this area. This area is also called a mismatched area.

[0094] It should be noted that although the perception result of the front sensor and the detection result of the rear sensor are compared here to determine whether there is a dynamic obstacle leaving the initial position in the target environment. However, the embodiments of the present application are not limited. The self-mobile device can also use other methods to determine whether there is a dynamic obstacle in the target environment and leaving the initial position, and other methods include but are not limited to visual detection methods and the like.

[0095] Adopting this solution, the self-mobile device combines the front sensor and the rear sensor to determine whether there is a dynamic obstacle leaving the initial position, so as to determine whether there is an area where the map is not built, and achieve the purpose of improving the quality of the environmental map.

[0096] In the embodiments of the present application, during each process of traveling a certain distance, when it is determined that there is a dynamic obstacle that has left the initial position, the self-mobile device travels to the reference position and performs a rotation operation at the reference position to obtain detection data using the rear sensor. Among them, the distance between the reference position and the initial position is less than the preset distance.

[0097] Exemplarily, when the self-mobile device determines that there is a dynamic obstacle that has left the initial position, a reference position is determined according to the initial position, and the distance between the reference position and the initial position is less than the preset distance. The preset distance is less than or equal to the effective scanning radius of the rear sensor. For example, if the effective scanning radius of the rear sensor is 3.5 meters, the preset distance is 1 meter, 2 meters, etc. Since the distance between the reference position and the initial position is less than the preset distance, when the self-mobile device performs a rotation operation at the reference position, the initial position is within the effective detection range of the rear sensor, and the entire mismatched area including the initial position is within the effective range of the rear sensor. Therefore, when the self-mobile device performs a rotation operation at the reference position, the rear sensor can re-detect the mismatched area. If it is found that the known area is further expanded after detection, it indicates that there was indeed a dynamic obstacle at the initial position before; if it is found that the known area is not expanded after performing the rotation operation at the reference position, it indicates that a misjudgment has occurred, that is, there is no dynamic obstacle that has left the initial position in the target environment.

[0098] Adopting this solution, during the mapping process, when the self-mobile device finds that a dynamic obstacle in the known area has left, a reference position is determined according to the initial position of the dynamic obstacle, and further detection is performed at this reference position to avoid missing the mapping and achieve the purpose of improving the quality of the environmental map.

[0099] The embodiments of the present application do not limit the timing of the self - moving device performing a rotation operation at the reference position. For example, during the exploration and mapping process of the self - moving device, once it is determined that there is a dynamic obstacle that has left the initial position in the target environment, a reference position is immediately determined based on the initial position, and the device travels to the reference position to perform the rotation operation.

[0100] For another example, during the operation process where the self - moving device repeatedly travels a certain distance and then rotates in place, when it is determined that there is a dynamic obstacle that has left the initial position and after detecting other areas in the target environment, it travels to the reference position. That is to say, during the exploration and mapping process of the self - moving device, it continuously creates an environmental map using the detection data of the rear sensor. Even if it compares the perception result of the front sensor and the detection result of the rear sensor and determines that there is a dynamic obstacle that has left the initial position, and determines a reference position based on the initial position, it will not immediately travel to the reference position but temporarily ignore it, that is, temporarily ignore the mismatched area. When the self - moving device has detected other areas in the target environment, it then travels to the reference position and performs a rotation operation at the reference position. Other areas refer to the areas in the target environment except for the areas that have not been explored due to the existence of dynamic obstacles. Suppose at time t0, a child is at the door of room a, and the self - moving device performs a rotation operation at the current position and marks the child in the currently created local environmental map according to the detection data of the rear sensor. After marking, the child leaves the door of room a at time t1. After the self - moving device finishes rotating, at time t2, it uses the front sensor to sense the surrounding environment at the current position or during the traveling process and obtains a perception result, and finds that the perception result of the front sensor is inconsistent with the detection result obtained based on the detection data of the rear sensor before. The self - moving device believes that a dynamic obstacle has left the door of room a. Therefore, the self - moving device determines the initial position of the child according to the detection data of the rear sensor, that is, determines the initial position according to the detection data at time t0, and determines the reference position according to the initial position. Then, the self - moving device continues to detect other areas. After detecting other areas, the self - moving device travels to the reference position and fully detects the initial position at the reference position.

