WO2022249486A1 - 情報処理システム、作業機及びプログラム - Google Patents
情報処理システム、作業機及びプログラム Download PDFInfo
- Publication number
- WO2022249486A1 WO2022249486A1 PCT/JP2021/020544 JP2021020544W WO2022249486A1 WO 2022249486 A1 WO2022249486 A1 WO 2022249486A1 JP 2021020544 W JP2021020544 W JP 2021020544W WO 2022249486 A1 WO2022249486 A1 WO 2022249486A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- work
- route
- information processing
- processing system
- area
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 173
- 238000001514 detection method Methods 0.000 claims description 37
- 238000012545 processing Methods 0.000 claims description 36
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 17
- 230000008859 change Effects 0.000 claims description 11
- 238000010586 diagram Methods 0.000 description 26
- 230000006870 function Effects 0.000 description 10
- 238000003860 storage Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 230000001133 acceleration Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 244000025254 Cannabis sativa Species 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003337 fertilizer Substances 0.000 description 1
- 230000035876 healing Effects 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 239000002096 quantum dot Substances 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000003971 tillage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/60—Intended control result
- G05D1/648—Performing a task within a working area or space, e.g. cleaning
- G05D1/6484—Performing a task within a working area or space, e.g. cleaning by taking into account parameters or characteristics of the working area or space, e.g. size or shape
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/20—Control system inputs
- G05D1/24—Arrangements for determining position or orientation
- G05D1/247—Arrangements for determining position or orientation using signals provided by artificial sources external to the vehicle, e.g. navigation beacons
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/60—Intended control result
- G05D1/648—Performing a task within a working area or space, e.g. cleaning
- G05D1/6482—Performing a task within a working area or space, e.g. cleaning by dividing the whole area or space in sectors to be processed separately
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D2101/00—Details of software or hardware architectures used for the control of position
- G05D2101/10—Details of software or hardware architectures used for the control of position using artificial intelligence [AI] techniques
- G05D2101/15—Details of software or hardware architectures used for the control of position using artificial intelligence [AI] techniques using machine learning, e.g. neural networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D2105/00—Specific applications of the controlled vehicles
- G05D2105/15—Specific applications of the controlled vehicles for harvesting, sowing or mowing in agriculture or forestry
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D2107/00—Specific environments of the controlled vehicles
- G05D2107/20—Land use
- G05D2107/23—Gardens or lawns
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D2109/00—Types of controlled vehicles
- G05D2109/10—Land vehicles
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D2111/00—Details of signals used for control of position, course, altitude or attitude of land, water, air or space vehicles
- G05D2111/30—Radio signals
- G05D2111/36—Radio signals generated or reflected by cables or wires carrying current, e.g. boundary wires or leaky feeder cables
Definitions
- the present invention relates to an information processing system, a work machine and a program.
- Patent Document 1 discloses a technique in which a virtual marker in an image is identified, boundary data indicating the boundary of the work area is generated, and the boundary data is generated within the work area of a robot lawnmower. A technique for controlling the operation of the is disclosed.
- the present invention provides an information processing system etc. that can improve the possibility of setting a route with high work quality in the work area.
- an information processing system In the acquisition step in this information processing system, boundary information indicating the boundary of the work area targeted by the work machine capable of autonomous travel is acquired. In the setting step, a work machine path is set to eliminate an unreached area from the work area based on the obtained boundary information, and the work quality when using the path set by the first method does not meet a predetermined standard. If the conditions are not satisfied, the route is set by the second method.
- FIG. 1 is a diagram showing the appearance of a lawnmower robot 1.
- FIG. 4 is a diagram showing an example of a travel route of the lawnmower robot 1.
- FIG. 2 is a diagram showing the hardware configuration of the lawnmower robot 1.
- FIG. 1 is a diagram showing a functional configuration of a lawnmower robot 1;
- FIG. It is a figure which shows the operation
- FIG. 4 is a diagram illustrating a first example of a work target area;
- FIG. 4 is a diagram showing an example of a route set by a first route setting method;
- FIG. 10 is a diagram showing an example of a route set by a second route setting method;
- FIG. 10 is a diagram showing the processing load in the work target area A10;
- FIG. 10 is a diagram illustrating a second example of a work target area;
- FIG. 4 is a diagram showing an example of a route set by a first route setting method;
- FIG. 10 is a diagram showing an example of a route set by a second route setting method;
- FIG. It is a figure which shows an example of the route set by route setting processing.
- FIG. 10 is a diagram showing the processing load in the second example;
- the program for realizing the software appearing in this embodiment may be provided as a non-transitory computer-readable medium (Non-Transitory Computer-Readable Medium), or may be downloaded from an external server. It may be provided as possible, or may be provided so that the program is activated on an external computer and the function is realized on the client terminal (so-called cloud computing).
- the term “unit” may include, for example, a combination of hardware resources implemented by circuits in a broad sense and software information processing that can be specifically realized by these hardware resources.
- various information is handled in the present embodiment, and these information are, for example, physical values of signal values representing voltage and current, and signal values as binary bit aggregates composed of 0 or 1. It is represented by high and low, or quantum superposition (so-called quantum bit), and communication and operation can be performed on a circuit in a broad sense.
- a circuit in a broad sense is a circuit realized by at least appropriately combining circuits, circuits, processors, memories, and the like.
- Application Specific Integrated Circuit ASIC
- Programmable Logic Device for example, Simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), and field It includes a programmable gate array (Field Programmable Gate Array: FPGA)).
- the hardware configuration of the self-propelled lawnmower robot which is the work machine according to the present embodiment, will be described. That is, in this embodiment, the work performed by the work machine is lawn mowing. Further, the work machine is a work machine capable of autonomous travel, and includes each part of an information processing system for controlling operations related to work (lawn mowing).
- FIG. 1 is a diagram showing the appearance of the lawnmower robot 1.
- the lawn mowing robot 1 includes a blade disk, tires, a motor for rotating them, a battery, and the like inside a housing.
- the lawn mowing robot 1 cuts the lawn to a predetermined height by rotating the blade disk by the work motor while traveling on a grassy field by means of a drive motor.
- the lawn-mowing robot 1 sets its own running route in the field by the above-described information processing system provided in its own device, and mows the lawn while traveling along the set running route.
- FIG. 2 is a diagram showing an example of the travel route of the lawnmower robot 1.
- FIG. FIG. 2 shows a rectangular work area A1.
- FIG. 2 shows a direction D1 and a direction D3 along the short sides of the work area A1, and a direction D2 and a direction D4 along the long sides of the work area A1.
- the work target area A1 has a boundary B1 that serves as a boundary between the inside and outside of the own area. It is assumed that wires are embedded along the boundary B1 in the work target area A1.
