Method for creating an environment map

Abstract

Method for creating an environmental map of an area for the operation of a mobile, self-driving device (10), in particular a floor cleaning device such as a vacuum and / or sweeping and / or mopping robot, comprising the following method steps: - Detecting the surrounding area of ​​the device (10) with at least one first sensor (1) to create a first horizontal plane (5a) of the environment map, - Detecting the surrounding area of ​​the device (10) with at least one second sensor (4) to create a second horizontal plane (5b) of the environment map that differs from the first horizontal plane (5a), and - Planning a movement path of the device (10) based on the first and second levels (5a, 5b) of the environment map, in particular to achieve the greatest possible soil cultivation in the surrounding area.

Description

[0001] The invention relates to a method for creating an environmental map of an environmental area for the operation of a mobile, self-propelled device, in particular a floor cleaning device, such as a vacuum and / or sweeping and / or wiping robot, as well as a mobile, self-propelled device comprising such a method.

[0002] Mobile, self-driving devices, such as robot vacuums, are designed to autonomously clean as entire a floor area as possible. These robot vacuums use sensors to perceive their surroundings, allowing them to move autonomously without colliding with obstacles, determine ideal paths of travel, and clean an optimal, particularly maximum, portion of the floor area to be cleaned. In addition to the bumper familiar to those in the know, which can detect collisions with obstacles at the level of the device's body, robot vacuums are often also equipped with a lidar sensor, which scans the robot's surroundings, usually with a 360° field of view, contactlessly and at a distance of several meters, and records these measurements on an environmental map. The lidar sensor is typically integrated into a tower-like structure on the device's housing, giving the lidar sensor a clear field of view above the robot vacuum's body.

[0003] Such methods for creating environmental maps for vacuum robots are known, for example, from the documents DE 10 2016 124 856 A1, DE 10 2017 220 180 A1 and WO 2019 / 114 221 A1.

[0004] It is possible that an area of ​​the surrounding area remains above and / or below a measurement plane of the lidar sensor that is neither detected by the lidar sensor nor by the bumper, resulting in a risk of the robot vacuum colliding with obstacles, objects, and furniture. The robot vacuum with the lidar tower cannot drive under obstacles that are at the same height as the lidar sensor. However, it is possible for the robot vacuum to clean at least the edge of such obstacles, especially if the lidar tower is located on the robot vacuum at a distance from the cleaning system.

[0005] It is known to additionally equip the lidar sensor with a separate, dedicated bumper. If the robot vacuum hits an obstacle with its lidar bumper, the state-of-the-art robot vacuum moves back a bit and then tries again a few centimeters to see if it can pass. This behavior is repeated until the robot vacuum finds a path that allows it to reach its destination without hitting anything. Conventional robot vacuums do not have a memory for the areas where the lidar bumper has detected an object. Therefore, when the robot vacuums repeat the maneuver along the corresponding path, the lidar sensor will hit the respective obstacle again.

[0006] The object of the invention is to be able to create an environmental map that ensures a logical evaluation of given environmental situations and, in particular, the avoidance of repeated collisions while simultaneously ensuring the cleaning of accessible areas.

[0007] This object is achieved by a method for creating an environmental map of an environmental area for the operation of a mobile, self-propelled device having the features of claim 1. Furthermore, this object is achieved by a mobile, self-propelled device having the features of claim 10. Advantageous embodiments and further developments are the subject of the subclaims.

[0008] According to the invention, in a method for creating an environmental map of an environmental area for the operation of a mobile, self-propelled device, in particular a floor-cleaning device such as a vacuum and / or sweeping and / or mopping robot, the environmental area of ​​the device is detected with at least one first sensor in order to create a first horizontal plane of the environmental map. Furthermore, the environmental area of ​​the device is detected with at least one second sensor in order to create a second horizontal plane of the environmental map, which differs from the first horizontal plane. A movement path of the device is planned based on the first and second planes of the environmental map, in particular in order to achieve the greatest possible soil cultivation in the environmental area.