[0101] It should be noted that although the above uses a child as an example to describe the embodiments of the present application in detail, the embodiments of the present application do not limit the type of the dynamic obstacle. In other embodiments, the dynamic obstacle can also be a door that can be opened and closed, a table or chair that is moved away within a certain period of time, etc.

[0102] For another example, during the operation process where the self - moving device repeatedly moves forward a certain distance and then rotates in place, when it is determined that there is a dynamic obstacle that has left the initial position, during the process of returning to the base station, it moves to the reference position. That is to say, during the exploration and mapping process of the self - moving device, when it is determined that there is a dynamic obstacle that has left the initial position, that is, when a mismatched area is determined, the mismatched area is temporarily ignored, but the initial position is marked, and the detection continues until the traversable area is completely detected. After that, the self - moving device plans a path back to the base station according to the created environmental map. If the distance between the return path and the initial position is relatively small, it moves to the reference position and performs a rotation operation.

[0103] Adopting this solution, after the self - moving device has explored the traversable area in the target environment, it then moves to the reference position or moves to the reference position during the process of returning to the base station, and performs a rotation operation at the reference position, avoiding the self - moving device constantly moving to the reference position, thus achieving the purpose of improving the mapping efficiency.

[0104] Optionally, in the above - mentioned embodiment, after the self - moving device moves to the reference position, it rotates 360 degrees at the reference position to fully detect the surrounding environment of the reference position, avoiding having an obstacle in the environmental map that substantially blocks some areas in the target environment, that is, avoiding missing mapping, and achieving the purpose of improving the quality of the environmental map.

[0105] Optionally, in the above - mentioned embodiment, when the self - moving device moves from the current position to the reference position and performs a rotation operation at the reference position to obtain detection data, it uses the detection data to continue expanding the known area in the environmental map. If the known area is expanded, that is, the area of the known area increases, it moves towards the initial position to continue expanding the known area.

[0106] Exemplarily, after the self - moving device performs a rotation operation at the reference position to obtain detection data, it uses the detection data to expand the known area. If the area of the known area increases, that is, the known area is further expanded, it indicates that there was indeed a dynamic obstacle at the initial position before. At this time, the self - moving device moves towards the initial position so as to enter through the door that was blocked by the dynamic obstacle before, and then detect a new area, such as room a mentioned above. If it is found that the known area has not been expanded after performing the rotation operation at the reference position, it indicates a misjudgment, that is, there is no dynamic obstacle that has left the initial position in the target environment, or the dynamic obstacle has not left. At this time, the self - moving device does not need to move towards the initial position, but determines the target detection point and moves towards the target detection point, and determines the position where rotation is required and rotates during the movement process.

[0107] In the embodiments of the present application, the mismatched area is converted into a known area, causing the known area to be expanded. Exemplarily, please refer to Figure 11A andFigure 11B , Figure 11A is a schematic diagram of a mismatched area in the environmental map creation method provided by an embodiment of the present application, Figure 11B is Figure 11A a schematic diagram of the expansion of a known area in

[0108] Please refer to Figure 11A , there is a mismatched area in the target environment. The reason for the appearance of the mismatched area is as follows: When the self-mobile device creates a map using the detection data of the rear sensor, it is found that there are obstacles in this area, so the obstacles are marked on the environmental map. However, after marking, the self-mobile device discovers according to the perception result of the front sensor that there are no obstacles in this area and assumes that the obstacles have left. At this time, the self-mobile device determines a reference position based on the initial position of the dynamic obstacle that has left and travels to the reference position, such as the position of the self-mobile device in Figure 11A . After that, the self-mobile device performs a rotation operation at the reference position, thereby obtaining the detection data of the mismatched area again and expanding the known area based on the obtained detection data, that is, transforming the mismatched area into a known area. After expansion, it is found that the area of the known area has increased.

[0109] When the self-mobile device performs a rotation operation at the reference position and discovers that the known area has been expanded, it travels towards the initial position to continue expanding the known area. The initial position is located within the mismatched area. Please refer to Figure 11B , the self-mobile device can explore unknown areas by traveling towards the initial position, thereby avoiding missing map building.