- the lawnmower robot 1 By recognizing the position of the wire in a two-dimensional or three-dimensional space, the lawnmower robot 1 recognizes the shape and size formed by the boundary B1, that is, the shape and size of the outer periphery of the work area A1. The lawnmower robot 1 sets a travel route based on the recognized shape and size of the work area A1. The lawnmower robot 1 sets a work path C1 in the example of FIG. The work route C1 has a starting point P1 at the corner of the direction D1 and the direction D4 of the work target area A1.
- the work path C1 starts from the starting point P1 in the direction D2, and when it reaches the boundary B1, it changes direction in the direction D3, moves by the width of the lawnmower robot 1, and then changes direction again in the direction D4. It is the route to The work route C1 is a route that turns back along the route that has been advanced due to the direction change of two degrees in this way.
- the direction of movement during folding (here, direction D3) is hereinafter referred to as the "folding direction".
- the work route C1 is a route that travels in the direction D4 and turns back in the direction D2 with the direction D3 as the turning direction when reaching the boundary B1.
- the work path C1 is thus a reciprocating path that repeats movements in the directions D2 and D4 while gradually deviating in the direction D3, which is the turn-back direction. The details of how to set the work path will be described later.
- FIG. 3 is a diagram showing the hardware configuration of the lawnmower robot 1. As shown in FIG.
- the lawn mowing robot 1 includes a control section 11 , a storage section 12 , a communication section 13 , a sensor section 14 , a traveling section 15 and a working section 16 .
- control unit 11 The control unit 11 is, for example, a central processing unit (CPU) (not shown).
- the control unit 11 implements various functions related to the lawnmower robot 1 by reading a predetermined program stored in the storage unit 12 . That is, information processing by software stored in the storage unit 12 can be specifically realized by the control unit 11 which is an example of hardware, and can be executed as each function unit included in the control unit 11 . These are further detailed in the next section. Note that the control unit 11 is not limited to a single unit, and a plurality of control units 11 may be provided for each function. A combination thereof may also be used.
- the storage unit 12 stores various information defined by the above description. This may be, for example, a storage device such as a solid state drive (SSD) that stores various programs related to the lawnmower robot 1 executed by the control unit 11, or a temporary storage device related to program calculation. It can be implemented as a memory such as a Random Access Memory (RAM) that stores information (arguments, arrays, etc.) required for the .
- the storage unit 12 stores various programs, variables, etc. related to the lawnmower robot 1 executed by the control unit 11 .
- the communication unit 13 is configured to be able to transmit various electrical signals from the lawnmower robot 1 to external components. Also, the communication unit 13 is configured to be capable of receiving various electrical signals from external components to the lawnmower robot 1 . More preferably, the communication unit 13 has a network communication function, so that various information can be communicated between the lawnmower robot 1 and an external device via a network such as the Internet.
- the sensor unit 14 has a group of sensors that measure various values used by the lawnmower robot 1 as processing parameters.
- the sensor unit 14 includes, for example, a positioning sensor, an orientation sensor, an angular velocity sensor, an acceleration sensor, a blade disk torque sensor, a wire detection sensor, an object detection sensor, and the like.
- the positioning sensor measures the position of its own device by GPS (Global Positioning System) or the like.
- the orientation sensor measures the orientation of the device itself.
- the angular velocity sensor measures the angular velocity when the device rotates.
- the acceleration sensor measures the acceleration of its own device.
- a torque sensor on the blade disc detects the load on the motor as the mowing blade rotates.
- the aforementioned control section 11 calculates the estimated work amount of the blade disk.
- the wire detection sensor detects wires embedded in the boundary B1 or the like in FIG.
- the object detection sensor emits millimeter waves or the like and detects objects that reflect them.
- the running unit 15 includes tires, a motor, and the like, and runs the own device.
- the traveling unit 15 is controlled by the control unit 11 to move straight, reverse, change directions, and the like.
- the working unit 16 includes a blade disk, a motor, etc., and is controlled by the control unit 11 to rotate the blade disk to mow the lawn.
- FIG. 4 is a diagram showing the functional configuration of the lawnmower robot 1.
- the lawn mowing robot 1 includes a boundary detection unit 111 , an object detection unit 112 , a setting unit 113 , a determination unit 114 , a travel control unit 115 and a work control unit 116 .
- the boundary detection unit 111 detects the boundary of the work target area on which the own device performs work. For example, in the example of FIG. 2, the boundary detection unit 111 first detects wires embedded along the outer circumference of the work target area A1.
- the boundary detection unit 111 detects a wire using, for example, a well-known technique used in wire-based autonomous traveling described in Japanese Unexamined Patent Application Publication No. 2016-208950.
- the boundary detection unit 111 detects the boundary B1 of the work target area A1 by running along the detected wire on the outer circumference.
- the boundary detection unit 111 acquires detection result data represented by a set of coordinates on the two-dimensional or three-dimensional space of the detected boundary B1.
- the boundary detection unit 111 is an example of a function that executes the "acquisition step", and the detection result data is an example of "boundary information”.
- the boundary detection unit 111 (acquisition unit) in the information processing system acquires boundary information indicating the boundary of the work area targeted by the autonomously traveling lawnmower robot 1 (working machine).
- the boundary detection unit 111 supplies the acquired detection result data to the setting unit 113 .
- the object detection unit 112 detects objects existing in the work area.
- the object detection unit 112 is an example of a function that executes a "detection step".
- Objects that is, obstacles
- Obstacles are, for example, natural objects such as stones and rocks, and artificial objects such as tools left behind by people.
- the object detection unit 112 also detects a wire embedded along the boundary B1 as an obstacle.
- the object detection unit 112 can detect the object even if the distance from the object is somewhat long.
- the object detection unit 112 supplies object information indicating the position and direction of the detected object (obstacle) to the setting unit 113 and the travel control unit 115 .
- the setting unit 113 Based on the boundary information acquired by the boundary detection unit 111, the setting unit 113 sets the route of the work machine to eliminate the unreached area from the work area.
- the setting unit 113 is an example of a function that executes a “setting step”.
- the setting unit 113 sets the work path by two methods, a first path setting method and a second path setting method, which will be described below. Both the first route setting method and the second route setting method are methods of setting a route based on a learned model learned by a machine learning technique.
- NN Neural Network
- NN is a learning model having a multi-layered network structure such as an input layer, an intermediate layer, and an output layer.
- a learning model is acquired by optimizing multiple model parameters inside the network using an algorithm such as error backpropagation using learning data that indicates the relationship between input data and output data. can be done.
- the first route setting method is a method of learning a model for selecting the optimum one from a plurality of predetermined action rules by the NN technique and setting a route based on the learned model. be.
- a classifier system is a set of rules (classifiers) consisting of a condition part and an action part, and learns the appropriateness of the rules based on the results of outputting actions in response to input from the environment according to those rules.
- the second route setting method acquires environmental information each time a certain distance is traveled and learns a model that sets an optimal route using an NXCS (Neural-network eXtended Classifier System).
- NXCS Neuro-network eXtended Classifier System
- the first route setting method it is possible to simplify processing related to route selection within relatively simple terrain and improve work efficiency.