[0009] In this case, a second level of the environment map is introduced for obstacle detection. This allows for the evaluation of which ground areas are inaccessible to the device's body and therefore cannot be cleaned, and which ground areas are exclusively impassable for the first and / or second sensor. Through appropriate mission and path planning, ground areas near obstacles can then be systematically and, in particular, optimally cleaned.

[0010] For example, the device body is significantly lower than the first and / or second sensor. The first sensor, for example, is a lidar sensor located on the device body in a rear area of ​​the device. With a front area of ​​the device, it can therefore reach and, in particular, clean floor areas where the device would otherwise collide with obstacles if only the lidar sensor were used.

[0011] These at least partially negotiable obstacles are stored in the second level of the environmental map. This advantageously prevents repeated collisions with these obstacles with the lidar sensor. Distinguishing between obstacles at the height of the device body (first horizontal level) and obstacles at the height of the lidar sensor (second horizontal level) advantageously enables repeatable partial negotiation of obstacles, thereby increasing the reachable or cleanable portion of the entire floor area. The targeted evaluation of both levels of the environmental map enables optimized planning of defined driving maneuvers near or along obstacles.

[0012] A mobile, self-propelled device is specifically defined as a floor cleaning device that autonomously cleans floors, particularly in the home. These include, among others, vacuuming, mopping, and / or sweeping robots, such as robot vacuum cleaners and / or robot lawn mowers. These devices preferably operate without, or with as little as possible, user intervention during operation (cleaning or mowing). For example, the device moves automatically to a specified room to clean the floor according to a predefined and programmed process strategy.

[0013] The floor area to be cleaned refers to any room area to be cleaned. This includes, but is not limited to, parts of individual rooms, individual areas of an apartment, individual rooms within an apartment, and / or the entire floor area of ​​the entire apartment or living space.

[0014] In order to take into account any specific environmental characteristics, an exploratory drive with the mobile, self-propelled device is preferably carried out. An exploratory drive is understood to be a reconnaissance drive suitable for exploring the area of ​​land to be worked for obstacles, spatial layout, and the like, and in particular for detecting the area surrounding the device. The goal of an exploratory drive is, in particular, to be able to assess and / or represent the conditions of the soil to be worked.

[0015] After the exploratory drive, the mobile, self-driving device is familiar with its surroundings and can communicate this information to the user in the form of a map, for example, in a mobile app. The map can provide the user with the opportunity to interact with the mobile, self-driving device. The user can advantageously view information in the map and change and / or adapt it as needed.

[0016] An environmental map is understood to mean, in particular, any map suitable for depicting the surroundings of the soil cultivation area with all its obstacles and objects. For example, the environmental map shows a sketch of the soil cultivation area, including the furniture and walls within it. The environmental map has the first and second horizontal planes, which are aligned horizontally to the ground surface and run essentially parallel to each other. Each plane represents its own height-relevant obstacles and objects, so that the planes differ at least in some areas.

[0017] Obstacles are understood to mean any objects and / or items that are located in a soil cultivation area, for example lying or standing there, and that influence the cultivation by the mobile, self-propelled device, in particular hinder and / or interfere with it, such as furniture, walls, curtains, carpets and the like.

[0018] The surrounding map with the obstacles is preferably displayed in the app on a portable device. This primarily serves to visualize a possible interaction for the user.

[0019] In the present case, an additional device is understood to mean in particular any device that is portable for a user, that is arranged outside the mobile, self-propelled device, in particular that is external and / or differentiated from the mobile, self-propelled device, and that is suitable for displaying, providing, transmitting and / or transmitting data, such as a mobile phone, a smartphone, a tablet and / or a computer or laptop.

[0020] An app, in particular a cleaning app, is preferably installed on the portable attachment. This app serves to communicate between the mobile, self-propelled device and the attachment and, in particular, enables visualization of the floor-treatment area, i.e., the living space or apartment or living area to be cleaned. The app preferably shows the user the area to be cleaned as a map of the surrounding area, along with any obstacles.

