JP4163150B2 - Self-propelled vacuum cleaner - Google Patents

Description

  The present invention relates to a self-propelled cleaner that performs cleaning while self-propelled on a floor surface of a space to be cleaned.

  There are regular and random driving modes for self-propelled cleaners. Regular travel is a travel mode that travels uniformly on the floor of the space to be cleaned (the room to be cleaned) according to a predetermined travel route, and is regular only for the travel time or travel distance determined according to the shape and dimensions of the room. Cleaning is completed by running. On the other hand, in the case of random running, when a wall surface or an obstacle in the room to be cleaned comes into contact with or is about to come in contact, the traveling direction is repeatedly changed so as to reflect in a random angular direction from the room floor surface. Since this is a traveling mode and there is no standard for completion of cleaning as in regular traveling, the cleaning is completed by traveling for a travel time or travel distance set as appropriate.

  In order to perform the regular traveling as described above, it is necessary to set a traveling manner until the completion of cleaning according to the shape and dimension of the room. For this, it is possible to obtain a three-dimensional shape data of a room in advance and generate a driving style based on the data. However, it takes time and effort to create the three-dimensional shape data of the room, and there are problems such as a large amount of data, which is not practical. On the other hand, based on the wall surface of the room, the shape of the room is obtained by first making a round of the self-propelled vacuum cleaner along the wall surface of the room, and then the shape of the room obtained by traveling around the wall A method (for example, Patent Document 1) for performing regular traveling based on the dimensions is more realistic.

Japanese Patent Laid-Open No. 5-46239

  From the viewpoint of the efficiency of cleaning between regular running and random running, it can be said that regular running is better because random running is more likely to have duplicate cleaning parts. However, there is one problem with regular driving. That is, in the case of regular running, as described above, first, the self-propelled vacuum cleaner makes a round along the wall surface of the room to be cleaned, and the shape and dimensions of the room are acquired. Realistic. However, in this case, factors such as the accumulated error of sensors provided in the self-propelled vacuum cleaner to detect the wall surface of the room, etc., or the door that partitions the room is open and the self-propelled vacuum cleaner is outside the room. If you run away, you may not be able to complete the round trip along the wall. Further, if the internal structure of the room to be cleaned is complicated, it may take a long time to complete one round along the wall surface. In such a situation, in reality, regular travel after traveling around the wall cannot be performed, and cleaning must be interrupted.

  Therefore, it is desired that cleaning can be performed as efficiently as possible by appropriately using regular driving and random driving depending on the situation of the room to be cleaned. In addition, there is a problem of the cleaning efficiency as described above for the random running, and it is desired to improve the cleaning efficiency by generating as few overlapping cleaning portions as possible. In addition to the problem of cleaning efficiency, there is also a problem that it is easy to generate uncleaned parts in random running. That is, for example, when there is a recess in an obstacle such as furniture installed in the room to be cleaned, it is reflected in a random angle direction at the corner of the recess and cannot enter the interior of the recess. There is a possibility that the inside may be left uncleaned, and it is desired to effectively prevent such uncleaned residue.

  The present invention has been made in the background of the circumstances as described above, and it is an object of the present invention to provide a self-propelled cleaner that enables traveling control to properly use regular traveling and random traveling according to the condition of a room to be cleaned. In addition, an object of the present invention is to provide a self-propelled cleaner that can effectively prevent unclean leftovers in random running.

