METHOD FOR DETERMINING THE DISTANCE OF A CLEANING ROBOT TO AN OBSTACLE

Description

[0001] For some time now, cleaning robots have employed methods to determine their distance from an obstacle. Conventional cleaning robots equipped for such methods typically include distance measurement systems that determine the robot's distance to the obstacle based on optoelectronic distance measurement—e.g., laser triangulation, time-of-flight measurement, phase difference measurement—the use of a stereo camera, or ultrasonic sensors. Alternatively or additionally, cleaning robots are often equipped with touch sensors that detect a zero distance—i.e., contact—between the robot and an obstacle.

[0002] Such investigative procedures are known, for example, from documents US 10 422 648 B2, WO 2019 / 238958 A1, US 2018 / 249872 A1 and US 6 124 694 A.

[0003] The disadvantages of commonly used non-contact distance measurement systems – e.g., optoelectronic distance measurement, stereo cameras, and / or ultrasonic sensors – include comparatively high implementation costs and inherent inaccuracies in the measurement range of the respective system. A disadvantage of contact-based distance measurement systems is that they require physical contact between the cleaning robot and the obstacle to detect the zero distance. This leads to negative effects such as wear, abrasion, and contamination of either the cleaning robot, the obstacle, or both.

[0004] It is therefore an object of the present invention to show new ways for methods to determine the distance of a cleaning robot to an obstacle - in particular to eliminate the disadvantages mentioned above.

[0005] This problem is solved by the method according to independent claim 1. Preferred embodiments are the subject of dependent claims.

[0006] The basic idea of ​​the invention is to implement a method for determining the distance of a cleaning robot to an obstacle, wherein the cleaning robot comprises two different distance measuring systems, one of which provides reliable measurement results for the actual distance of the cleaning robot to the obstacle in a long-range measuring range and the other in a short-range measuring range. A second of the two distance measuring systems, which is configured for distance measurement in the short-range measuring range, includes a camera directed at the obstacle. Advantageously, the distance of the cleaning robot to the obstacle can thus be precisely determined even in a short-range measuring range using the method according to the invention, whereby a camera already provided on the cleaning robot for mapping the area to be cleaned by the cleaning robot can be used.This results in cost advantages, as additional sensors can be dispensed with by exploiting synergies.

[0007] A method according to the invention for determining the distance of a cleaning robot to an obstacle can preferably be used to prevent collisions between the cleaning robot and the obstacle. The cleaning robot, which is configured to carry out the method, has a first distance measuring system configured for measuring distances in a predetermined long-range measuring area. The cleaning robot also has a second distance measuring system, distinct from the first and configured for measuring distances in a predetermined near-range measuring area. This second distance measuring system of the cleaning robot includes a camera directed towards the obstacle. The near-range measuring area is a distance range close to the obstacle, and the long-range measuring area is a distance range further away from the cleaning robot.The near-field measuring range can therefore extend directly away from the cleaning robot, whereas there is a distance between the end of the far-field measuring range facing the cleaning robot and the robot itself. The far-field and near-field measuring ranges can thus be arranged in a staggered, shell-like arrangement. The near-field and far-field measuring ranges overlap within an overlap zone. In this overlap zone, the distance of the cleaning robot to the obstacle can be reliably measured using both the first and the second distance measuring systems. The method comprises three measures: a), b), and c). Measure a) stipulates that the distance of the cleaning robot to the obstacle is determined at least using the first distance measuring system if the obstacle is located within the far-field measuring range.According to measure b) of the method, the distance of the cleaning robot to the obstacle is determined using the first and second distance measuring systems, provided the obstacle is located in the overlap area. In measure c) of the method, if the obstacle is located in the near-measuring area, the distance of the cleaning robot to the obstacle is determined using at least the second distance measuring system. In measures b) and c), the optical flow of an image sequence captured by the camera of the second distance measuring system is determined and evaluated to determine the distance of the cleaning robot to the obstacle. In determining and evaluating the optical flow of the image sequence, at least one distinctive image point is tracked, which moves along a flow vector with a flow vector length in the image sequence.

[0008] The advantage of this method is that it allows for a reliable determination of the actual distance between the cleaning robot and the obstacle in both long-range and short-range measurement areas. For short-range distance measurement, a camera is used that the cleaning robot can already use to map the area to be cleaned. This means that the method enables the determination of the cleaning robot's distance to the obstacle in short-range measurement areas without significant additional costs.