[0021] A device housing refers specifically to the outer casing of the device, which encloses the device from the outside. The inner workings of the device are located inside the device housing. The first sensor extends beyond the device housing, for example, vertically, i.e., in the Z-direction. For example, the first sensor is located on a top surface in a rear or central area of ​​the device housing.

[0022] A first and a second sensor are understood to mean, in particular, any sensor suitable for reliably detecting obstacles. This is preferably tactile, scanning, touch-sensitive, laser-based, and / or camera-based. The first sensor is preferably a LIDAR sensor and / or a laser tower that scans its surroundings in a horizontal plane through a 360° rotation. Measuring beams, in particular laser beams, are emitted by the first sensor at regular intervals and are used for distance measurement. The rotation of the first sensor occurs about a rotation axis, in particular about a z-axis, relative to the device housing and is performed by a motor. The second sensor is preferably a bumper that is arranged on the LIDAR sensor or on the device body (LIDAR bumper).This preferentially detects obstacles and objects upon contact, so that obstacles at the height level of the lidar sensor can be reliably detected, even if the lidar sensor itself cannot detect the obstacles.

[0023] A horizontal plane is understood to mean, in particular, any plane that runs parallel to the floor surface and in a horizontal and / or horizontal direction. In particular, the horizontal plane runs parallel to the top of the device housing. Preferably, the first horizontal plane is located at a short distance, i.e., just above the top of the device housing. The second horizontal plane preferably extends above the first plane at the highest level of the lidar sensor.

[0024] A movement path is understood to be a planned movement of the device, the goal of which is to achieve the greatest possible soil cultivation in the area surrounding the device. Maximum possible soil cultivation is understood to mean optimal cultivation in the largest possible soil area, including areas below obstacles and objects, if the device's body height (regardless of the lidar sensor height) allows this.

[0025] In an advantageous embodiment, paths beneath obstacles that can be driven under at least in certain areas are integrated into the movement path. In particular, areas beneath obstacles that cannot be driven under with the first sensor, in particular the lidar sensor, due to the height of the lidar sensor, but which are greater than the device housing and can therefore be driven under with the device housing, can thus be cleaned in certain areas. A suction mouth with a brush roller of the device, which is used for cleaning the floor, is advantageously located on a side of the device housing opposite the first sensor, so that optimal cleaning of the floor can be ensured. This allows the suction mouth to reach and clean areas of the floor where the device would otherwise collide with the obstacle with its lidar sensor.

[0026] In a further advantageous embodiment, the first level and the second level are at different heights above the ground. In particular, the first level is closer to the ground than the second level. The first level therefore contains obstacles closer to the ground than the second level. The first level thus contains and displays obstacles that are lower in height than the obstacles on the second level. The levels of the environment map therefore differ in content, particularly due to the different heights of the obstacles entered.

[0027] In a further advantageous embodiment, the first level contains obstacles that are impassable for a device body of the device, and the second level contains obstacles that are impassable for the second sensor. Preferably, the obstacles in the second level are passable for the device body. For example, measurement results from the lidar sensor (first sensor) and / or collision results from a bumper arranged on a front of the device housing are entered in the first level. Collision results from the lidar bumper (second sensor) are entered in the second level. The two separate levels allow a clear distinction to be made as to which obstacles occur at which height level (device housing height or lidar sensor height).

[0028] In a further advantageous embodiment, the movement path is a collision-free path. The movement path is preferably planned such that the device body at least partially passes under the obstacles of the second level. In particular, after the second level has been created, a journey plan is carried out which creates the movement path of the device. For this purpose, each point to be traversed on one of the planned trajectories is checked in both levels of the environmental map for a movement path. In the first level, a check is made to determine whether the device body can pass through or under obstacles, and in the second level, whether the lidar sensor can pass past or under the obstacles. The evaluation is carried out separately, for example, and linked in the program using an "AND" criterion in order to obtain the collision-free movement path.This linked evaluation of both levels now enables the device to automatically drive over, drive under and thus clean the maximum amount of floor space while still avoiding obstacles for the lidar sensor.