In order to achieve the above object, in the present invention, in a self-propelled cleaner that performs cleaning while traveling on a floor surface of a space to be cleaned by a traveling mechanism, a regular traveling mode that allows traveling according to a predetermined traveling route is used. A regular travel generation means for generating, a random travel generation means for generating a random travel mode for performing a travel in which the travel direction is changed at random, and an output from each of the regular travel generation means and the random travel generation means is selected by the travel mechanism. A first determination method for determining success or failure of one-round traveling along the wall based on an elapsed time from the travel start time , and a reference rotation angle in which the rotation direction angle from the travel start time is set in advance The success / failure of the round trip along the wall by combining the judgment made by comparing with the judgment made by comparing the travel distance from the start of travel with the preset reference travel distance A second determination method for determining, a determination based on the distance between the current position during travel and the travel start position, and a rotation direction angle from the travel start time larger than the reference rotation angle α is set in advance. Any of the third determination methods for determining the success or failure of the round trip along the wall by combining the determination made by comparing with β and the determination made by comparing the travel distance from the start of travel with a preset reference travel distance One or both of these determination methods are combined, and a round trip along the wall is determined. The round trip along the wall is first tried, and the success or failure is judged by the round trip judgment along the wall. If the round trip is successful, the switch outputs the output from the regular trip generation means to the running mechanism to run in the regular trip mode, and the round trip along the wall is unsuccessful. If there is characterized in that the are output from the random travel generating means outputs to the driving mechanism so as to perform a travel in the random movement pattern by the switch.

Further, in the present invention, for the automatic cleaner as described above, a traveling mode is generated for an obstacle that causes a random traveling direction change in the random traveling so as to follow the peripheral surface of the obstacle. And an image pickup means for picking up an image of the obstacle detected during the drive with the obstacle as a center, and an image picked up by the image pickup means together with a position / direction angle of the self-propelled cleaner. An image captured by the imaging unit and an image stored in the storage unit based on a position / direction angle of the self-propelled cleaner during travel of the self-propelled cleaner. Is compared with the surrounding image obtained by pattern matching to determine whether the obstacle has passed or not passed by a determination method for determining whether the obstacle has passed or not passed. Comprises a pre-pass / not pass judgment means, the copying run by an output from the copying traveling generating means to the driving mechanism for the pre-pass / not pass is determined as non-passed by the determination means obstacle It is designed to run in a style.

  In the present invention, first, a round trip along the wall is tried, and whether or not the room to be cleaned is suitable for regular running is determined by determining the success or failure by a predetermined determination method. Random driving is selected for cleaning. For this reason, according to the present invention, it is possible to perform traveling control that properly uses regular traveling and random traveling according to the state of the room to be cleaned, and it is possible to perform cleaning more efficiently.

  Also, in the present invention, it is determined whether or not the obstacle has been passed / not passed during random running, and only the obstacle that has been judged as not passing can be surely entered into the recess of the obstacle and cleaned. I am trying to do. For this reason, according to the present invention, it is possible to effectively prevent uncleaned leftovers in random running, and to effectively avoid the overlap of cleaning and increase the efficiency of cleaning.

  Hereinafter, preferred embodiments for carrying out the present invention will be described. FIG. 1 is a block diagram showing the configuration of a travel control system in the self-propelled cleaner according to the first embodiment. The self-propelled cleaner includes mechanical elements such as a suction cleaning mechanism system for cleaning by suction and a traveling mechanism system for self-propulsion, and a known configuration can be used for them. Therefore, description thereof is omitted below.

  The travel control system includes a position / direction angle measuring means 1, a current position / direction angle calculating means 2, a time measuring means 3, a wall-around round determining means 6, a travel route storage means 7, a regular travel generating means 8, and a random travel generating means. 9 and the switch 10 are provided.