[0009] According to an advantageous further development of the method, in measures a) and b), the distance of the cleaning robot to the obstacle in the remote measuring range is determined by means of optoelectronic distance measurement using the first distance measuring system. A first distance measuring system specifically based on laser triangulation provides particularly reliable distance measurement results in the remote measuring range.

[0010] According to the invention, in measures b) and c), an optical flow of an image sequence captured by the camera of the second distance measuring system is determined and evaluated to ascertain the distance of the cleaning robot to the obstacle. This offers a particularly precise way to determine the distance of the cleaning robot to the obstacle using the camera.

[0011] In a further advantageous embodiment of the method, step c) provides that the optical flux for determining the distance of the cleaning robot from the obstacle is continuously determined and evaluated. Preferably, the optical flux is continuously determined and evaluated assuming a nearly constant speed of the cleaning robot. This advantageously allows for a particularly reliable determination of the distance between the cleaning robot and the obstacle. Any deviations in the speed of the cleaning robot, determined via the wheel rotation of the cleaning robot's wheels, due to slippage, thus do not affect the distance measurement in the near-field measurement range.

[0012] According to the invention, at least one distinctive image point is tracked when determining and evaluating the optical flow of the image sequence. This distinctive image point moves along a flow vector of a given length within the image sequence. This offers a particularly simple method for determining the optical flow of the image sequence.

[0013] A further preferred embodiment of the method provides that a common end point near the obstacle for the near-measuring range and the overlapping range is defined by means of a calibration distance. This calibration distance is determined using the first distance measuring system in measure a). This advantageously ensures that the distance of the cleaning robot to the obstacle can be accurately determined across the entire distance measuring range – i.e., in the near-measuring range, the far-measuring range, and the overlapping range.

[0014] According to a further advantageous embodiment of the method, if the cleaning robot, upon initiation of measure b), crosses the obstacle-adjacent end of the near-measuring range and the overlap range in the direction of the obstacle, the flux vector length is calibrated to the calibration distance determined by the first distance measuring device in measure a). Advantageously, this ensures with particular reliability that, as soon as the cleaning robot is outside the far-measuring range, precise values ​​of the cleaning robot's distance to the obstacle can be determined by the second distance measuring system in the near-measuring range.

[0015] According to a further preferred embodiment of the method, in measure a) the second distance measuring system is inactive. In measure c), the first distance measuring system is inactive. Preferably, in measure b), both the first and the second distance measuring systems are active. Unnecessary energy consumption by one of the distance measuring systems is thus avoided if the obstacle is located at a distance from the cleaning robot that is outside the measuring range reliably monitored by the respective inactive distance measuring system.

[0016] With a further advantageous development of the method, the remote measuring range corresponds to an interval of 30 mm to 200 mm of the distance between the cleaning robot and the obstacle. Within this interval, the first distance measuring system operates particularly precisely and reliably.

[0017] According to a further advantageous embodiment of the method, the near-range measuring range corresponds to an interval of 0 mm to 50 mm of the distance between the cleaning robot and the obstacle. Within this interval, the distance of the cleaning robot to the obstacle can be determined particularly reliably and precisely using the second distance measuring system.

[0018] Further important features and advantages of the invention will become apparent from the dependent claims, the drawings and the associated description of the figures based on the drawings.

[0019] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or on their own, without leaving the scope of the present invention.

[0020] Preferred embodiments of the invention are shown in the drawings and are explained in more detail in the following description, wherein identical reference numerals refer to identical or similar or functionally identical components.

[0021] They show, each schematically Fig. 1 shows an exemplary cleaning robot configured to carry out a method according to the invention; Fig. 2a shows an exemplary snapshot taken during the execution of the method according to the invention; Fig. 2b shows another exemplary snapshot taken during the execution of the method according to the invention; Fig. 3a shows an exemplary image of a sequence of images taken by means of a camera of the cleaning robot during the execution of the method according to the invention; Fig. 3b shows another image of the sequence of images taken by means of the camera of the cleaning robot during the execution of the method according to the invention; Fig. 3c shows an exemplary determination of an optical flow of the sequence of images from the images of the Figures 3a and 3b In carrying out the method according to the invention, Fig. 4 shows a flow diagram illustrating the method according to the invention by way of example.

[0022] In the Figure 1Figure 2 shows an exemplary perspective view of a cleaning robot 2, which is configured to carry out a method 1 according to the invention for determining a distance A of the cleaning robot 2 to an obstacle 3. The cleaning robot 2 has a first distance measuring system 5 and a second distance measuring system 7 with a camera 8, which is distinct from the first. The first distance measuring system 5 is expediently based on laser triangulation.