[0110] Adopting this solution, after the self-mobile device performs a rotation operation at the reference position, when the known area is further expanded, the self-mobile device further travels towards the initial position, thereby accurately determining the traveling direction and achieving the purpose of improving the map building efficiency and accuracy.

[0111] In another way, the self-mobile device determines a target position based on a long-distance obstacle and performs a rotation operation at the target position.

[0112] In this way, when a long-distance obstacle is detected by performing the operation of traveling a certain distance and then rotating in place last time, the operation of traveling a certain distance and then rotating in place this time is performed; wherein, the distance traveled before the rotation operation in place this time is less than the distance traveled before the rotation operation in place last time, and the long-distance obstacle is an obstacle located outside the detection range of the rear sensor but can be sensed by the rear sensor when the self-mobile device performs the rotation operation.

[0113] In the embodiments of the present application, in the operations of moving forward a certain distance and then rotating in place for two adjacent executions, if a long-distance obstacle was detected last time, the distances of the two forward movements are different: the distance of the last forward movement is long, and the distance of the current forward movement is short. The path of the current rotation in place before moving forward is determined by the maximum detection radius and the effective detection radius of the rear detector. In order to improve the quality of the environmental map, a part of the actual detection range of the rear sensor is used as the effective range, and this effective range is used as the detection range of the rear sensor. In fact, the actual detection range of the rear sensor is larger than the effective range. For example, the maximum detection radius of the rear sensor is 5 meters, but only an effective range of 3.5 meters is taken when creating the environmental map. Obviously, considering the accuracy of the environmental map, the detection data of the rear sensor is not fully utilized.

[0114] In order to make full use of the rear sensor, in the embodiments of the present application, the detection data outside the effective range and within the actual detection range is fully utilized. Exemplarily, please refer to Figure 12A and Figure 12B .

[0115] Figure 12A Schematic diagram of detecting a long-distance obstacle during the last execution of the operation of moving forward a certain distance and then rotating in place. Figure 12B is Figure 12A The corresponding schematic diagram of the current execution of the operation of moving forward a certain distance and then rotating in place. Please refer to Figure 12A . During the last execution of the operation of moving forward a certain distance and then rotating in place, after the mobile device moves a distance 1 and then executes the rotation operation, when it rotates 180 degrees, the actual detection range of the rear sensor is shown by the dotted line in the figure, and the effective range is shown by the dashed line. Obviously, a long-distance obstacle appears outside the effective range and within the actual detection range. After detecting the long-distance obstacle, the mobile device determines the distance 2 of the current forward movement according to the maximum detection radius and the effective scanning radius of the rear sensor.

[0116] Please refer to Figure 12B . After the mobile device finishes the last rotation operation, after moving a distance 2 and then executing the rotation operation, when it rotates 180 degrees, the effective range of the rear sensor covers the long-distance obstacle, that is, the rear sensor can fully detect the long-distance obstacle, so as to obtain the obstacle point cloud of the long-distance obstacle. Since the distance is relatively close, this obstacle point cloud is a detailed point cloud. The mobile device makes decisions according to the distribution characteristics, density of the obstacle point cloud and the relative position of the long-distance obstacle, etc., so as to achieve the purpose of making full use of the obstacle point cloud for decision-making.

[0117] In the embodiments of the present application, if a far - distance obstacle is detected in the previous rotation operation, the self - moving device determines the distance to travel this time, hereinafter referred to as distance 2, and distance 2 is related to the maximum detection radius and the effective scanning radius of the rear sensor. In practice, distance 2 can be flexibly determined. For example, if the maximum detection radius is 5 meters and the effective scanning radius is 3.5 meters, then distance 2 is 5 - 3.5 = 1.5 meters; for another example, if the distance between the self - moving device and the far - distance obstacle is 4.8 meters and the effective scanning radius is 3.5 meters, then distance 2 is 4.8 - 3.5 = 1.3 meters. After the self - moving device travels 1.3 meters, the distance between it and the far - distance obstacle is 3.5 meters. Then, the self - moving device rotates, and the far - distance obstacle is just within the effective range of the rear sensor.

