This relates to information processing equipment, methods, and programs.

Abstract

To efficiently find a movement passage of a smooth trajectory, while securing safety of a moving body exhibiting an autonomous movement function.SOLUTION: An information processor comprises a condition acquisition part for acquiring information about a start point and an end point of a moving body inside a field, and about an impassable area in the field; a route-setting part for setting a route of an evaluation object reaching the end point from the start point via a plurality of middle points by setting the plurality of middle points inside the field; a first calculation part for calculating an index of a risk degree increased according to closeness of the route and the impassable area; a second calculation part for calculating an index indicating smoothness of a route trajectory; and a search processing part for optimizing positions of the middle points inside the field, based on the calculated index, in such a manner that a smooth route exhibiting a low risk degree is produced.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] This invention relates to an information processing device, a method, and a program. [Background technology]

[0002] When formulating an action plan to efficiently move an autonomous mobile device in an environment where the location of obstacles is known, it is conceivable to explore the device's movement path using an optimization problem.

[0003] For example, Patent Document 1 proposes a server device that provides the cost between any two points on a map based on a learning model. In the technology of Patent Document 1, the predicted cost between two nodes corresponding to points on a map is calculated based on graph information, the actual cost between the two points on the map corresponding to the two nodes is obtained, and training data for the learning model is generated from the relationship between the predicted cost and the actual cost. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Patent No. 6562431 [Overview of the project] [Problems that the invention aims to solve]

[0005] In this type of search, the more possible movement paths there are, the greater the computational power required to optimize the movement path of the moving object. Furthermore, obtaining a suitable solution becomes even more difficult when trying to make the movement path as smooth as possible while ensuring the safety of the moving object.

[0006] This invention was made against the background of the above-mentioned problems, and provides an information processing device that can efficiently determine a smooth trajectory while ensuring the safety of a mobile body with autonomous movement capabilities. [Means for solving the problem]

[0007] One aspect of the present invention provides an information processing apparatus (20) for setting a movement path when moving a moving body (10) having an autonomous movement function within a predetermined field (F). The information processing apparatus includes a condition acquisition unit (31) that acquires information on a starting point (P S ) and an end point (P E ) of the moving body within the field, and information on impassable regions (41) within the field, a path setting unit (33) that sets a plurality of intermediate points (P n ) within the field and sets an evaluation target path (42) that reaches the end point via the plurality of intermediate points from the starting point, a first calculation unit (34) that calculates an index of risk level that increases according to the proximity of the path to the impassable region, a second calculation unit (35) that calculates an index indicating the smoothness of the trajectory of the path, and a search processing unit (37) that optimizes the positions of the intermediate points within the field so that a path with a low risk level and smoothness is generated based on the calculated indices. In one embodiment described above, the path may be formed by using a starting point, an intermediate point, and an ending point as nodes, and connecting the nodes with links (43), and the second calculation unit (35) calculates the angle (θ) between the links at the intermediate point. n The first index, which corresponds to the sum of the curvatures of the curves at each midpoint, may be calculated using the above method, and a second index, which indicates the variability in the distances between nodes, may also be calculated. Furthermore, the search processing unit (37) may use an evaluation value obtained by normalizing and combining the first and second indexes to optimize the system so that the sum of the curvatures of the curves at each midpoint is small and the variability in the distances between nodes is small. Furthermore, in one embodiment described above, the system may also include a third calculation unit (36) that calculates the expected travel time of a moving object along a path based on the frequency of collision avoidance objects set in the field, and the search processing unit (37) may optimize the position of intermediate points in the field to shorten the expected travel time.