[0023] In the Figure 2aA schematic snapshot of the cleaning robot 2 and the obstacle 3 during the execution of the method 1 according to the invention is shown. The first distance measuring system 5 of the cleaning robot 2 is configured for distance measurement in a predetermined remote measuring range 4. The second measuring system 7 of the cleaning robot 2, which differs from the first, is configured for distance measurement in a predetermined near measuring range 6. The near measuring range 6 is a distance range in the vicinity, and the remote measuring range 4 is a distance range at a distance from the cleaning robot 2. The near measuring range 6 can therefore extend directly away from the cleaning robot 2, whereas there is a distance between the end of the remote measuring range 4 facing the cleaning robot 2 and the cleaning robot 2. The remote measuring range 4 and the near measuring range 6 can thus be arranged in a staggered, shell-like arrangement.The camera 8 of the second distance measuring system 7 is directed towards the obstacle 3. It can be seen that the near and far measuring ranges 6, 4 overlap in an overlap area 9. In this overlap area 9, the distance of the cleaning robot 2 to the obstacle 3 can be reliably measured using both the first and the second distance measuring systems 5, 7.

[0024] Figure 4The inventive method 1 is illustrated by means of a flowchart. It can be seen that method 1 comprises three measures a), b), and c). According to measure a) of method 1, the distance A of the cleaning robot 2 to the obstacle 3 is determined by means of at least the first distance measuring system 5 if the obstacle 3 is located within the far measuring range 4. In measure b), the distance A of the cleaning robot 2 to the obstacle 3 is determined by means of the first and second distance measuring systems 5, 7 if the obstacle 3 is located in the overlap area 9. According to measure c) of method 1, the distance A of the cleaning robot 2 to the obstacle 3 is determined by means of at least the second distance measuring system 7 of the cleaning robot 2 if the obstacle 3 is located in the near measuring range 6.

[0025] In the Figure 2aIn the snapshot shown during the execution of procedure 1, obstacle 3 is located in the remote measurement area 4 and outside the overlap area 9. Thus, in Figure 2a The distance A of the cleaning robot 2 to the obstacle 3 is determined using the first distance measuring system 5, since the condition of measure a) of procedure 1 is fulfilled and the conditions of measures b) and c) of procedure 1 are not fulfilled. In the example shown, the distance A of the cleaning robot 2 to the obstacle 3 is determined using the first distance measurement 5 in the remote measuring range 4 by means of laser triangulation. Thus, in both measure a) and measure b), the distance A of the cleaning robot 2 to the obstacle 3 in the remote measuring range 4 is determined using the first distance measuring system 5 by means of laser triangulation.

[0026] The one in Figure 2bThe illustrated snapshot taken during the execution of method 1 shows that the obstacle 3 is now located in the near-measuring area 6. The condition of measure c) of method 1 according to the invention is therefore met. Figure 2b fulfilled. Accordingly, in the example shown, the distance A of the cleaning robot 2 to the obstacle 3 is determined using the second distance measuring system 7. In measures b) and c), an optical flow of an image sequence captured by the camera 8 of the second distance measuring system 7 is determined and evaluated to determine the distance A of the cleaning robot 2 to the obstacle 3.

[0027] The Figures 3a and 3b Each image, in the form of a snapshot, shows the sequence of images captured by the camera 8 of the second distance measuring system 7 in measures b) and c).

[0028] Determining and evaluating the optical flow of the image sequence is in Figure 3cThis is illustrated by example. In measure c), the optical flow is continuously determined and evaluated, for example assuming a nearly constant speed v of the cleaning robot 2, to determine the distance A of the cleaning robot 2 to the obstacle 3. Figure 3c This therefore only shows a snapshot during the determination or evaluation of the image sequence. For this purpose, the images of obstacle 3 in the image sequence are superimposed, so that the image in the Figure 3c The illustrated image is obtained. It is evident that at least one prominent image point P is tracked during the determination and evaluation of the optical flow of the image sequence. In the Figures 3a to 3c Two such distinctive pixels P are marked. This illustrates the Figure 3c, that such a distinctive pixel P in the image sequence shifts along a flux vector F with a flux vector length. From this flux vector length, and given the known speed v of the cleaning robot 2, the distance A of the cleaning robot 2 to the obstacle 3 can be determined.