- learning by the classifier system as the second route setting method, it is possible to set the optimum route even in situations where it is difficult to respond with the first route setting method, maintaining efficiency and work quality in various terrains. work is performed. It should be noted that it is desirable to switch to the first route setting method when it is determined that the work by the second route setting method is completed. By combining them in this manner, the respective features of the first and second learning methods are utilized to the maximum extent, unnecessary processing load and processing time are eliminated, and work efficiency is further improved.
- the first route setting method is an example of the "first method”
- the second route setting method is an example of the "second method”.
- the first method and the second method are, as described above, methods of setting routes based on learned models learned by different machine learning methods, NN and NXCS in this embodiment, respectively. By using machine learning in this way, the more the route setting is repeated, the more desirable the work route is set.
- the setting unit 113 When setting a route using the first route setting method, the setting unit 113 first sets the route using the boundary information supplied from the boundary detection unit 111 . Specifically, the setting unit 113 sets a route using a method of setting a continuous route as a first method according to a rule that the working machine repeats movement and direction change at the boundary of the work area. By using such a rule, even if the size of the work target area changes, even if the distance traveled increases or the number of turnarounds increases slightly, the processing load when setting the work route changes. can be suppressed.
- the setting unit 113 has a plurality of rules in which at least one of the initial direction of travel and the direction after direction change is different, and selects the rule with the highest work quality from among the plurality of rules to set the route. method as the first method to set the route.
- the work quality is the quality of work performed by the lawnmower robot 1, which is a work machine. For example, when the lawnmower robot 1 has been working for a certain period of time, it can be said that the more areas where the work is completed, the higher the quality of the work is.
- the work quality when mowing the entire area is completed after working for 30 minutes and the work quality when mowing only half of the area is completed the former work quality is higher.
- the work quality evaluation method is not limited to this.
- the shorter the time required to complete the work for the entire area the higher the work quality, or the shorter the distance traveled to complete the work for the entire area, the higher the work quality. good.
- the work route can be determined by the first route setting method compared to the case where only one rule is fixed. It is possible to improve the work quality when it is set.
- the work quality selected and set in order to improve the work quality in this way is much higher than the work quality in the case of random travel without the travel rule or learning function.
- the plurality of rules are, for example, when the work target area is rectangular, a rule that uses the direction along the long side and the direction along the short side as the initial traveling direction and turning direction, respectively, as in the example of FIG. is.
- the setting unit 113 uses the following four rules in this embodiment.
- the setting unit 113 sets the route according to the obstacle avoidance rule when the lawnmower robot 1 travels along the work route set before work.
- the setting unit 113 uses the object information supplied from the object detection unit 112 to recognize that there is an obstacle on the work route. If the recognized obstacle can be bypassed, the setting unit 113 resets the work route that bypasses the obstacle and returns to the original work route. Further, when the recognized obstacle cannot be detoured (for example, when the boundary of the work target area is reached before detouring and returning to the original work route), the setting unit 113 returns to the point where the detour was started and Reconfigure the work route for turning back.
- the setting unit 113 uses any one of the rules 1 to 4 and the obstacle avoidance rule when resetting the work route in the first route setting method.
- the setting unit 113 also resets the work path when using the second path setting method, but differs from the first path setting method in that it is not restricted by a specific rule.
- the setting unit 113 sets the route based on the boundary information acquired by the boundary detection unit 111 and the position of the object detected by the object detection unit 112 each time the working machine travels a predetermined distance.
- a route is set using the healing method as the second route setting method.
- the setting unit 113 sets a new optimum route when, for example, an obstacle is found in the middle of the route and the route up to that point becomes impassable. In other words, there is a possibility that an area once inaccessible due to an obstacle or the like can be reached by a newly set route. In this way, when the second route setting method is used, it is possible to reduce areas that cannot be reached compared to the first route setting method, in which the route once set is not changed during driving. can.
- the setting unit 113 supplies route information indicating the set route to the determination unit 114 .
- the determination unit 114 determines whether or not a switching condition for switching the route setting method from the first route setting method to the second route setting method is satisfied. Specifically, the determination unit 114 determines, as a switching condition, whether or not the work quality when using the route set by the first route setting method satisfies a predetermined standard.
- the determination unit 114 determines that a predetermined criterion is satisfied when the difference between the determined turnaround position on the set route and the position where the vehicle actually runs and bumps into some object is less than a threshold. Conversely, the determination unit 114 determines that the work quality does not meet the predetermined criteria when the aforementioned difference is equal to or greater than the threshold. If the position is greatly deviated from the position when traveling according to the set route, for example, there are obstacles on the route, or the route is not flat and sloped or uneven, making it impossible to travel straight. Conceivable.
- the determination unit 114 determines that the work quality does not meet the predetermined criteria. If the determining unit 114 determines that the work quality does not satisfy the predetermined standard, it notifies the setting unit 113 of that effect.
- the setting unit 113 receives the notification from the determination unit 114, that is, when the work quality when using the route set by the first route setting method does not satisfy a predetermined standard, the setting unit 113 sets the second route. Set route by method.
- the running control unit 115 controls the motion of the own device (the lawn mowing robot 1) when running.
- the traveling control unit 115 acquires the position and orientation of the own device, and controls the traveling operation of the own device so as to move forward, backward, and change direction in a predetermined direction.
- the work control unit 116 controls the operation of the own device (the lawn mowing robot 1) during work. In this embodiment, the work control unit 116 controls the operation of rotating the blade disc to mow the lawn while the vehicle is running.
- the lawnmower robot 1 executes the route setting process for setting the route of the work target area using the functions described above.
- FIG. 5 is a diagram showing the operating procedure in the route setting process.
- the route setting process is started, for example, when the lawnmower robot 1 is arranged at the start point of the work area.
- the lawnmower robot 1 acquires boundary information indicating the boundary of the work area by detecting, for example, wires embedded along the outer periphery of the work area using the boundary detection unit 111 (step S11).
- the lawnmower robot 1 recognizes, for example, a portion where the distance between facing boundaries in the work target area is less than a threshold value as a "passage".
- step S12 the lawnmower robot 1 recognizes the areas separated by the passages as "partial areas" (step S12).
- step S21 the lawnmower robot 1 travels and reaches one of the recognized partial regions by the travel control unit 115 (step S21).
- the point reached by the lawnmower robot 1 is the starting point of the work in that partial area.
- step S22 the lawnmower robot 1 determines whether or not the reached partial area is the target area for the second route setting method by the setting unit 113 (step S22).
- the lawnmower robot 1 determines by the setting unit 113 that the reached partial area is not the target area for the second route setting method (NO), and the boundary information acquired in step S11 , a work path is set for that partial area by the first path setting method (step S31).
- the lawnmower robot 1 travels along the work route set in step S21 by the travel control unit 115 (step S32).