[0008] Another aspect of the present invention provides a method for setting a movement path when moving a moving body (10) having an autonomous movement function within a predetermined field (F). The method includes steps of: acquiring information on a starting point (P S ) and an end point (P E ) of the moving body within the field, and information on impassable regions (41) within the field (step S1); setting a plurality of intermediate points (P n ) within the field and setting an evaluation target path (42) that reaches the end point via the plurality of intermediate points from the starting point (step S3); calculating an index of risk level that increases according to the proximity of the path to the impassable region (step S4); calculating an index indicating the smoothness of the trajectory of the path (step S5); and optimizing the positions of the intermediate points within the field so that a path with a low risk level and smoothness is generated based on the calculated indices (steps S6 - S8). In another embodiment of the method described above, the path may be formed by having a starting point, an intermediate point, and an ending point as nodes, and connecting the nodes with links (43), and in step (S5) of calculating an index indicating the smoothness of the path's trajectory, the angle (θ) between the links at the intermediate point n A first index corresponding to the sum of the curvatures of the curves at each midpoint may be calculated using ), and a second index indicating the variation in distance between nodes may also be calculated. Furthermore, in the step of optimizing the position of the midpoints (S6-S8), an evaluation value obtained by normalizing and combining the first and second indices may be used to optimize so that the sum of the curvatures of the curves at each midpoint is small and the variation in distance between nodes is small. Furthermore, in another embodiment of the method described above, the method may further include a step (S12) of calculating the expected travel time of a moving object along a path based on the frequency of occurrence of collision avoidance objects set in the field, and in the step (S6-S8) of optimizing the position of the intermediate points, the position of the intermediate points in the field may be optimized to shorten the expected travel time.

[0009] Another aspect of the present invention provides a program for causing a computer (20) to execute a method for setting a movement path when moving a moving body (10) having an autonomous movement function within a predetermined field (F). The method includes a step (S1) of acquiring information on a starting point (P S ) and an end point (P E ) of the moving body within the field, and information on impassable regions (41) within the field, a step (S3) of setting a plurality of intermediate points (P n ) within the field and setting an evaluation target path (42) that reaches the end point via the plurality of intermediate points from the starting point, a step (S4) of calculating an index of risk that increases according to the proximity of the path to the impassable regions, a step (S5) of calculating an index indicating the smoothness of the trajectory of the path, and a step (S6 - S8) of optimizing the positions of the intermediate points within the field so that a path with low risk and smoothness is generated based on the calculated indices. In another embodiment of the program described above, the path may be formed by using a starting point, an intermediate point, and an ending point as nodes, and connecting the nodes with links (43), and in step (S5) of calculating an index indicating the smoothness of the path's trajectory, the angle (θ) between the links at the intermediate point is n A first index corresponding to the sum of the curvatures of the curves at each midpoint may be calculated using ), and a second index indicating the variation in distance between nodes may also be calculated. Furthermore, in the step of optimizing the position of the midpoints (S6-S8), an evaluation value obtained by normalizing and combining the first and second indices may be used to optimize so that the sum of the curvatures of the curves at each midpoint is small and the variation in distance between nodes is small. Furthermore, in the program of the other embodiment described above, a step (S12) may be further included in which the expected travel time of a moving object along a path is calculated based on the frequency of occurrence of collision avoidance objects set in the field, and in the step (S6-S8) of optimizing the position of the intermediate points, the position of the intermediate points in the field may be optimized to shorten the expected travel time.

Advantages of the Invention

[0010] According to the present invention, it is possible to efficiently obtain a movement path with a smooth trajectory while ensuring the safety of a moving body having an autonomous movement function.

Brief Description of the Drawings

[0011] [Figure 1] It is a diagram showing an example of the system configuration of the first embodiment. [Figure 2] It is a diagram showing an example of the software configuration of the information processing apparatus of the first embodiment. [Figure 3] It is a flowchart showing an example of the operation of the information processing apparatus of the first embodiment. [Figure 4] It is a diagram schematically showing an example of a field map. [Figure 5] This diagram schematically shows an example of a risk map. [Figure 6] This figure shows an example of the initial settings for the route to be evaluated. [Figure 7] This figure shows an example of an optimized path from the state shown in Figure 6. [Figure 8] This figure shows an example of the software configuration of the information processing device according to the second embodiment. [Figure 9] This is a flowchart showing an example of the operation of the information processing device 20 of the second embodiment. [Figure 10] This figure shows an example where areas with a high frequency of collision avoidance objects are set as the field. [Modes for carrying out the invention]

[0012] Hereinafter, embodiments of the information processing apparatus, method, and program of the present invention will be described with reference to the drawings. The configuration described below is an example (representative example) of the present invention and is not limited thereto.