[0118] Adopting this solution, the self - moving device makes full use of the obstacle point cloud outside the effective range of the rear sensor and within the actual detection range, so as to be able to intelligently process complex environmental information and achieve the purpose of improving the quality and creation efficiency of the environmental map.

[0119] Optionally, in the above - mentioned embodiments, when the self - moving device performs a rotation operation, it rotates 360 degrees or a target angle. When rotating 360 degrees, that is, rotating one full circle, the self - moving device uses the rear sensor to fully detect the surrounding environment. Taking the effective scanning radius of the rear sensor as 3.5 meters as an example, during the process of the self - moving device rotating 360 degrees, the rear sensor can detect a circular area with a radius of 3.5 meters centered on the self - moving device.

[0120] When rotating the target angle, at the position where the self - moving device needs to rotate, it determines the target angle according to the boundary line between the known area and the unknown area, the position where it needs to rotate, etc. Exemplarily, please refer to Figure 13 .

[0121] Figure 13 is a schematic diagram of the process of determining the target angle in the environmental map creation method provided by the embodiments of the present application. Please refer to Figure 13 , in the target environment, there is an area, which is an open area formed by the wall 1301. The self - moving device 1302 explores in this area. When located at the current position, this area includes a known area 1303 and an unknown area 1304, and the boundary line 1305 is used to divide the known area 1303 and the unknown area 1304. After the self - moving device 1302 reaches the position where it needs to rotate, it connects the two endpoints of the boundary line 1305 with the position where it needs to rotate respectively to obtain an included angle 1306. For example, the self - moving device takes the included angle 1306 as the target angle and rotates the corresponding angle with the angular bisector of the included angle 1306 as the rotation axis. For example, if the included angle 1306 is 50°, then it rotates 50 degrees.

[0122] For another example, the self - moving device determines the target angle according to the included angle 1306. For example, the sum of the included angle 1306 and 180 degrees is used as the target angle, and the target angle is rotated around the angular bisector of the included angle 1306. For example, if the included angle 1306 is 50 degrees, then the target angle is 230 degrees.

[0123] In addition, the target angle can also be a preset angle, such as 180 degrees, 160 degrees, etc., which is not limited in the embodiments of the present application. The larger the target angle, the larger the detected area and the higher the accuracy of the environmental map; the smaller the target angle, the smaller the detected area and the higher the efficiency of creating the environmental map.

[0124] It should be noted that, although Figure 13 it is described by taking the existence of a boundary line between the known area and the unknown area as an example. However, the embodiments of the present application are not limited thereto. In other feasible implementation manners, there can be two or more boundary lines between the known area and the unknown area. When there is more than one boundary line, the self - moving device selects one boundary line from multiple boundary lines and determines the target angle according to the selected boundary line. The selected boundary line can be, for example, the boundary line with the longest length, the boundary line closest to the self - moving device, etc., which is not limited in the embodiments of the present application; or, the self - moving device determines the target angle according to some of the boundary lines among all the boundary lines. For example, a connection line is obtained by connecting some of the boundary lines in sequence, and the target angle is determined according to the connection line.

[0125] Adopting this solution, when the self - moving device rotates 360 degrees, the quality of the environmental map can be improved, and when rotating the target angle, the efficiency of creating the environmental map can be increased.

[0126] Optionally, in the above - mentioned embodiments, during the process of traveling a certain distance each time, the self - moving device travels towards the target detection point on the boundary line to continue creating the environmental map of the target environment until the expanded known area completely covers the target environment. Wherein, the boundary line is used to divide the target environment into the known area and the unknown area, the boundary line is updated as the known area expands, the target detection point is the point with the largest benefit value on the boundary line, and the benefit value of the target detection point is greater than the benefit value of the previous detection point. The previous detection point is the detection point before the boundary line is updated, and the benefit value of the target detection point is used to indicate the amount of information of the unknown area obtained by the self - moving device from the target detection point.

[0127] Exemplarily, the self - moving device repeatedly performs the operation of moving forward a certain distance and then rotating in place. During the forward movement, the boundary line is continuously updated and the target detection point is determined. Each time the target detection point is determined, the self - moving device moves towards the target detection point. During the movement, when the self - moving device determines a new target detection point again, the previous target detection point becomes the previous detection point. At this time, even if the self - moving device has not reached the previous detection point, it does not need to continue moving towards the previous detection point, but instead moves towards the new target detection point again.