- the lawnmower robot 1 determines whether or not it has traveled a certain distance (step S33).
- the lawnmower robot 1 determines that it has traveled the predetermined distance (YES)
- it selects one of the rules 1 to 4 according to the first route setting method and resets the work route (step S34).
- the constant distance may be set to infinity or a very large number so that resetting is not performed during work in the partial area.
- the lawnmower robot 1 repeatedly detects an object by the object detection unit 112, and determines whether or not it has approached an obstacle (step S35).
- An obstacle is an object that approaches the lawnmower robot 1 excluding wires.
- the lawnmower robot 1 makes a determination in step S35 even when it is determined in step S32 that it has not traveled the predetermined distance (NO).
- the setting unit 113 and the travel control unit 115 reset the work route that bypasses the obstacle in accordance with the obstacle avoidance rule (step S36). ).
- the lawn mowing robot 1 determines whether or not it has reached the end point of the work path set in the current partial area (step S37).
- the lawnmower robot 1 makes the determination in step S37 even when it determines NO in step S35. If the lawnmower robot 1 determines in step S37 that it has not reached the end point of the work path (NO), it returns to step S32 and repeats the operation. If the lawnmower robot 1 determines in step S37 that it has reached the end point of the work path (YES), it next determines whether or not a switching condition for switching the current partial area setting method is satisfied. It judges (step S41).
- the switching condition is expressed by whether or not the work quality when using the route set by the first route setting method satisfies a predetermined standard.
- the condition that is satisfied when the number of locations where the difference between the determined turn-around position on the set route and the actually traveled turn-back position is greater than or equal to the distance threshold is greater than or equal to the number threshold is switched. Used as a condition. For example, if the number threshold is 1, the switching condition is satisfied if there is at least one point where the difference between the set turn-back position and the actual turn-back position is greater than or equal to the distance threshold.
- step S41 When the lawnmower robot 1 determines that the switching condition is satisfied (YES) in step S41, it switches the setting method of the work path for the current partial area (step S42). For example, when the lawnmower robot 1 determines to switch in the first operation, it memorizes that the current partial area setting method has been switched from the first route setting method to the second route setting method. After step S42, or when it is determined that the switching condition is not satisfied (NO) in step S41, the lawnmower robot 1 determines whether or not the work end condition is satisfied (step S43).
- the work end condition is a condition for ending the work in the work target area by the lawnmower robot 1 .
- the work end condition is satisfied, for example, when the area of the unreached area in the work target area becomes less than 5% of the entire area. This work end condition is used to prevent an extreme increase in the difficulty of machine learning due to minute uncutting.
- the work target area has a simple shape as in the examples shown in FIGS. 2 and 6, there is almost no uncutting. It may be 0%.
- a method of setting the value used as the work end condition to, for example, 10% as an initial value and decreasing it gradually or step by step as the amount of learning increases can also be adopted. According to this method, the optimum end condition is set according to the contents of learning without requiring manual setting of detailed conditions prior to work.
- step S43 When the lawnmower robot 1 determines that the work end condition is satisfied (YES) in step S43, it ends this operation procedure.
- the setting unit 113 and the travel control unit 115 determine the movement path to move toward the next partial area through the passage leading to the current partial area. , and moves to the next partial area along the set moving route (step S44).
- the lawnmower robot 1 specifically sets the movement route using Rule 5 or Rule 6 for movement.
- Rule 5 Proceed clockwise around the perimeter of the work target area, and when reaching the entrance of the passage, proceed toward that passage, and when reaching the next partial area, turn left or right and proceed to the corner of that partial area.
- Rule 6 Replace “clockwise” in rule 5 with "counterclockwise”
- the setting unit 113 selects the rule 5 or rule 6, whichever has the shorter travel distance to reach the passage, and uses it for route setting.
- the lawn mowing robot 1 After moving to the next partial area, the lawn mowing robot 1 returns to step S21 and continues its operation. A series of work from the start in step S11 to the end of work in step S43 is called an "episode". Even if one episode ends, the lawnmower robot 1 starts the next episode if the grass grows over time. In the second and subsequent episodes, there may be partial areas that have been switched as target areas for the second route setting method.
- the lawnmower robot 1 determines by the setting unit 113 that the reached partial area is the target area for the second route setting method (YES), and based on the boundary information acquired in step S11, determines that partial area.
- a work path is set for the area by the second path setting method (step S51).
- the lawnmower robot 1 travels along the work path set in step S51 by the travel control unit 115 (step S52).
- the lawnmower robot 1 repeatedly detects an object by the object detection unit 112 while traveling along the work path.
- step S53 the lawn mowing robot 1 determines whether or not it has traveled a certain distance (step S53), and repeats the operation of step S53 until it determines that it has traveled (YES). Then, when the lawnmower robot 1 determines that it has traveled a certain distance, the robot 1 resets the work path by the second path setting method, taking into consideration the information on the objects detected so far (step S54). In the lawnmower robot 1, in step S53, it is determined whether or not the approach of an object has been detected. You may
- the lawnmower robot 1 determines whether or not it has reached the end point of the work path set in the current partial area (step S55). If the lawnmower robot 1 determines in step S55 that it has not reached the end point of the work path (NO), it returns to step S52 and repeats the operation. When the lawnmower robot 1 determines that the end point of the work path has been reached (YES) in step S55, it next determines whether or not the switching condition is satisfied in the current partial area (step S61). A condition for switching from the second route setting method to the first route setting method is satisfied, for example, when all obstacles detected as objects other than boundaries have shapes, positions, and sizes that allow detours. Conditions are used.
- step S61 determines in step S61 that the switching condition is satisfied (YES)
- step S62 switches the setting method of the current work path for the partial area (step S62).
- step S62 determines whether or not the work end condition is satisfied (step S63).
- step S63 When the lawnmower robot 1 determines that the work end condition is satisfied (YES) in step S63, it ends this operation procedure.
- the setting unit 113 and the travel control unit 115 determine the movement path to move toward the next partial area through the passage leading to the current partial area. , and moves to the next partial area along the set moving route (step S64).
- An example of setting a work route by route setting processing will be described below.
- FIG. 6 is a diagram showing a first example of the work target area.
- FIG. 6 shows a work target area A10 including a rectangular partial area A11 and a partial area A13 connected to the partial area A11 by a path E12.
- directions D1 to D4 similar to those in FIG. 2 are represented.
- FIG. 7 is a diagram showing an example of routes set by the first route setting method.
- the setting unit 113 sets the work route using the corner of the direction D1 and the direction D4 of the partial area A11 as the starting point P11-1.
- the end point P12-1 at the end of the partial area A11 is set as the end point of the work route.
- the setting unit 113 sets the movement route using Rule 5 or Rule 6 above.
- rule 5 requires a shorter travel distance to the passage E12 than rule 6. Therefore, the setting unit 113 uses rule 5 to determine the route from the end point P12-1 to the passage E12.