[0013] <First Embodiment> In the first embodiment, an example of a system applied to a bulldozer having an autonomous movement function is described as an example of a mobile body. Figure 1 is a diagram showing an example of the system configuration of the first embodiment. The system of the first embodiment comprises a bulldozer 10 and an information processing device 20.

[0014] The dozer 10 is a construction machine-type robot used at civil engineering and construction sites, and has the function of unwinding embankments in an automated driving manner and autonomous movement. As shown in Figure 1, the dozer 10 comprises a dozer blade 11 for pushing out embankments, a drive unit 12 for driving the dozer 10, a sensor unit 13, a position detection unit 14, and a control unit 15.

[0015] Figure 1 shows a tracked vehicle in which the drive unit 12 is composed of crawlers as an example of a bulldozer 10, but the bulldozer 10 may also be a wheeled vehicle. Furthermore, the bulldozer 10 may be configured to be operated by a person if necessary. The bulldozer 10 may be configured to allow the bulldozer blade 11 to be controlled up, down, left, and right, or it may be configured to allow the bulldozer blade 11 to be controlled only up and down.

[0016] The sensor unit 13 is a sensor module for acquiring information about the surroundings of the dozer 10. For example, the sensor unit 13 may include an object recognition sensor that captures images to detect surrounding objects, or it may include a distance sensor that detects the presence or absence of objects in a predetermined direction using infrared light, ultrasound, or the like. The position detection unit 14 is a module for detecting the current position of the bulldozer 10, and is composed of, for example, a GPS (Global Positioning System).

[0017] The control unit 15 is a computer that comprehensively controls each part of the bulldozer 10 to perform automatic operation. For example, based on movement path information provided from an external source, the control unit 15 performs autonomous movement of the bulldozer 10 between multiple work areas set up within a civil engineering construction site. In addition, the control unit 15 drives the bulldozer 10 according to a predetermined algorithm within the work areas to perform the unloading of embankment soil. The control unit 15 also includes a storage unit 16 that holds information on the movement path.

[0018] The information processing device 20 is a computer for setting the movement path of the bulldozer 10, which moves autonomously between work areas. The information processing device 20 sets the movement path of the bulldozer 10 within a field where the bulldozer's movement path is designed. Information about the bulldozer 10's movement path is input from the information processing device 20 to the bulldozer 10, for example, by transfer via a storage medium or by wired or wireless communication.

[0019] Here, the movement path of the bulldozer 10 is required to allow the bulldozer 10 to travel safely without colliding with various obstacles (Condition 1). Also, if the movement path of the bulldozer 10 is a complex trajectory, the ground after leveling will become rough due to the turning of the bulldozer 10. Therefore, it is preferable that the movement path of the bulldozer 10 satisfies the above Condition 1 and also has a smooth trajectory with few turns (Condition 2).

[0020] As shown in Figure 1, the information processing device 20 is a computer equipped with a CPU (Central Processing Unit) 21, ROM (Read Only Memory) 22, RAM (Random Access Memory) 23, storage device 24, input I / F 25, and display I / F 26. Each element of the information processing device 20 is connected to one another via a bus 27.

[0021] The CPU 21 is a processor that performs various arithmetic operations according to programs stored in the ROM 22 or RAM 23. The ROM 22 is a non-volatile memory area where programs such as the BIOS are stored. The RAM 23 is a volatile memory area used as temporary memory when the CPU 21 performs various arithmetic operations.

[0022] The storage device 24 is a non-volatile, high-capacity storage medium, such as a hard disk drive (HDD) or a solid-state drive (SSD). The storage device 24 stores the operating system (OS), programs that implement various functions of the information processing device 20, and data used by said programs.