[0029] In a further advantageous embodiment, the movement path of the device corresponds to a sawtooth pattern when driving under it. The planned movement path enables the device to clean areas under obstacles, for example furniture, such as the front areas under sofas or cupboards, using a targeted movement. Targeted evaluation of the areas not covered by standard movement patterns, such as "driving along edges" or "cleaning the room in meandering motions", enables, among other things, special driving maneuvers of the device. Areas that are partially driveable under, for example, are not cleaned by driving along sideways, but using a sawtooth pattern in which the device repeatedly drives head-on under the obstacle, then reverses and drives under the obstacle again a little to the side.

[0030] In a further advantageous embodiment, the first sensor is a lidar sensor on the device body and / or a bumper on the device body, and the second sensor is a bumper for the lidar sensor and / or an optical sensor on the lidar sensor. In particular, the second sensor and the resulting creation of the second level are not limited to collision events with the first sensor, i.e., the lidar sensor. Camera systems or other measurement systems at the level of the lidar sensor can also be used alternatively or additionally to fill the second level of the environmental map with data and thereby make this level even more precise. The second sensor can be mounted or attached directly to the lidar sensor or to the device body at the detection height of the second level.

[0031] The invention further relates to a mobile, self-propelled device, in particular a floor cleaning device for the autonomous processing of floor surfaces, such as a vacuum and / or sweeping and / or wiping robot, which comprises a device body, an evaluation unit, a first sensor and a second sensor. The first sensor is arranged on the device body such that it detects a first horizontal plane of a surrounding area. The second sensor is arranged on the first sensor and / or on the device body such that it detects a second horizontal plane of the surrounding area, which plane differs from the first horizontal plane. The evaluation unit creates a map of the surrounding area based on the first horizontal plane and the second horizontal plane, and based thereon creates a movement path of the device with the greatest possible soil processing in the surrounding area.

[0032] Any features, configurations, embodiments and advantages relating to the method also apply in connection with the device according to the invention, and vice versa.

[0033] The invention is explained in more detail with reference to the following embodiments, which are merely examples. They show: Fig. 1A - 1C: each a schematic view of an embodiment of a mobile, self-propelled device suitable for carrying out the method according to the invention, Fig. 2A - 2C: each a schematic view of an embodiment of a method according to the invention for creating an environment map, and Fig. 3A, Fig. 3B: each a schematic view of an embodiment of a mobile, self-propelled device according to the prior art, Fig. 4A, Fig. 4B: each a schematic view of an embodiment of a mobile, self-propelled device suitable for carrying out the method according to the invention, and Fig. 5: a flowchart of an embodiment of a method according to the invention for creating an environment map.

[0034] In the Fig. 1A, Fig. 1B, Fig. 1C shows a mobile, self-propelled device, in particular a vacuum robot 10, which is in the process of cleaning, in particular vacuuming, a floor. In order to be able to perceive its surroundings, to move autonomously without colliding with obstacles, to determine ideal movement paths for cleaning runs, and to clean a maximum portion of the floor area to be cleaned, the vacuum robot 10 has a first sensor, in particular a lidar sensor 1. The lidar sensor 1 is arranged on a device body 2 in a rear central region of the device body 2 and scans a first horizontal plane 3 of the surroundings in a 360° field of view of the vacuum robot, which plane is located just above an upper side of the device body (see Fig. 1B). The measured values ​​detected in this way are entered into an environmental map. In particular, the lidar sensor 1 has a tower-like structure on the device body 2, so that the lidar sensor 1 has a clear field of view above the device body 2.

[0035] In addition to the lidar sensor 2, the robot vacuum cleaner 10 can have a bumper on the device body 2 (not shown). This device bumper detects obstacles and objects located below the measurement plane of the lidar sensor 1. In particular, the device bumper bumps into objects and furniture that were not previously detected by the lidar sensor 1, i.e., particularly very flat obstacles such as shoes, toys, baseboards, and the like.