  The position / direction angle measuring means 1 is a combination of a position measuring means using an encoder attached to a traveling motor of a traveling mechanism system and a direction angle measuring means using a gyro sensor attached to a self-propelled cleaner, for example. According to the traveling of the self-propelled cleaner, a detection signal of its position and traveling direction angle is generated. The current position / direction angle calculation means 2 receives the detection data of the position and direction angle of the self-propelled cleaner detected by the position / direction angle measurement means 1, and calculates the current position and direction angle of the self-propelled cleaner. . The time measuring means 3 cumulatively measures the traveling time from the time when the self-propelled cleaner starts traveling. The round-around-around determination means 6 receives outputs from the current position / direction angle calculation means 2, the time measurement means 3 and the travel route storage means 7, and performs self-propelled cleaning along the wall surface of the room to be cleaned as described later. The success or failure of the round trip along the aircraft wall is judged. The travel route storage means 7 obtains the time series value of the position / direction angle of the self-propelled cleaner from the output of the time measurement means 3 and the current position / direction angle output from the current position / direction angle calculation means 2. Remember. The regular travel generation means 8 generates a servo command value for causing the self-propelled cleaner to travel in the regular travel mode. The random travel generation means 9 generates a servo command value for causing the self-propelled cleaner to travel in a random travel mode. The switch 10 switches the servo command value output to the servo drive system 11. That is, the servo command value for the regular travel generated by the regular travel generation means 8 or the servo command value for the random travel generated by the random travel generation means 9 is switched according to the determination result by the one-round determination means 6 along the wall to the servo drive system 11. The traveling mechanism system 12 causes the self-propelled cleaner to perform regular traveling or random traveling under the control of the servo drive system 11 that receives the output.

  FIG. 2 shows an example of traveling patterns for regular traveling and random traveling in a rectangular room. FIG. 2a shows an example of regular running. In this example, regular running is a run that regularly repeats zigzag wrapping by a combination of straight running in the direction parallel to the wall of the room and 90 ° direction change. The vehicle travels along a travel route predetermined by the pattern. In the case of such regular driving, the self-propelled vacuum cleaner overlaps the starting point of the cleaning start or close to the starting point by performing regular driving for the driving time or driving distance determined according to the shape and dimension of the room. It will be in the state which returns to the end point of the position of, and cleaning is completed. On the other hand, b in FIG. 2 is an example of random running. In random running, when you touch or are about to touch a wall or obstacle in the room to be cleaned, you run on the floor of the room repeatedly changing the direction of travel so that it reflects in a random angular direction. It will be. In such random travel, the position where the travel is completed for the travel time or travel distance set appropriately is the end point and the cleaning is completed.

  Next, the success / failure determination of the round trip along the wall by the round trip judgment means 6 along the wall will be described. FIG. 3 shows the flow of processing in the success / failure determination of the one-round travel along the wall. The one-round traveling along the wall is a traveling mode in which the room to be cleaned travels around the wall along the wall. However, when there is furniture or the like that is fixedly installed along the wall of the room, the outer peripheral surface of the furniture or the like is also included in the “room wall” in the round traveling along the wall. There are first to third determination methods for determining whether or not the vehicle travels along the wall. Each of the first to third determinations is based on the logic of first determining unsuccessful one-round travel along the wall and determining successful if not unsuccessful. The success / failure judgment of the round trip along the wall is made.

  The first determination is made in a determination block (processing step) 101. In the determination block 101, a determination process is performed based on the elapsed time of travel from the travel start time. Specifically, the elapsed time from the travel start time is counted by the time measuring means 3, and the elapsed time is a fixed value, that is, a reference time which is set in advance as required for one round travel along the wall according to the size of the room. In this case, it is assumed that the vehicle has traveled as much as necessary to travel around the wall, but has not yet succeeded in traveling around the wall. When the round turn along the wall judging means 6 rejects the round turn judgment, the switch 10 outputs the servo command value from the random running generating means 9 to the servo drive system 11, whereby the self-propelled vacuum cleaner runs randomly. To clean the room. On the other hand, if the elapsed time does not exceed the reference time, the round trip along the wall is not unsuccessful. In the case of only the first determination, the switch 10 outputs the servo command value from the regular travel generation means 8 to the servo drive system 11 based on this one-round determination, whereby the self-propelled cleaner is Clean the room by regular driving.