[0029] In Figure 2a It is also illustrated that a common end 10 near the obstacle of the near-measuring range 6 and the overlapping range 9 is defined by means of a calibration distance 11. The calibration distance 11 is determined using the first distance measuring system 5 in measure a). In a case where the cleaning robot 2, initiating measure b), crosses the obstacle-near end 10 of the near-measuring range 6 and the overlapping range 9 in the direction of the obstacle 3, the flux vector length is calibrated to the calibration distance 11 determined using the first distance measuring device 5 in measure a).

[0030] As illustrated in the examples, in measure a), the second distance measuring system 7 is inactive. In measure c), the first distance measuring system 5 is inactive. In measure b), both the first and the second distance measuring systems 5, 7 can be active. The long-range measuring range 4 corresponds, for example, to a distance A interval between cleaning robot 2 and obstacle 3 from 30 mm to 200 mm. The short-range measuring range 6 corresponds, for example, to a distance A interval of 0 mm to 50 mm between cleaning robot 2 and obstacle 3. Reference symbol list

[0031] 1. Procedure 2. Cleaning robot 3. Obstacle 4. Far measuring range 5. First distance measuring system 6. Near measuring range 7. Second distance measuring system 8. Camera 9. Overlap range 10. Obstacle furthest end 11. Calibration distance ADistance FFlow vector PPixel vVelocity

Claims

1. Method (1) for determining a distance (A) between a cleaning robot (2) and an obstacle (3), wherein the cleaning robot (2) has a first distance measuring system (5) designed to measure the distance in a predetermined remote measuring range (4) and a second distance measuring system (7) with a camera (8) directed onto the obstacle (3) designed to measure the distance in a predetermined close measuring range (6), wherein the close and the remote measuring range (6, 4) overlap in an overlap region (9), in which the distance (A) between the cleaning robot (2) and the obstacle (3) can be reliably measured by means of the first and also by means of the second distance measuring system (5, 7), wherein the method (1) comprises the following measures: a) if the obstacle (3) is located within the remote measuring range (4) - determining the distance (A) between the cleaning robot (2) and the obstacle (3) by means of at least the first distance measuring system (5); b) if the obstacle (3) is located in the overlap region (9) - determining the distance (A) between the cleaning robot (2) and the obstacle (3) by means of the first and the second distance measuring system (5 ,7); c) if the obstacle (3) is located in the close measuring range (6) - determining the distance (A) between the cleaning robot (2) and the obstacle (3) by means of at least the second distance measuring system (7), wherein in measures b) and c) an optical flow of an image sequence detected by means of the camera (8) of the second distance measuring system (7) is determined and evaluated in order to determine the distance (A) between the cleaning robot (2) and the obstacle (3), characterised in that when the optical flow of the image sequence is determined and evaluated, at least one striking pixel (P) is traced, which moves in the image sequence along a flow vector (F) with a flow vector length.

2. Method (1) according to claim 1, characterised in that in measures a) and b) the distance (A) between the cleaning robot (2) and the obstacle (3) in the remote measuring range (4) is determined by means of the first distance measuring system (5) using optoelectronic distance measurement.

3. Method (1) according to claim 1 or 2, characterised in that in measure c), the optical flow is determined and evaluated continuously in particular by assuming an almost constant velocity (v) of the cleaning robot (2) in order to determine the distance (A) between the cleaning robot (2) and the obstacle (3).

4. Method (1) according to one of the preceding claims, characterised in that a shared end (10) of the close measuring range (6) and the overlap region (9) near to the obstacle is defined by means of a calibration distance (11), wherein the calibration distance (11) is determined by means of the first distance measuring system (5) in measure a).

5. Method (1) according to one of the preceding claims, characterised in that in one case, in which the cleaning robot (2) passes over the end (10) of the close measuring range (6) near to the obstacle and of the overlap region (9) in the direction of the obstacle (3), by introducing measure b), a calibration of the flow vector length with the calibration distance (11) determined by means of the first distance measuring device (5) in measure a) is carried out.

6. Method (1) according to one of the preceding claims, characterised in that in measure a), the second distance measuring system (7) and in measure c) the first distance measuring system (5) is inactive, wherein in measure b) both the first and also the second distance measuring system (5, 7) are preferably active.

7. Method (1) according to one of the preceding claims, characterised in that the remote measuring range (4) corresponds to an interval of 30 mm to 200 mm of the distance (A) between the cleaning robot (2) and the obstacle (3).

8. Method (1) according to one of the preceding claims, characterised in that the close measuring range (6) corresponds to an interval of 0 mm to 50 mm of the distance (A) between the cleaning robot (2) and the obstacle (3).