[0128] In the embodiments of the present application, the target detection point is a position point in the target environment that helps to improve the mapping efficiency and quality. The target detection point is also called a high - gain target point and includes but is not limited to the following types:

[0129] a. Points on the boundary of the unknown area

[0130] The boundary of the area not yet detected by the self - moving device. Going to these positions can quickly expand the known area.

[0131] b. Points in the information - rich area

[0132] The information - rich area refers to an area containing a large number of features, which helps to improve the positioning and environmental map accuracy. Information - rich areas such as the corners of walls, areas where furniture is placed, etc.

[0133] c. Points in the high - uncertainty area

[0134] In the environmental map, areas with high uncertainty. Going to these areas can reduce uncertainty and thus improve the quality of the environmental map.

[0135] d. Loop - closing points

[0136] Areas where loop - closing may occur. Going to this area helps to correct errors and improve the consistency of the environmental map.

[0137] e. Sensor data conflict areas

[0138] Places where sensor data is inconsistent. Going to these areas helps to correct the accumulated error and improve the accuracy of the environmental map.

[0139] f. Dynamically changing areas

[0140] Places in the target environment with dynamic changes. Going to this area helps to update the environmental map in a timely manner and ensure the timeliness of the environmental map.

[0141] g. Key areas for path planning

[0142] This area is crucial for path planning intuitively. Going to this area can optimize the path planning effect.

[0143] With this solution, the self - moving device does not move in a straight line during its travel, but moves towards the newly determined target detection point, achieving the purpose of improving the mapping efficiency and quality.

[0144] The following is an embodiment of the device of the present application, which can be used to execute the method embodiment of the present application. For details not disclosed in the device embodiment of the present application, please refer to the method embodiment of the present application.

[0145] Figure 14 It is a schematic diagram of an environmental map creation device provided by an embodiment of the present application. This environmental map creation device 1400 is integrated on a self - moving device. The self - moving device has a front - end sensor and a rear - end sensor. The detection range of the front - end sensor is smaller than that of the rear - end sensor. This environmental map creation device includes: a travel module 1401 and a processing module 1402.

[0146] The travel module 1401 is used to control the self - moving device to perform the operation of rotating in place after traveling a certain distance multiple times during the process of the self - moving device traveling in a target environment to create an environmental map, so as to utilize the detection data obtained by the rear - end sensor;

[0147] The processing module 1402 is used to create an environmental map by using the detection data obtained by the rear - end sensor. The detection data is used to expand the known area, and the known area is the area in the target environment that has been detected.

[0148] In a feasible implementation, the processing module 1402 is used to determine the information gain of the current position during each process of the self - moving device traveling a certain distance. The information gain is used to indicate the amount of information of the unknown area obtained by the self - moving device at the current position. The unknown area is the area in the target environment that has not been detected;

[0149] The travel module 1401 is used to control the self - moving device to perform a rotation operation at the current position when the information gain is greater than a preset threshold, so as to create an environmental map by using the detection data obtained by the rear - end sensor.

[0150] In a feasible implementation, the travel module 1401 is further used to control the self - moving device to continue traveling from the current position when the information gain is less than or equal to the preset threshold.

[0151] In a feasible implementation manner, the traveling module 1401 is configured to, during the process that the self-moving device travels a distance each time, when it is determined that there is a dynamic obstacle that has left the initial position, control the self-moving device to travel to a reference position, where the distance between the reference position and the initial position is less than a preset distance, and the initial position refers to the position where the dynamic obstacle is detected; perform a rotation operation at the reference position to create an environmental map by using the detection data obtained by the rear sensor.

[0152] In a feasible implementation manner, the traveling module 1401 is configured to, when it is determined that there is a dynamic obstacle that has left the initial position and after detecting other areas in the target environment, control the self-moving device to travel to the reference position; or,

[0153] In a feasible implementation manner, the traveling module 1401 is configured to, when it is determined that there is a dynamic obstacle that has left the initial position, control the self-moving device to travel to the reference position during the process that the self-moving device returns to the base station.

[0154] In a feasible implementation manner, the rotation angle of the rotation operation performed at the reference position is 360 degrees.