  The second determination is performed by the determination block 102 and the determination block 103. In the determination block 102, the direction angle of the self-propelled cleaner from the start of travel, that is, the amount of rotation angle from the start of travel to the present by the direction rotation operation that the self-propelled cleaner performs to change the travel direction during travel. Judgment is made. Specifically, the accumulated rotation angle (direction angle) from the start of travel to the present is compared with a reference rotation angle (reference direction angle) α that is set in advance according to the shape of the room. When the rotation angle is equal to or larger than the reference rotation angle α, the round determination is made. When only the second determination or only the combination of the first determination and the second determination is performed, the switch 10 converts the servo command value from the regular travel generation means 8 to the servo drive system based on this one-round determination. 11 to output. Note that the reference rotation angle α is set so as to travel around the wall according to the shape of the room as described above. However, in the case of a general square room having four corners, 360 is 360. What is necessary is just to set arbitrarily in the range of less than degree. Further, as shown in FIG. 4, the cumulative rotation angle of the self-propelled cleaner is integrated with the rotation angle in the direction change during traveling, with the counterclockwise direction being positive with respect to the traveling direction and the clockwise direction being negative. By doing so, even when the wall surface of the room is uneven as in the example of FIG. 4, the integrated rotation angle after traveling around the wall is ideally 360 degrees.

  If it is determined in the determination block 102 that the rotation angle has not reached the reference rotation angle α, the determination by the determination block 103 is subsequently performed. In the determination block 103, the determination is made based on the travel distance from the travel start time. Specifically, when the travel distance accumulated from the start of travel is a fixed value, that is, a reference travel distance that is set in advance as required for one round travel along the wall according to the size of the room, the reference rotation angle α Although it is not rotating, it has traveled more than the distance necessary for a round trip along the wall, so that it has not succeeded in a round trip along the wall, and the round judgment is rejected. As described above, the random running is selected when it is determined that the one-round determination is rejected. If it is determined in the determination block 103 that the travel distance is equal to or less than the reference travel distance, the process returns to the determination block 101.

  The third determination is made in determination blocks 104-106. In the determination block 104, determination based on the distance between the current position of the self-propelled cleaner and the travel start position is performed. Specifically, if the distance between the current position / travel start position is equal to or less than a predetermined value, that is, equal to or less than a preset reference distance, it is determined that the travel has been completed along the wall and returned to the travel start position. When the determination by the determination block 107 described later is omitted, the switch 10 causes the servo drive system 11 to output the servo command value from the regular travel generation means 8 based on this one-round determination. On the other hand, if the distance between the current position and the travel start position is equal to or greater than the reference distance, the determination by the determination block 105 is subsequently performed.

  In the determination block 105, the same determination as in the determination block 102 is performed. However, if the reference rotation angle β different from the reference rotation angle α in the determination block 102 is used as the reference rotation angle in the determination block 105 and the rotation angle from the start of travel is equal to or greater than the reference rotation angle β, it is far from the travel start position. Although it is away, it has already made a rotation of the reference rotation angle β or more, so it is assumed that it has not succeeded in the one-round traveling along the wall, and the one-round determination is rejected. On the other hand, if the rotation angle from the start of traveling has not reached the reference rotation angle β, the determination by the determination block 106 is subsequently performed. The reference rotation angle β may be arbitrarily set as a larger angle than the reference rotation angle α.

  In the determination block 106, similarly to the determination block 103, the travel distance from the travel start time is determined. If the mileage is greater than or equal to the reference mileage, it does not succeed in making a round trip along the wall because it does not rotate at the reference rotation angle β, but runs more than the distance required for a round trip along the wall. It is assumed that it is a round and the round is not judged. If it is determined in the determination block 106 that the travel distance is equal to or less than the reference travel distance, the process returns to the determination block 101.

  For the third determination as described above, the determination accuracy can be improved by adding a determination by the determination block 107. The determination block 107 makes a determination based on the deviation between the current direction angle of the self-propelled cleaner and the initial direction angle at the start of the travel. The success or failure of the round trip along the road is judged. Specifically, the direction angle deviation between the current direction angle of the self-propelled vacuum cleaner and the initial direction angle at the start of traveling is equal to or less than a preset reference deviation as a condition for successful round-the-wall traveling. If it is satisfied, the round determination is made, and if it is not satisfied, the round determination is denied.