[0023] Input I / F 25 is an interface that connects to an external input device 28 and accepts input from the input device 28. Examples of input devices 28 include keyboards and pointing devices.

[0024] The display I / F 26 is an interface that connects to an external display device 29 and is responsible for outputting to the display device 29. The display device 29 can be, for example, a monitor device such as a liquid crystal display.

[0025] In the information processing device 20, after startup, the CPU 21 executes the BIOS, and the OS is loaded from the storage device 24 into the RAM 23 in an executable state. The CPU 21 loads the software modules of the information processing device 20 from the storage device 24 into the RAM 23 in an executable state as the OS operates. The loaded software modules then operate when the CPU 21 executes a program.

[0026] (Software configuration of information processing equipment) Figure 2 shows an example of the software configuration of the information processing device 20 of the first embodiment. The information processing device 20 includes, as a software module, a condition acquisition unit 31, a map generation unit 32, a route setting unit 33, a first calculation unit 34, a second calculation unit 35, a search processing unit 37, and an output processing unit 38.

[0027] The condition acquisition unit 31 acquires a field F for designing the movement path of the bulldozer 10 and the starting point P of the movement path of the bulldozer 10. S and the terminus P E The system then receives input from the operator for initial conditions such as impassable areas 41 within field F. Based on this, the condition acquisition unit 31 obtains a field map 40 that shows the specifications of field F.

[0028] The map generation unit 32 generates a hazard map M that indicates the degree of danger within field F, based on the arrangement of impassable areas 41 within field F.

[0029] The route setting unit 33 performs the process of setting the route 42 to be evaluated within field F. The first calculation unit 34 calculates an index indicating the degree of risk of the route 42. The second calculation unit 35 calculates an index indicating the smoothness of the trajectory of the path 42.

[0030] The search processing unit 37 optimizes the path 42 based on the above indicators. The output processing unit 38 controls the display device 29 to display the candidate movement paths optimized by the search processing unit 37.

[0031] (Example of operation of an information processing device) Figure 3 is a flowchart showing an example of the operation of the information processing device 20 of the first embodiment. The processing in Figure 3 is started in response to a start command from the operator via the input device 28.

[0032] In S1, the condition acquisition unit 31 acquires a field map 40 that shows the specifications of field F. As an example, the condition acquisition unit 31 in S1 generates the field map 40 by receiving the following initial conditions via the input device 28. • Specify the shape and dimensions of Field F • Starting point P of route 42 S Location • Starting point P S Specify the direction in which the dozer 10 can move. • End point P of route 42 E Location • End point P E Specify the direction in which bulldozer 10 can enter. • Specify the number, location, and dimensions of the impassable areas 41.

[0033] Figure 4 is a schematic diagram showing an example of a field map 40. Field F is set up to correspond to a civil engineering and construction site where the bulldozer 10 operates autonomously, and in this embodiment, for simplicity, it is represented as a two-dimensional shape in an xy coordinate system. Note that the various initial conditions in S1 may be input to the information processing device 20 in, for example, a GUI format.

[0034] The field map 40 shown in Figure 4 includes the starting point P. S , final stop P E The location and three impassable areas 41 are defined. Also, the starting point P S The direction in which the bulldozer 10 can move is set to the left in the diagram, and the endpoint P EThe direction in which the bulldozer 10 can enter is set to upward in the diagram.

[0035] Furthermore, impassable areas 41 include, for example, fixed obstacles such as walls and pipes present in a civil engineering or construction site, as well as holes, ditches, and steps that are too large for the dozer 10 to pass through. In the field map 40, the impassable areas 41 for the dozer 10 and passable areas that do not fall under the impassable areas 41 are shown, for example, by a binary distribution of 0 and 1. For example, on the field map 40, the coordinates of passable areas are associated with a setting value indicating passability ("0" value), and the coordinates of impassable areas 41 are associated with a setting value indicating impassability ("1" value).