- a movement path C12-1 is set that advances to the position and passes through the path E12 from there. Further, the setting unit 113 sets a moving path C12-1 that, when reaching the boundary of the partial area A13, turns to the right and proceeds to the corners of the directions D2 and D3 of the partial area A13.
- the lawnmower robot 1 travels along the movement path C12-1 thus set by the travel control unit 115 to move to the partial area A13.
- the end point of the movement route C12-1 is the start point P13-1 of the next work route.
- the lawnmower robot 1 determines that the work end condition is satisfied when the area of the unreached area in the work target area is less than 5% of the entire area.
- FIG. 8 shows an example of a work route set using the second route setting method in the work target area A10.
- FIG. 8 is a diagram showing an example of routes set by the second route setting method.
- the setting unit 113 departs from the departure point P11-2, which is the same as in the example of FIG.
- a work path C11-2 is set that deviates in the direction D2 while reciprocating to and ends at the end point P12-2.
- the setting unit 113 sets a movement route C12-2 from the end point P12-2 to the start point P13-2 of the partial area A13 through the path E12, and sets the work route C13-2 in the partial area A13. ing.
- the object detection unit 112 resets the work route each time the vehicle travels a certain distance.
- the processing load is compared between the case where the route is set by the first route setting method and the case where the route is set by the second route setting method for the entire work target area A10. do.
- FIG. 9 is a diagram showing the processing load in the work target area A10.
- FIG. 9 shows the number of steps and the processing time when the process of setting the work path is performed in the work target area A10. In both cases, the larger the numerical value, the larger the processing load. More specifically, the period from the start point of the entire work path in the work target area to the satisfaction of the work end condition is called one episode, and the NN used in the first and second route setting methods from an unlearned state. and 500 episodes while learning for NXCS.
- the results for the best (minimum number of steps) and worst (maximum number of steps) trials among the 10 trials and the average value of the number of steps for the 10 trials are shown.
- the number of steps was minimum 2353.0, maximum 8139.0, average 2752.3, and average processing time was 984.7 seconds.
- the number of steps was minimum 2366.0, maximum 30076.0, average 17243.6, and processing time averaged 1406.8 seconds.
- the second route setting method it is difficult to set the movement route to the next partial area, and the average number of steps and average processing time are longer than in the first route setting method.
- the first route setting method since rules 1 to 10 allow the vehicle to travel efficiently, both the average number of steps and the average processing time are smaller than in the second route setting method. In this way, the first route setting method tends to have a smaller processing load than the second route setting method.
- the lawnmower robot 1 sets the work route only by the first route setting method. It's getting smaller. In this manner, the lawnmower robot 1 basically uses the first route setting method with a low processing load, and uses the second route setting method with a high processing load only when the switching condition is satisfied. By using this method, it is possible to reduce the processing load when setting the work route, compared to the case where the second route setting method is used for the basic processing.
- FIG. 10 is a diagram showing a second example of the work target area.
- FIG. 10 shows a work target area A20 including a rectangular partial area A21 and a partial area A23 connected to the partial area A21 by a path E22.
- a spiral wall F23 exists in the partial area A23, and an inner area A24 of the wall F23 is included.
- FIG. 10 shows directions D1 to D4 similar to FIG.
- FIG. 11 is a diagram showing an example of routes set by the first route setting method.
- the setting unit 113 sets the work route using the corner of the direction D1 and the direction D4 of the partial area A21 as the starting point P21-1.
- the end point P22-1 at the end of the partial area A21 is set as the end point of the work route.
- the setting unit 113 sets a movement route C22-1 passing through the path E22, and sets a work route C23-1 for the partial area A23 from the starting point P23-1 for the partial area A23. As shown in FIG. 11, in the first route setting method, the route cannot be set to the depth of the inner region A24 of the spiral wall F23. On the other hand, in the second route setting method, the route can be set to the depth of the internal area A24.
- FIG. 12 is a diagram showing an example of routes set by the second route setting method.
- the setting unit 113 sets a work path C21-2 that travels throughout the partial area A21 from the starting point P21-2 at the corner of the directions D1 and D4. Further, the setting unit 113 sets a movement route C22-2 from the partial area A21 to the partial area A23 through the path E22, and sets a work route C23-2 in the partial area A23. Furthermore, the setting unit 113 sets a work path C24-2 leading to the inner region A24 of the spiral wall F23.
- FIG. 13 is a diagram showing an example of the route set by the route setting process.
- the work path C21-3 is set by
- the setting unit 113 also sets a movement path C22-3 that moves from the end point of the work path C21-3 to the partial area A23.
- the setting unit 113 uses the second path setting method to set a work path C23-3 in the partial area A23 and a work path C24-3 leading to the inner area A24 of the wall F23.
- the setting unit 113 sets the work path for both the partial area A21 and the partial area A23 by the first path setting method for the first time, and sets the work path for the partial area A23 by the second path setting method for the second and subsequent times. set.
- FIG. 14 is a diagram showing the processing load in the second example.
- FIG. 14 shows the number of steps and the processing time when the process of setting the work path is performed in the work target area A20.
- FIG. 14 also shows the results for the best (minimum number of steps) and worst (maximum number of steps) out of 10 trials, and the average value of the number of steps for 10 trials, with 500 episodes performed as one trial.
- the number of steps was minimum 2433.0, maximum 8059.8, average 2908.3, and average processing time was 7458.3 seconds.
- the minimum number of steps was 2386.7
- the maximum was 30054.7
- the average was 17131.2
- the average processing time was 85616.8 seconds.
- the minimum number of steps is 2460.7, the maximum number is 12881.7, and the average number is 5580.8.
- the average time was 50605.1 seconds.
- the first route setting method is basically used, but the second route setting method is used depending on the area.
- the processing load it is possible to reduce the processing load while reducing the unreachable area.
- the amount of resources is limited and the upper limit of the load that can be processed is likely to occur.
- the first route setting method and the second route setting method are used properly to reduce the unreached area and improve the work quality in the work area. It is possible to improve the possibility that a route will be set.
- NN Neuron
- NXCS Neuro-network XCS
- Other machine learning methods include, for example, XCS (eXtended Classifier System), SVM (Support Vector Machine), Deep Learning, CNN (Convolutional Neural Network), RNN (Recurrent Neural Network, etc.), PCA (Private Computer) may be used.
- the setting unit 113 uses a route setting method of setting a route based on a learned model learned by a machine learning method.
- a route setting method include a method of setting a route using an algorithm devised by humans and a method of setting a route created by a method other than machine learning among AI (Artificial Intelligence) methods.
- AI Artificial Intelligence
- the determination unit 114 may use a switching condition different from the embodiment.
- the determination unit 114 may determine, for example, whether or not the switching condition is satisfied according to the length of the set work path. In general, the shorter the work path, the lower the power consumption of the work machine and the higher the work quality. Therefore, the determining unit 114 calculates, for example, the total length of an ideal working path that does not overlap at all, and calculates the ratio of the length of the set working path to the total length. The closer the ratio is to 1, the higher the work quality, and the farther from 1, the lower the work quality. The determination unit 114 determines that the switching condition is satisfied when the calculated ratio is equal to or greater than the threshold.