  FIG. 5 and FIG. 6 show examples of running states when the round trip along the wall succeeds and fails. FIG. 5a is an example of success. As shown in the example of this figure, when the round trip along the wall is successful, the end point position of the self-propelled cleaner that has returned around the wall overlaps with the trip start point, and the direction angle at that time is 360 degrees + a minute angle Δ, and the difference from the initial direction angle 0 degree is a minute angle Δ, so that it is determined that the one-round traveling along the wall is successful by the determination process described above.

  FIG. 5B shows an example of a case where it is determined that the vehicle has made an unsuccessful round trip along the wall in the determination process described above. In the example of FIG. 5b, the direction angle is 355 degrees, and if the reference rotation angle α is set to 350 degrees in the determination block 102 in the second determination, for example, the round is determined because it exceeds this. However, the distance between the current position / travel start position when the direction angle becomes 355 degrees is d, and the distance d between the current position / travel start position exceeds the reference distance. It is determined that the one-round determination is negative. Such unsuccessful round trips along the wall occur when the accumulated error of the sensors that detect the position and direction angle is large, or when a large slip occurs between the self-propelled cleaner body and the floor. obtain.

  FIG. 5c is another example of the case where it is determined that the one-round traveling along the wall is unsuccessful in the determination process described above. FIG. 3c shows a case where the room to be cleaned is partly partitioned by a partition having a complicated arrangement, and this partition is also used as a wall of the room, and traveling along the partition is performed when traveling around the wall. In this example, the vehicle continues to travel even after the directional angle is already 370 degrees at the position X1, and the vehicle is still traveling even when the travel distance from the travel start point becomes L at the position X2. It has become. Therefore, the reference rotation angle α in the determination block 102 is 350 degrees, and the reference distance for the current position / travel start position distance in the determination block 104 is smaller than the current position / travel start position distance D at the position X2, If the reference rotation angle β in 105 is set to 400 degrees and the reference travel distance in the determination block 106 is set to a value smaller than the travel distance L, it is determined that the one-round determination is not made in the determination process described above.

  FIG. 6 is still another example in the case where it is determined in the above-described determination processing that unsuccessful traveling around the wall is unsuccessful. In the example of FIG. 6, when the square room A is intended to be cleaned with the door closed, the door opens and the self-propelled cleaner goes out of the room A while traveling around the wall. This is the case. In this case, the current direction angle is 180 degrees smaller than the reference rotation angle α in the determination block 102, but the elapsed time from the start of travel exceeds the reference time in the determination block 101, and the travel distance from the start of travel is This is a situation where the reference travel distance in the determination block 103 is exceeded. Therefore, it is determined that the one-round determination is rejected in the first determination or the second determination in the determination process described above.

  As described above, in the present invention, first, a round trip along the wall is attempted, and whether or not the room to be cleaned is suitable for regular running is determined by determining whether or not its success or failure is determined by a predetermined determination method. In this way, regular driving and random driving are selected for cleaning. For this reason, according to the present invention, it is possible to perform traveling control that properly uses regular traveling and random traveling according to the state of the room to be cleaned, and it is possible to perform cleaning more efficiently.

  Hereinafter, a second embodiment will be described. FIG. 7 is a block diagram showing the configuration of the travel control system in the self-propelled cleaner according to the second embodiment. The travel control system in the self-propelled cleaner according to the present embodiment has a configuration in which elements related to obstacles are added to the travel control system in the self-propelled cleaner according to the first embodiment. Only the portion is shown, and the illustration of the configuration common to the first embodiment is omitted.

  The travel control system for obstacles treats obstacles such as walls of the room to be cleaned and furniture placed in the room as obstacles, especially with recesses. In the case of performing random running, with respect to an obstacle that may leave the inside of the recess cleaned by reflecting in the direction of the random angle at the corner of the recess, the inside of the recess is surely cleaned, In addition, running control is performed so as to effectively avoid the redundant cleaning and increase the cleaning efficiency.