[0036] Even if the vacuum robot 10 with its lidar sensor 1 cannot completely drive under all obstacles that a vacuum robot without a comparable tower structure can drive under, it is still possible for the vacuum robot 10 to at least clean the edge areas below the obstacles that are too low for the lidar sensor 1, but not for the device body 2. To ensure that the cleaning is targeted, systematic, and, in particular, maximal, the measured values ​​of the individual obstacle sensors are evaluated accordingly.

[0037] In particular, the vacuum robot 10 has a second sensor for this purpose, which is designed as a bumper on the lidar sensor (lidar bumper 4). This lidar bumper 4 detects obstacles in the environment that are arranged above the first level 3 in a second level. The second level is incorporated into the environment map as an additional level. This second level only includes measured values ​​detected by the lidar bumper 4. In combination with the first level, which includes the measured values ​​of the device bumper and / or the lidar sensor, it is advantageously possible to evaluate not only where the vacuum robot 10 cannot plan a movement path because it would collide with an obstacle, but also how the vacuum robot 10 must move in order to clean as systematically as possible and as far as possible under an obstacle that can be partially driven under.

[0038] Fig. 1A shows an isometric view of a vacuum robot 10 having a bumper 4 in the lidar sensor 1, which detects obstacles that are not themselves detected by the lidar sensor 1, but can lead to collisions with the lidar sensor 1. Fig. 1B shows a side view. Fig. Figure 1C shows a top view of the Fig. 1A, Fig. 1B.

[0039] In the Fig. 2A, Fig. 2B and Fig. Figure 2C illustrates the process for filling the first layer 5a and second layer 5b with detected obstacles to create the environment map according to the invention. All measurement results from the lidar sensor and, optionally, collision events from the device bumper, which is arranged on a front of the device housing, are entered into the first layer 5a. Collision events from the lidar bumper are entered into the second layer 5b.

[0040] During an exploration drive, the first level 5a is initially filled, as the lidar sensor continuously scans the environment. When an obstacle comes into contact with the lidar bumper, obstacles are entered into the second level 5b. At the beginning of the exploration drive, the second level 5b is only sparsely filled (see Fig. 2A, Fig. 2B). As the robot systematically travels across the floor during cleaning runs, it increasingly comes into contact with obstacles 6 whose underride height leads to a collision with the lidar sensor 1. With the increasing number of cleaning runs and thus with each additional collision with an obstacle 6, the second level 5b is expanded and thus becomes more accurate to the environment. The two levels of the environment map allow distinguishing which obstacles occur at which height level (robot height or lidar height).

[0041] The Fig. 2A to 2C show the progressive exploration of the environment and the concomitant progressive filling of the second level 5b. In Fig. Figure 2B shows the vacuum robot 10, the obstacles 6 and the superimposed layers 5a, 5b of the current environment map. Fig. Figure 2C shows the final environment map after multiple detections. The robot vacuum cleaner can clearly distinguish which obstacles 6 affect the device body and which obstacles 6 affect the lidar sensor.

[0042] An optimal movement path can be implemented after introducing the second level 5b. For this purpose, for each path, each point to be traversed on the planned trajectory is checked in both levels 5a and 5b. In the first level 5a, it is checked whether the robot vacuum cleaner can pass all obstacles 6. In the second level 5b, it is checked whether the lidar sensor can get past the obstacles. For example, separate evaluations are used for optimal movement path planning, which are linked in the program with an "AND" criterion to obtain a collision-free path. This joint evaluation of both levels 5a and 5b advantageously enables the robot vacuum cleaner to automatically cover and, in particular, clean the maximum amount of floor space, while still being able to avoid obstacles 6 for the lidar sensor.

[0043] As a result, the robot vacuum cleaner 10 is able to clean specific areas under obstacles such as furniture, such as the front area under sofas or cupboards, by means of a targeted movement.

[0044] Fig. 3A, Fig. 3B show a standard movement pattern F, in particular driving along the edges of the obstacles 6. Here, the vacuum robot 10 only just passes under the obstacle 6 in order to clean along the obstacle 6. Maximum cleaning, especially in areas below the obstacle 6, is thus impossible.