- the determination unit 114 may determine whether or not the switching condition is satisfied according to the number of times of direction change in the set work route. In general, the smaller the number of direction changes, the less time lost during direction changes and the higher the work quality. Therefore, the determining unit 114 calculates, for example, the number of direction changes on the work route that minimizes the number of direction changes, and calculates the ratio of the number of direction changes on the set work route to the number of times. This ratio also indicates that the closer to 1, the higher the work quality, and the farther from 1, the lower the work quality. The determination unit 114 determines that the switching condition is satisfied when the calculated ratio is equal to or greater than the threshold.
- the determination unit 114 may determine whether or not the switching condition is satisfied according to the ratio of the reached areas included in the set work route.
- the percentage of the reached area is the area where the work has actually been reached by the working machine that has traveled all of the set work paths. The smaller the ratio of the reached area, the less wasted travel and the higher the work quality.
- the determination unit 114 calculates the ratio of the reached area by detecting the load amount of the motor that rotates the blade disk provided in the self-machine.
- the amount of load on the motor is maximized when the grass is cut to the full width of the blade disc, and becomes smaller as the width of the blade disc includes more areas that have already been reached.
- the determining unit 114 creates a graph in which the travel distance and the load amount of the motor are associated with each other, and determines that the switching condition is satisfied when the area obtained by collecting the differences from the maximum value of the load amount of the motor is equal to or greater than the threshold value. to decide. It should be noted that it may be determined that the switching condition is satisfied when the distance for traveling the reached area again is accumulated and the accumulated distance reaches a threshold value.
- the determining unit 114 may reduce the accumulated distance according to the distance traveled in the unreached area.
- the determination unit 114 calculates the distance traveled in the unreached area by detecting the load amount of the motor described above. Unreached Area By doing so, it is less likely that the route setting method will be switched in the middle even though the quality of the work route is gradually improving, compared to the case where the accumulated distance is not reduced.
- the determination unit 114 may similarly determine by accumulating the time to travel the reached area again instead of the distance described above. Further, the determination unit 114 may determine that the switching condition is satisfied when the degree of complexity of the shape of the boundary detected by the boundary detection unit 111 exceeds a predetermined criterion. This is because the more complicated the shape of the boundary, the more difficult it becomes to set a route with good work quality under the rules used in the first route setting method.
- the determination unit 114 should use, as the switching condition, a condition that is satisfied in a situation where it is highly likely that it is difficult to set a good route using the first route setting method. Furthermore, the determination unit 114 preferably uses, as a switching condition, a condition that is satisfied in a situation where it is difficult to set a good route using the first route setting method but is likely to be able to set a good route using the second route setting method. By using such a switching condition, it is possible to improve the possibility of setting a route with high work quality in the work area, as in the embodiment.
- the setting unit 113 may use work end conditions different from those in the embodiment.
- the setting unit 113 may use, for example, a condition that is satisfied when a predetermined amount of time has passed or a condition that is satisfied when the battery of the working machine has decreased to a predetermined amount as the work end condition.
- the setting unit 113 may use, as the work end condition, a condition that is satisfied when the work in the work target area is almost completed.
- the work performed by the working machine is not limited to lawn mowing, and may be, for example, tillage, fertilizer spreading, or chemical spraying by an autonomously traveling agricultural robot, or an autonomously traveling vacuum cleaner robot. Cleaning by
- the first method tends to have a smaller processing load than the second method.
- the first method is a method of setting a continuous route according to a rule that the working machine repeats movement and direction change at the boundary of the working area.
- the first method includes a plurality of rules in which at least one of the initial direction of travel and the direction after direction change is different, and the work quality is the best among the plurality of rules. The one that is the way to choose and set the route.