  The obstacle traveling control system having the traveling control function as described above includes an obstacle detecting means 21, an obstacle position calculating means 22, an obstacle position storing means 23, a map generating means 24, a map storing means 25, and a position comparing means. 26, an imaging means 27, an image processing means 28, an image database creation means 29, an image database 30, an image comparison means 31, an identical obstacle passage determination means 32, a switch 33 and a follow travel generation means 34. The data is taken in from the position / direction angle calculation means 2 in the travel control system in the embodiment, and the random travel servo command value generated by the random travel generation means 9 in the travel control system in the first embodiment is also taken in. Has been.

  The obstacle detection means 21 is configured by appropriately combining, for example, a non-contact type position detection sensor such as an ultrasonic sensor or an infrared sensor, or a contact type touch sensor that emits a signal when contacted with an obstacle. When detected, a signal is generated. The obstacle position calculation means 22 obtains the position of the obstacle from the data detected by the obstacle detection means 21 and stores the obtained obstacle position in the obstacle position storage means 23.

  The map generation means 24 is configured to detect obstacles on the travel route from the current position / direction angle of the self-propelled cleaner obtained by the position / direction angle calculation means 2 and the position of the obstacle obtained from the obstacle position storage means. Generate map information about. This obstacle map information is, for example, divided into a route of travel, that is, the floor of the room to be cleaned, into sections of a certain size, whether there are obstacles in each section, and if there are obstacles, It is configured by describing information such as whether the obstacle is an obstacle placed along the wall of the room or along the wall, or an obstacle (internal obstacle) placed away from the wall. A configuration example of such obstacle map information is shown in FIG. In this example, M1, M2, M3,... Mn are given to the sections as the section names, and the respective center positions of the sections are assigned coordinate values (X1, Y1), (X2, Y2), (X3, Y3),. (Xn, Yn) is recorded in the form of a table that records the attributes (room walls, internal obstacles, etc.) when there are obstacles for each section, and “none” when there are no obstacles. Obstacle map information is configured. Such obstacle map information generated by the map generation means 24 is stored in the map storage means 25.

  The position comparison means 26 is one of the means for determining whether the obstacle has already passed or not, and uses the obstacle map information from the map storage means 25 to determine whether the obstacle has passed or not passed. The determination is made as follows. Now, it is assumed that the self-propelled cleaner detects an obstacle while traveling and the position coordinates (xi, yi) are obtained for the obstacle. Then, referring to the obstacle map information, the record of the obstacle is examined for the section Mi having the center position coordinate (Xi, Yi) closest to the position coordinate (xi, yi). If the result is that there is no obstacle, it is determined that the detected obstacle is an unpassed obstacle, and “following traveling” is set in the traveling condition flag. On the other hand, when an obstacle has been recorded, “random traveling” is set in the traveling condition flag on the assumption that the detected obstacle is an already-passed obstacle. The accuracy of the obstacle map information decreases if there is a cumulative error in the position / direction angle measurement by the position / direction angle measurement means 1. Therefore, it is preferable to improve the accuracy of the determination by constantly updating the obstacle map information at appropriate time intervals to limit the target range for the determination. The same applies to the determination by the image comparison means 31 described later.

  The imaging unit 27 is configured by, for example, a CCD camera and acquires an image about the periphery of the traveling self-propelled cleaner. The surrounding image acquired by the imaging unit 27 is subjected to processing such as binarization by the image processing unit 28, and then the position / direction angle data regarding the imaging point from the position / direction angle calculation unit 2 in the image database creation unit 29. The image database is combined with the image data and stored in the image database 30. An example of image data stored in the image database 30 is shown in FIG. In this example, Pict1, Pict2,... Are given to each image as image names, and position / direction angle data (Rx1, Ry1, Rθ1), (Rx2, Ry2, Rθ2),. The image data is configured in the form of a table recording such as.