[0155] In a feasible implementation manner, the traveling module 1401 is further configured to, when the known area is expanded by performing a rotation operation at the reference position, control the self-moving device to travel towards the initial position to continue expanding the known area.

[0156] In a feasible implementation manner, the processing module 1402 is further configured to sense the target environment by using the front sensor at the current position to obtain a sensing result, when the sensing result is inconsistent with the detection result of the most recent detection data, determine that there is a dynamic obstacle in the target environment; determine the initial position according to the detection result of the detection data.

[0157] In a feasible implementation manner, the traveling module 1401 is configured to, when a long-distance obstacle is detected by performing an operation of rotating in place after traveling a distance last time, control the self-moving device to perform an operation of rotating in place after traveling a distance this time; wherein, the distance traveled before the current rotation operation in place is less than the distance traveled before the last rotation operation in place, and the long-distance obstacle is an obstacle located outside the detection range of the rear sensor but can be sensed by the rear sensor when the self-moving device performs the rotation operation.

[0158] In a feasible implementation, the traveling module 1401 is configured to, during each process of the self-mobile device traveling a certain distance, control the self-mobile device to travel towards a target detection point on the boundary line to create an environmental map by using the detection data obtained by the rear sensor. The boundary line is used to divide the target environment into the known area and the unknown area, and the boundary line is updated as the known area expands. The target detection point is the point with the maximum benefit value on the boundary line, and the benefit value of the target detection point is greater than the benefit value of the previous detection point. The previous detection point is the detection point before the boundary line is updated. The benefit value of the target detection point is used to indicate the amount of information of the unknown area obtained by the self-mobile device from the target detection point.

[0159] In a feasible implementation, the rotation operation includes rotating 360 degrees or rotating a target angle. The target angle is an angle determined according to the self-mobile device and the boundary line. The boundary line is used to divide the target environment into the known area and the unknown area, and the boundary line is updated as the known area expands.

[0160] The environmental map creation device provided in the embodiments of the present application can perform the actions of the self-mobile device in the above embodiments. The implementation principle and technical effects are similar and will not be elaborated here.

[0161] Figure 15 It is a schematic structural diagram of a self-mobile device provided in the embodiments of the present application. As Figure 15 shown, the self-mobile device 1500 includes:

[0162] A processor 1501, a memory 1502, a front sensor 1503, and a rear sensor 1504. The detection range of the front sensor 1503 is smaller than the detection range of the rear sensor 1504;

[0163] The memory 1502 stores computer instructions;

[0164] The processor 1501 executes the computer instructions stored in the memory 1502, so that the processor 1501 executes the environmental map creation method implemented by the self-mobile device as above.

[0165] The specific implementation process of the processor 1501 can be referred to in the above method embodiments. The implementation principle and technical effects are similar and will not be elaborated here in this embodiment.

[0166] Optionally, the self-mobile device 1500 further includes a communication component 1505. Among them, the processor 1501, the memory 1502, and the communication component 1505 can be connected through a bus 1506.

[0167] The embodiments of the present application further provide a computer-readable storage medium, in which computer instructions are stored, and when the computer instructions are executed by a processor, they are used to implement the environmental map creation method implemented by the self-moving device as described above.

[0168] The embodiments of the present application further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the environmental map creation method implemented by the self-moving device as described above.

[0169] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0170] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A method for creating an environmental map, characterized in that, Applied to a self - moving device, the self - moving device having a front sensor and a rear sensor, the detection range of the front sensor being smaller than the detection range of the rear sensor, the method comprising: During the process of traveling in a target environment to create an environmental map, performing the operation of traveling a certain distance and then rotating in place multiple times to create an environmental map using the detection data obtained by the rear sensor, the detection data being used to expand the known area, the known area being the area in the target environment that has been detected.

2. The method according to claim 1, wherein The multiple times of performing the operation of traveling a certain distance and then rotating in place to create an environmental map using the detection data obtained by the rear sensor includes: During each process of traveling a certain distance, determining the information gain of the current position, the information gain being used to indicate the amount of information of the unknown area obtained by the self - moving device at the current position, the unknown area being the area in the target environment that has not been detected; When the information gain is greater than a preset threshold, performing a rotation operation at the current position to create an environmental map using the detection data obtained by the rear sensor.