[0045] Fig. 4A, Fig. 4B, on the other hand, depicts a targeted movement of the vacuum robot based on the two levels of the environment map. The targeted analysis enables special maneuvers in which affected floor areas are not only cleaned by moving sideways, as in the Fig. 3A, Fig. 3B. The special driving maneuver includes, for example, a sawtooth pattern S, in which the vacuum robot repeatedly drives head-on under the obstacle 6, then reverses and drives slightly offset to the side again under the obstacle 6, as shown in the Fig. 4A, Fig. 4B. This allows the vacuum robot 10 to push its suction mouth as far back as possible under the obstacles 6, thus ensuring maximum floor cleaning even under obstacles 6.

[0046] In Fig.Figure 5 shows a flowchart for creating an environment map based on two levels and movement path planning. In step 100, the robot vacuum performs an exploration or cleaning run. The robot vacuum detects an obstacle with its lidar sensor or device bumper (step 101a) and enters this detected obstacle into the first level of the environment map (step 102a). If the robot vacuum detects an obstacle with the lidar bumper (step 101b), the robot vacuum enters this detected obstacle into the second level of the environment map (step 102b). The robot vacuum then plans its optimal or maximum movement path based on the created first and second levels (step 103). In particular, the robot vacuum checks this planned movement path on the first level of the environment map (step 104a). Furthermore, the robot vacuum checks the planned movement path on the second level of the environment map (step 104b).Finally, the robot vacuum follows the planned path to complete its cleaning (step 105).

Claims

[1] Method for creating an environmental map of an environmental area for the operation of a mobile, self-propelled device (10), in particular a floor cleaning device such as a vacuum and / or sweeping and / or wiping robot, comprising the method steps: - detecting the surrounding area of ​​the device (10) with at least one first sensor (1) to create a first horizontal plane (5a) of the environment map, - detecting the surrounding area of ​​the device (10) with at least one second sensor (4) to create a second horizontal plane (5b) of the environment map, which differs from the first horizontal plane (5a), and - Planning a movement path of the device (10) based on the first and second levels (5a, 5b) of the environmental map, in particular to achieve the greatest possible soil cultivation in the surrounding area. [2] Method according to claim 1, wherein paths below obstacles (6) which can be driven under at least in some areas are integrated into the movement path. [3] Method according to one of the preceding claims, wherein the first level (5a) and the second level (5b) are different height levels above the ground. [4] Method according to one of the preceding claims, wherein the movement path is a collision-free path. [5] Method according to one of the preceding claims, wherein the first level (5a) contains obstacles (6) which are not passable for a device body (2) of the device (10), and the second level (5b) contains obstacles (6) which are not passable for the second sensor (4). [6] Method according to claim 5, wherein the obstacles (6) of the second level (5b) are passable for the device body (2). [7] Method according to claim 6, wherein the movement path is planned such that the obstacles (6) of the second level (5b) are at least partially passed under by the device body (2). [8] Method according to claim 7, wherein during the underrun the movement path of the device (10) corresponds to a sawtooth pattern (S). [9] Method according to one of the preceding claims, wherein the first sensor (1) is a lidar sensor on the device body (2) and / or a bumper on the device body (2), and the second sensor (4) is a bumper on the lidar sensor and / or an optical sensor on the lidar sensor. [10] Mobile, self-propelled device (10), in particular a floor cleaning device for autonomously processing floor surfaces, such as a vacuuming and / or sweeping and / or wiping robot, comprising a device body (2), an evaluation unit, a first sensor (1) and a second sensor (4), wherein - the first sensor (1) is arranged on the device body (2) in such a way that it detects a first horizontal plane (5a) of an ambient area, - the second sensor (4) is arranged on the first sensor (1) and / or on the device body (2) in such a way that it detects a second horizontal plane (5b) of the surrounding area which differs from the first horizontal plane (5a), and - the evaluation unit creates an environmental map based on the first horizontal plane (5a) and the second horizontal plane (5b), and based thereon creates a movement path of the device (10) with the greatest possible soil cultivation in the surrounding area.

Images