  The image comparison means 31 is another means for determining whether an obstacle has already passed or not, and uses the data from the image database 30 to determine whether the obstacle has passed or not passed. The determination is made as follows. When the self-propelled cleaner detects an obstacle while traveling, the imager 27 captures an image around the obstacle, and the position / direction angle of the self-propelled cleaner at the imaging point is determined to be a position / Obtained from the direction angle calculation means 2. Then, the surrounding image acquired when the obstacle is detected is compared with the surrounding image stored in the image database 30. Specifically, an already-stored surrounding image having a position / direction angle closest to the position / direction angle in the surrounding image acquired at the time of obstacle detection is selected from the image database 30, and this already-stored surrounding image and the surrounding image acquired at the time of obstacle detection are selected. Are matched and pattern matching is performed. If the patterns of both images coincide with each other within a predetermined error range, “random running” is set in the running condition flag, assuming that the detected obstacle is an already passed obstacle. On the other hand, if the patterns of the two images do not match, pattern matching is performed again after the image is subjected to processing such as rotation, reduction, and enlargement. If the pattern matches both images in this second pattern matching, “Random driving” is set in the driving condition flag on the assumption that the detected obstacle is an obstacle that has already passed, and pattern matching is performed again. However, if the patterns do not match, “following running” is set in the running condition flag, assuming that the detected obstacle is an unpassed obstacle. At the same time, the surrounding image acquired at the time of obstacle detection regarding the obstacle that has not passed is stored in the image database 30 together with the position / direction angle thereof.

  The same obstacle passage determination means 32 is sent from each of the position comparison means 26 and the image comparison means 31 to determine whether the obstacle detected during the traveling of the self-propelled cleaner is already passed or not passed. A determination is made based on the coming travel condition flag, and the result is output to the switch 33. Upon receiving this, the switch 10 switches the servo command value output to the servo drive system 11. That is, the servo drive system 11 is switched by switching the servo command value for the follow travel generated by the follow travel generation unit 34 or the servo command value for the random travel generated by the random travel generation unit 9 according to the determination result by the same obstacle passage determination unit 32. In response to this, under the control of the servo drive system 11, the traveling mechanism system 12 (FIG. 1) causes the self-propelled cleaner to perform traveling or random traveling. In order to determine that the obstacle has not passed in the same obstacle passage determination means 32, it is determined that both of the traveling condition flags sent from the position comparison means 26 and the image comparison means 31 are “following traveling”. The condition is preferable in terms of determination accuracy, but it is not always necessary to do so.

  Here, “following traveling” means that after traveling for a predetermined distance or a predetermined time so as to follow the uneven structure of the peripheral surface such as the concave portion Sr in the obstacle S as in the example shown in FIG. This is a driving mode that returns to random driving. By using this driving mode, it is possible to reduce the generation of uncleaned parts as much as possible. That is, when only random driving is performed, the obstacle has a concave portion (this is not only in the concave portion Sr as in the example of FIG. 10a on the side of the obstacle, but also in an obstacle having a lower space such as a table or a chair. If there is a lower space), there is a high possibility that the inside of the recess will be cleaned and left behind by reflecting in a random angular direction at the corner of the recess. Thus, it is possible to surely enter the inside of the recess and perform cleaning, and it is possible to reduce the generation of a portion left uncleaned as much as possible. However, there is a possibility that the same obstacle will hit in random driving, and if the driving is repeated for the same obstacle in that case, cleaning will be duplicated and the cleaning efficiency will be reduced. To prevent this, the above-mentioned obstacle passing / non-passing judgment regarding the obstacle and switching between the following traveling and the random traveling based on the determination. That is, by this switching operation, the following traveling is performed only for the obstacle that is first encountered, and for the obstacle that has already been encountered, random traveling is performed as in the example shown in FIG. Thus, it is possible to effectively prevent the occurrence of repeated cleaning.