[0036] In S2, the map generation unit 32 generates a hazard map M that indicates the degree of danger within field F, based on the arrangement of impassable areas 41 within field F. The hazard map M is a multi-level map that shows the degree of danger for each coordinate in field F, based on its proximity to the impassable areas 41.

[0037] Figure 5 schematically shows an example of a hazard map M. As an example, the map generation unit 32 of S2 generates the hazard map M by multiplying the field map 40 by a smoothing filter (e.g., a Gaussian filter or an averaging filter). The set value of each coordinate in the hazard map M is smoothed by the set values ​​of the surrounding coordinates and converted into a state where it takes a multi-gradient value between "0" and "1". The set value of each coordinate in the hazard map M approaches "1" the closer it is to the impassable area 41 and approaches "0" the further it is from the impassable area 41. Note that in Figure 5, the smoothed impassable area 41a is shown as a gradient.

[0038] In S3, the route setting unit 33 performs the process of setting the route 42 to be evaluated within field F. At this time, the route setting unit 33 sets multiple intermediate points P within field F. n (wherein n is an integer greater than 1) is set appropriately. Then, the route setting unit 33 sets the starting point P S From multiple intermediate points P n The route goes through these points in sequence to the final destination P. EA path 42 leading to is set. In this embodiment, path 42 is set by connecting nodes (start point, intermediate point, end point) in order with straight links 43. Also, intermediate point P n The number of intermediate points and the order in which they are passed are predetermined by the operator. Figure 6 shows an example of the initial settings for the route 42 to be evaluated.

[0039] In S4, the first calculation unit 34 calculates an index indicating the risk level of the route 42 to be evaluated using the following method. First, the first calculation unit 34 identifies the coordinates through which the route 42 passes on field F and obtains the set values ​​for each coordinate on the route 42 from the risk map M. Then, the first calculation unit 34 calculates the sum of the set values ​​for the route 42 and the maximum value of the set values ​​on the route 42 as the risk index.

[0040] Here, the coordinate settings are high in or near the impassable area 41. Therefore, in the route 42 with a large sum of settings, events such as the bulldozer 10 colliding with an obstacle in the impassable area 41 or the bulldozer 10 traveling near an obstacle occur, compared to the route 42 with a small sum of settings, making it difficult to safely operate the bulldozer 10 autonomously. Furthermore, if the maximum value of the set value exceeds the threshold, it can be determined that there is a high probability that the dozer 10 will collide with or come very close to an obstacle in the impassable area 41 somewhere along the path 42, resulting in a dangerous situation.

[0041] Therefore, the information processing device 20 can select a route 42 with a low risk during travel by selecting a route 42 with a low sum of the above-mentioned setting values ​​or a route 42 with a low maximum value of the setting values.

[0042] In S5, the second calculation unit 35 calculates two types of indices that indicate the smoothness of the trajectory of the path 42. The first index is an index that indicates the size of the curve on the path 42, and the second index is an index that indicates the variation in the distances between nodes on the path 42.

[0043] The second calculation unit 35 calculates the first index by the following method. First, the second calculation unit 35 calculates the angle between the links 43 at each midpoint of the path 42. Here, any midpoint P connecting the two links 43 n The angle between links is θ n This is how it is written.

[0044] Next, the second calculation unit 35 calculates the midpoint P n Curve size Cv n The value of Cv in equation (1) is calculated using the following formula (1). n is, θ n The angle is close to 180 degrees, and the midpoint P n The value becomes smaller when the curvature of the curve is small (i.e., the curve formed by the two links 43 is gentle), and θ n The angle deviates from 180 degrees, and the midpoint P n The value increases as the curvature of the curve increases. Cv n =|180°-θ n | …(1)

[0045] Then, the second calculation unit 35 calculates the above Cv at all n midpoints. n The second calculation unit 35 then calculates the value of Cv for all midpoints. n The sum of the nth powers of is obtained as the first index. Here, the first indicator will show a larger value the more curves with large curvatures there are on the path 42. Therefore, by selecting a path 42 with a small value for the first indicator, the information processing device 20 can select a path 42 that results in a smoother trajectory with less turning for the dozer 10. In addition, the second calculation unit 35 calculates the link angle θ n By using this method to obtain an index representing the curvature of path 42 through a simple calculation, the computational load can be reduced.