3. The method according to claim 2, characterized in that It further includes: When the information gain is less than or equal to the preset threshold, continuing to travel from the current position.

4. The method according to claim 1, wherein The multiple times of performing the operation of traveling a certain distance and then rotating in place to create an environmental map using the detection data obtained by the rear sensor includes: During each process of traveling a certain distance, when it is determined that there is a dynamic obstacle that has left the initial position, traveling to a reference position, the distance between the reference position and the initial position being less than a preset distance, the initial position referring to the position where the dynamic obstacle was detected; Performing a rotation operation at the reference position to create an environmental map using the detection data obtained by the rear sensor.

5. The method according to claim 4, characterized in that The traveling to the reference position when it is determined that there is a dynamic obstacle that has left the initial position includes: When it is determined that there is a dynamic obstacle that has left the initial position and after detecting other areas in the target environment, traveling to the reference position; or, When it is determined that there is a dynamic obstacle that has left the initial position, during the process of returning to the base station, traveling to the reference position.

6. The method according to claim 4, characterized in that, The rotation angle of the rotation operation performed at the reference position is 360 degrees.

7. The method according to claim 4, wherein It further includes: When performing the rotation operation at the reference position expands the known area, traveling towards the initial position to continue expanding the known area.

8. The method according to claim 4, wherein Before traveling to the reference position when it is determined that there is a dynamic obstacle that has left the initial position during each process of traveling a certain distance, it further includes: Sensing the target environment using the front sensor at the current position to obtain a sensing result; When the sensing result is inconsistent with the detection result of the most recent detection data, determining that there is a dynamic obstacle in the target environment; Determining the initial position according to the detection result of the detection data.

9. The method according to claim 1, wherein The multiple times of performing the operation of traveling a certain distance and then rotating in place to create an environmental map using the detection data obtained by the rear sensor includes: When a long-distance obstacle is detected during the operation of rotating in place after traveling a certain distance in the previous execution, perform the operation of rotating in place after traveling a certain distance this time; wherein, the distance traveled before the rotation operation in place this time is less than the distance traveled before the rotation operation in place in the previous time, and the long-distance obstacle is an obstacle located outside the detection range of the rear sensor but can be sensed by the rear sensor when the self-mobile device performs the rotation operation.

10. The method according to any one of claims 1 to 9, characterized in that, The multiple operations of rotating in place after traveling a certain distance to create an environmental map using the detection data obtained by the rear sensor include: During the process of traveling a certain distance each time, travel towards the target detection point on the boundary line to create an environmental map using the detection data obtained by the rear sensor. The boundary line is used to divide the target environment into the known area and the unknown area, and the boundary line is updated as the known area expands. The target detection point is the point with the maximum benefit value on the boundary line, and the benefit value of the target detection point is greater than the benefit value of the previous detection point. The previous detection point is the detection point before the boundary line is updated. The benefit value of the target detection point is used to indicate the amount of information about the unknown area obtained by the self-mobile device from the target detection point.

11. The method according to any one of claims 1 to 3 and 9, wherein The rotation operation includes rotating 360 degrees or rotating a target angle, and the target angle is an angle determined according to the self-mobile device and the boundary line. The boundary line is used to divide the target environment into the known area and the unknown area, and the boundary line is updated as the known area expands.

12. A self - moving device, characterized in that, including: The fuselage; Drive wheels rotatably provided on the fuselage for driving the self-mobile device to travel; A front sensor provided at the front end of the fuselage; A rear sensor provided at the rear end of the fuselage, and the detection range of the rear sensor is larger than the detection range of the front sensor; A control device electrically connected to the drive wheels, the front sensor, and the rear sensor, and is used to control the self-mobile device to travel in the target environment to create an environmental map. During this process, the control device controls the self-mobile device to perform multiple operations of rotating in place after traveling a certain distance to create an environmental map using the detection data obtained by the rear sensor. The detection data is used to expand the known area, and the known area is the area in the target environment that has been detected.

13. A self - moving device, comprising a fuselage and drive wheels, characterized in that, It further includes: A processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when executed by the processor, implements the method according to any one of claims 1 to 11.

Images