  The present invention can improve cleaning efficiency by properly using regular traveling and random traveling according to the situation of the room to be cleaned for the self-propelled cleaner. The present invention as described above can be widely applied as useful in the field of self-propelled cleaners.

It is a figure showing the composition of the run control system in the self-propelled cleaner by a 1st embodiment in the form of a block diagram. It is a figure which shows the example of each driving | running pattern of regular driving | running | working and random driving | running | working in a square room. It is a figure which shows the flow of the process in the success-and-failure determination of a round trip along a wall. It is a figure explaining the integration | accumulation of the rotation angle in one round run along a wall. It is a figure which shows the example of the driving | running | working state in a round run along a wall. It is a figure which shows the other example of the driving | running | working state in a round run along a wall. It is a figure showing the composition of the run control system in the self-propelled cleaner by a 1st embodiment in the form of a block diagram. It is a figure which shows the structural example of obstacle map information. It is a figure which shows the example of image data. It is a figure explaining the switching of follow driving and random running.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Position / direction angle measurement means 2 Current position / direction angle calculation means 3 Time measurement means 6 Round circumference determination means 7 Travel route storage means 8 Regular travel generation means 9 Random travel generation means 10 Switch 11 Servo drive system 12 Travel mechanism System 21 Obstacle detection means 22 Obstacle position calculation means 23 Obstacle position storage means 25 Map storage means 26 Position comparison means 27 Imaging means 28 Image processing means 29 Image database creation means 30 Image database 31 Image comparison means 32 Same obstacle passing Determining means 33 Switch 34 Profiling generation means

Claims (2)

In a self-propelled cleaner that performs cleaning while traveling on the floor surface of the space to be cleaned by a traveling mechanism,
A regular travel generation means for generating a regular travel mode for traveling according to a predetermined travel route, a random travel generation means for generating a random travel mode for performing a travel in which the travel direction is changed randomly, and the regular travel generation means; A switch for selectively outputting the output from each of the random travel generation means to the travel mechanism, and a first determination method for determining the success or failure of the one-round travel along the wall based on the elapsed time of travel from the travel start time; A round trip along the wall combining the determination made by comparing the rotation direction angle from the start of travel with a preset reference rotation angle α and the determination made by comparing the travel distance from the start of travel with a preset reference travel distance Second determination method for determining success or failure of travel, determination based on distance between current position during travel and travel start position, and rotation direction from travel start time Along the wall by combining the determination made by comparing with the preset reference rotation angle β greater than the reference rotation angle α and the determination made by comparing the travel distance from the start of travel with the preset reference travel distance. One of the third determination methods for determining the success or failure of the one-round traveling, or a one- round determination means along the wall that combines these determination methods , and the first round-trip traveling along the wall is first tried and the success or failure is determined along the wall. According to the result, if the round trip along the wall is successful, the switch outputs the output from the regular trip generation means to the running mechanism according to the result. When traveling around the wall is unsuccessful, the output from the random travel generating means is output to the travel mechanism by the switch to travel in the random travel mode. Self-propelled cleaner, characterized in that it is adapted to perform. The vehicle is provided with a travel generation means for generating a travel mode for causing the obstacle to cause a random travel direction change at the time of the random travel to follow the peripheral surface of the obstacle, and is detected during the travel. The self-propelled cleaning device includes: an imaging unit that captures an image of the obstacle centered on the obstacle; and a storage unit that stores a captured image captured by the imaging unit together with a position / direction angle of a self-propelled cleaner. Comparison between an image captured by the imaging unit and a surrounding image obtained by pattern matching between the image captured by the imaging unit and the image stored in the storage unit based on the position / direction angle of the self-propelled cleaner while the machine is running A passing / non-passing determining means for determining whether the obstacle has passed or not passed by a determination method for determining whether the obstacle has passed or not passed, In claim 1 for the non-passed and the determined obstacle over determining means are adapted to perform a travel in the copying movement pattern by outputting an output from the copying traveling generating means to the driving mechanism The self-propelled vacuum cleaner described.

Images