- the work quality is represented by any one of the set length of the route, the number of directional changes in the route, and the ratio of reached areas included in the route.
- the percentage of the unreached area or the reached area is calculated by detecting the load amount of a motor provided in the working machine.
- the detection step detects an object existing in the work area
- the second method includes detecting the acquired boundary information and the detected object each time the work machine travels a predetermined distance. a method of resetting said path based on the position of said object.
- the first method and the second method are methods of setting a route based on learned models respectively learned by different machine learning techniques.
- the first method uses a machine learning method using a neural network that can be selected from a plurality of types of patterned paths
- the second method uses a machine learning method using a classifier system.
- the work performed by the work machine is lawn mowing.
- a work machine capable of autonomous travel which is configured to execute each step of the information processing system.
- control unit 12 storage unit 13: communication unit 14: sensor unit 15: running unit 16: working unit 111: boundary detection unit 112: object detection unit 113: setting unit 114: determination unit 115: running Control unit 116: work control unit
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
Description
本節では、本実施形態に係る作業機である自走式の芝刈りロボットのハードウェア構成について説明する。すなわち、本実施形態においては、作業機によって行われる作業は、芝刈りである。また、作業機は、自律走行可能な作業機であり、作業(芝刈り)に関する動作を制御するための情報処理システムの各部を備える。
制御部11は、例えば不図示の中央処理装置(Central Processing Unit:CPU)である。制御部11は、記憶部12に記憶された所定のプログラムを読み出すことによって、芝刈りロボット1に係る種々の機能を実現する。すなわち、記憶部12に記憶されているソフトウェアによる情報処理が、ハードウェアの一例である制御部11によって具体的に実現されることで、制御部11に含まれる各機能部として実行されうる。これらについては、次節においてさらに詳述する。なお、制御部11は単一であることに限定されず、機能ごとに複数の制御部11を有するように実施してもよい。またそれらの組合せであってもよい。
記憶部12は、前述の記載により定義される様々な情報を記憶する。これは、例えば、制御部11によって実行される芝刈りロボット1に係る種々のプログラム等を記憶するソリッドステートドライブ(Solid State Drive:SSD)等のストレージデバイスとして、あるいは、プログラムの演算に係る一時的に必要な情報(引数、配列等)を記憶するランダムアクセスメモリ(Random Access Memory:RAM)等のメモリとして実施されうる。記憶部12は、制御部11によって実行される芝刈りロボット1に係る種々のプログラムや変数等を記憶している。
通信部13は、芝刈りロボット1から種々の電気信号を外部の構成要素に送信可能に構成される。また、通信部13は、外部の構成要素から芝刈りロボット1への種々の電気信号を受信可能に構成される。さらに好ましくは、通信部13がネットワーク通信機能を有し、これによりインターネット等のネットワークを介して、芝刈りロボット1と外部機器との間で種々の情報を通信可能に実施してもよい。
センサ部14は、芝刈りロボット1が処理のパラメータとして用いる各種の値を測定するセンサ群を有する。センサ部14は、例えば、測位センサ、方位センサ、角速度センサ、加速度センサ、ブレードディスクのトルクセンサ、ワイヤー検出センサ及び物体検知センサ等を備える。測位センサは、GPS(Global Positioning System)等により自装置の位置を測定する。方位センサは、自装置の向きを測定する。角速度センサは、自装置が回転する際の角速度を測定する。加速度センサは、自装置の加速度を測定する。ブレードディスクのトルクセンサは、芝刈りブレードの回転に伴うモータの負荷を検出する。この検出値に基づいて、前述した制御部11がブレードディスクの推定仕事量を算出する。ワイヤー検出センサは、図2の境界B1等に埋設されたワイヤーを検出する。物体検知センサは、ミリ波等を照射し、それを反射する物体を検知する。
走行部15は、タイヤ及びモータ等を備え、自装置を走行させる。走行部15は、制御部11により制御され、直進、後進及び方向転換等を行う。
作業部16は、ブレードディスク及びモータ等を備え、制御部11により制御され、ブレードディスクを回転させて芝を刈り込む。
本節では、本実施形態の機能構成について説明する。前述の通り、記憶部12に記憶されているソフトウェアによる情報処理がハードウェアの一例である制御部11によって具体的に実現されることで、制御部11に含まれる各機能部が実行されうる。
ルール1:初期方向=方向D1、折返し方向=方向D4、折り返し後の進行方向=D3、障害物回避
ルール2:初期方向=方向D2、折返し方向=方向D3、折り返し後の進行方向=D4、障害物回避
ルール3:初期方向=方向D3、折返し方向=方向D2、折り返し後の進行方向=D1、障害物回避
ルール4:初期方向=方向D4、折返し方向=方向D1、折り返し後の進行方向=D2、障害物回避
ルール5:作業対象領域の外周を右回りに進行し、通路の入り口に到達したらその通路に向けて進行し、次の部分領域に到達したら左右いずれかに曲がってその部分領域の角まで進行。
ルール6:ルール5の「右回り」を「左回り」に置き換え
設定部113は、ルール5及びルール6のうち、通路に到達するまでの走行距離が短い方を選んで経路設定に用いる。
図8は、第2の経路設定方法により設定された経路の一例を示す図である。図8の例では、設定部113は、図7の例と同じ出発地点P11-2から、方向D3に向かって出発して境界に到達したあとに方向D2にずれて折返し、方向D1及び方向D3に往復しながら方向D2にずれていって終了地点P12-2に至る作業経路C11-2を設定している。
図10は、作業対象領域の第2の例を示す図である。図10では、長方形の部分領域A21と、部分領域A21に通路E22で接続された部分領域A23とを備える作業対象領域A20が表されている。部分領域A23には、渦巻状の壁F23が存在し、壁F23の内部領域A24が含まれている。図10には図2と同様の方向D1~D4が表されている。
芝刈りロボット1に例示される作業機に関して、以下のような態様を採用してもよい。
前記情報処理システムにおいて、前記第1の方法は、前記第2の方法よりも処理の負荷が小さくなりやすい、もの。
前記情報処理システムにおいて、前記第1の方法は、前記作業機が進行と前記作業領域の境界における方向変換とを繰り返すルールにより、連続する経路を設定する方法である、もの。
前記情報処理システムにおいて、前記第1の方法は、初期の進行方向及び方向変換後の方向の少なくとも1つが異なる複数の前記ルールを有し、当該複数のルールのうち作業品質が最も良好になるものを選んで経路を設定する方法である、もの。
前記情報処理システムにおいて、前記作業品質が、設定された前記経路の長さ、当該経路における方向変換の回数、当該経路に含まれる到達済領域の割合のいずれかで表される、もの。
前記情報処理システムにおいて、未到達領域又は到達済領域の割合は、作業機に設けられたモータの負荷量を検出することによって算出される、もの。
前記情報処理システムにおいて、検知ステップでは、前記作業領域に存在する物体を検知し、前記第2の方法は、前記作業機が所定の距離を走行するたびに、取得された前記境界情報及び検知された前記物体の位置に基づいて前記経路を設定し直す方法である、もの。
前記情報処理システムにおいて、前記第1の方法及び前記第2の方法は、異なる機械学習手法によってそれぞれ学習された学習済みモデルに基づいて、経路を設定する方法である、もの。
前記情報処理システムにおいて、前記第1の方法では、パターン化された複数種類の経路から選択可能なニューラルネットワークを利用した機械学習方法が用いられ、前記第2の方法では、クラシファイアシステムを利用した機械学習方法が用いられる、もの。
前記情報処理システムにおいて、前記作業機によって行われる作業は、芝刈りである、もの。
自律走行可能な作業機であって、前記情報処理システムの各ステップを実行するように構成される、もの。
プログラムであって、コンピュータに、前記情報処理システムの各ステップを実行させる、もの。
もちろん、この限りではない。
11 :制御部
12 :記憶部
13 :通信部
14 :センサ部
15 :走行部
16 :作業部
111 :境界検知部
112 :物体検知部
113 :設定部
114 :判定部
115 :走行制御部
116 :作業制御部
Claims (12)
- 情報処理システムであって、
取得ステップでは、自律走行可能な作業機が対象とする作業領域の境界を示す境界情報を取得し、
設定ステップでは、取得された前記境界情報に基づき前記作業領域から未到達領域をなくすための前記作業機の経路を設定し、第1の方法により設定される前記経路を用いたときの作業品質が所定の基準を満たさない場合は第2の方法により前記経路を設定する、もの。 - 請求項1に記載の情報処理システムにおいて、
前記第1の方法は、前記第2の方法よりも処理の負荷が小さくなりやすい、もの。 - 請求項2に記載の情報処理システムにおいて、
前記第1の方法は、前記作業機が進行と前記作業領域の境界における方向変換とを繰り返すルールにより、連続する経路を設定する方法である、もの。 - 請求項3に記載の情報処理システムにおいて、
前記第1の方法は、初期の進行方向及び方向変換後の方向の少なくとも1つが異なる複数の前記ルールを有し、当該複数のルールのうち作業品質が最も良好になるものを選んで経路を設定する方法である、もの。 - 請求項1~請求項4の何れか1つに記載の情報処理システムにおいて、
前記作業品質が、設定された前記経路の長さ、当該経路における方向変換の回数、当該経路に含まれる到達済領域の割合のいずれかで表される、もの。 - 請求項5に記載の情報処理システムにおいて、
未到達領域又は到達済領域の割合は、作業機に設けられたモータの負荷量を検出することによって算出される、もの。 - 請求項1~請求項6の何れか1つに記載の情報処理システムにおいて、
検知ステップでは、前記作業領域に存在する物体を検知し、
前記第2の方法は、前記作業機が所定の距離を走行するたびに、取得された前記境界情報及び検知された前記物体の位置に基づいて前記経路を設定し直す方法である、もの。 - 請求項1~請求項7の何れか1つに記載の情報処理システムにおいて、
前記第1の方法及び前記第2の方法は、異なる機械学習手法によってそれぞれ学習された学習済みモデルに基づいて、経路を設定する方法である、もの。 - 請求項8に記載の情報処理システムにおいて、
前記第1の方法では、パターン化された複数種類の経路から選択可能なニューラルネットワークを利用した機械学習方法が用いられ、前記第2の方法では、クラシファイアシステムを利用した機械学習方法が用いられる、もの。 - 請求項1~請求項9の何れか1つに記載の情報処理システムにおいて、
前記作業機によって行われる作業は、芝刈りである、もの。 - 自律走行可能な作業機であって、
請求項1~請求項10の何れか1つに記載の情報処理システムの各ステップを実行するように構成される、もの。 - プログラムであって、
コンピュータに、請求項1~請求項10の何れか1つに記載の情報処理システムの各ステップを実行させる、もの。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/020544 WO2022249486A1 (ja) | 2021-05-28 | 2021-05-28 | 情報処理システム、作業機及びプログラム |
JP2023523940A JPWO2022249486A1 (ja) | 2021-05-28 | 2021-05-28 | |
EP21943123.6A EP4350463A4 (en) | 2021-05-28 | 2021-05-28 | INFORMATION PROCESSING SYSTEM, WORKING MACHINE AND PROGRAM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/020544 WO2022249486A1 (ja) | 2021-05-28 | 2021-05-28 | 情報処理システム、作業機及びプログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022249486A1 true WO2022249486A1 (ja) | 2022-12-01 |
Family
ID=84229592
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/020544 WO2022249486A1 (ja) | 2021-05-28 | 2021-05-28 | 情報処理システム、作業機及びプログラム |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP4350463A4 (ja) |
JP (1) | JPWO2022249486A1 (ja) |
WO (1) | WO2022249486A1 (ja) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10234634A (ja) * | 1997-02-26 | 1998-09-08 | Fujitsu General Ltd | 床清掃ロボットの制御方法 |
WO2002023297A1 (fr) * | 2000-09-11 | 2002-03-21 | Kunikatsu Takase | Systeme de commande de mouvement de corps mobiles |
JP2016208950A (ja) | 2015-05-13 | 2016-12-15 | シャープ株式会社 | ガード機構及び芝刈機 |
JP2019121364A (ja) * | 2018-01-05 | 2019-07-22 | アイロボット・コーポレーション | 状況認識のためのモバイル清掃ロボット人工知能 |
WO2020012944A1 (ja) * | 2018-07-09 | 2020-01-16 | ソニー株式会社 | 制御装置、制御方法、およびプログラム |
JP2020524330A (ja) * | 2017-06-14 | 2020-08-13 | ズークス インコーポレイテッド | ボクセルベースのグランド平面推定およびオブジェクト区分化 |
US20210018927A1 (en) | 2019-07-15 | 2021-01-21 | Deere & Company | Robotic mower boundary detection system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110312418B (zh) * | 2017-11-16 | 2022-05-10 | 南京泉峰科技有限公司 | 智能割草系统 |
-
2021
- 2021-05-28 JP JP2023523940A patent/JPWO2022249486A1/ja active Pending
- 2021-05-28 EP EP21943123.6A patent/EP4350463A4/en active Pending
- 2021-05-28 WO PCT/JP2021/020544 patent/WO2022249486A1/ja active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10234634A (ja) * | 1997-02-26 | 1998-09-08 | Fujitsu General Ltd | 床清掃ロボットの制御方法 |
WO2002023297A1 (fr) * | 2000-09-11 | 2002-03-21 | Kunikatsu Takase | Systeme de commande de mouvement de corps mobiles |
JP2016208950A (ja) | 2015-05-13 | 2016-12-15 | シャープ株式会社 | ガード機構及び芝刈機 |
JP2020524330A (ja) * | 2017-06-14 | 2020-08-13 | ズークス インコーポレイテッド | ボクセルベースのグランド平面推定およびオブジェクト区分化 |
JP2019121364A (ja) * | 2018-01-05 | 2019-07-22 | アイロボット・コーポレーション | 状況認識のためのモバイル清掃ロボット人工知能 |
WO2020012944A1 (ja) * | 2018-07-09 | 2020-01-16 | ソニー株式会社 | 制御装置、制御方法、およびプログラム |
US20210018927A1 (en) | 2019-07-15 | 2021-01-21 | Deere & Company | Robotic mower boundary detection system |
Non-Patent Citations (1)
Title |
---|
See also references of EP4350463A4 |
Also Published As
Publication number | Publication date |
---|---|
EP4350463A1 (en) | 2024-04-10 |
JPWO2022249486A1 (ja) | 2022-12-01 |
EP4350463A4 (en) | 2024-04-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11175670B2 (en) | Robot-assisted processing of a surface using a robot | |
Galceran et al. | A survey on coverage path planning for robotics | |
US8306659B2 (en) | Autonomous robot | |
US8498788B2 (en) | Method and system for determining a planned path of a vehicle | |
JP7220285B2 (ja) | 経路計画 | |
CN113064408B (zh) | 自主机器人及其控制方法、计算机存储介质 | |
JP6722302B2 (ja) | 無人作業システム、サーバーコンピューター、及び無人作業機 | |
CN113766825B (zh) | 节能草坪养护车辆 | |
KR102315678B1 (ko) | 잔디 깎기 로봇 및 그 제어 방법 | |
US20220248599A1 (en) | Lawn mower robot and method for controlling the same | |
US20230384795A1 (en) | Robot localization and mapping accommodating non-unique landmarks | |
Wasif | Design and implementation of autonomous Lawn-Mower Robot controller | |
Taïx et al. | Path planning for complete coverage with agricultural machines | |
Abdulazeez et al. | Vision-based mobile robot controllers: a scientific review | |
WO2022249486A1 (ja) | 情報処理システム、作業機及びプログラム | |
CN114937258A (zh) | 割草机器人的控制方法、割草机器人以及计算机存储介质 | |
Zhang et al. | Intelligent in-orchard bin-managing system for tree fruit production | |
US11457558B1 (en) | Autonomous vehicle navigation | |
Cheng et al. | Distributed area coverage using robot flocks | |
Barak et al. | Two Dimensional Mapping by using Single Ultrasonic Sensor. | |
CN115298633A (zh) | 控制装置、控制方法以及计算机程序 | |
WO2024045357A1 (zh) | 自移动设备的控制方法、装置和自移动设备 | |
Majdi et al. | AGV path planning in unknown environment using fuzzy inference systems | |
US20220330477A1 (en) | Control device and work machine | |
Verma | Safe Navigation of a Non-Holonomic Robot with Low Computational-power in a 2D Dynamic Environment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21943123 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2023523940 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18563231 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2021943123 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2021943123 Country of ref document: EP Effective date: 20240102 |