[0046] Furthermore, the second calculation unit 35 calculates the second index by the following method. First, the second calculation unit 35 calculates the length of each link 43 connecting the nodes of the path 42. Then, as a second indicator, the second calculation unit 35 calculates the standard deviation of the link lengths 43.

[0047] Generally, paths 42 with large variations in distance between nodes are often poorly optimized and inefficient, and tend to unnecessarily form curves with large curvatures. Therefore, the information processing device 20 can select a path with less variation in distance between nodes using the second indicator, thereby enabling the dozer 10 to select a path with less turning and a smoother trajectory.

[0048] Here, in the search for path 42, the intermediate point P n Increasing the number of intermediate points makes the path trajectory smoother, but the computational cost increases significantly. To reduce the computational cost of searching for path 42, intermediate point P n It is also conceivable to reduce the number of nodes and use curve interpolation between them to smooth the actual trajectory of the bulldozer 10. However, when generating the actual trajectory of the bulldozer 10 by curve interpolation between nodes in this way, if the distance between nodes is large, the curve-interpolated actual trajectory is more likely to deviate from the link's trajectory, which could cause the bulldozer 10 to unintentionally approach an obstacle in an impassable area. Therefore, when selecting nodes suitable for curve interpolation, the second indicator becomes more effective.

[0049] In S6, the search processing unit 37 determines whether the safety of path 42 is ensured. In S6, the search processing unit 37 determines that the safety of path 42 is ensured when the maximum value of the set values ​​on path 42 is less than a predetermined first threshold and there are no collisions with obstacles, and the sum of the set values ​​on path 42 is less than a predetermined second threshold and a safety margin from obstacles is ensured.

[0050] If the security of route 42 is guaranteed (Yes), the process proceeds to S7. On the other hand, if the security of route 42 is not guaranteed (No), the process proceeds to S8.

[0051] In S7, the search processing unit 37 determines whether the current path 42 satisfies the optimization termination conditions.

[0052] For example, in S7, the search processing unit 37 calculates an evaluation value for the path 42 by normalizing the first and second indicators calculated in S5 and performing a linear combination. The evaluation value is smaller the closer the path 42 is to a smooth trajectory. The search processing unit 37 then determines that the optimization termination condition is met when the evaluation value is smaller than a predetermined termination condition value. However, in the first processing, the search processing unit 37 determines that the termination condition is not met.

[0053] If the optimization termination condition is not met (No), the process proceeds to S8. On the other hand, if the optimization termination condition is met (Yes), the process proceeds to S9.

[0054] In S8, the search processing unit 37 uses an algorithm such as the Monte Carlo method to find an intermediate point P. n The position is reset. After S8, the process returns to S4 and the above process is repeated. Through the loop processing in S8, the information processing device 20 searches for a path that has a lower evaluation value than the current path 42 and ensures safety, and the travel path is optimized.

[0055] Figure 7 shows an example of an optimized route 42a from the state in Figure 6. In Figure 6, the second and third intermediate points P2 and P3 from the starting point overlapped with the smoothed impassable area 41a, but in Figure 7, intermediate points P1-P4 are set to avoid the impassable area 41a.

[0056] In S9, the search processing unit 37 saves the current path 42 as a candidate for the travel path. In S10, the search processing unit 37 determines whether the number of saved candidate movement paths is greater than or equal to a predetermined number greater than 1. If the number of candidate movement paths is greater than or equal to the predetermined number (Yes), the process proceeds to S11. On the other hand, if the number of candidate movement paths is less than the predetermined number (No), the process returns to S3, and a new candidate movement path is searched for.

[0057] In S11, the search processing unit 37 determines the optimal movement path from a predetermined number of candidate movement paths using a predetermined algorithm. When optimizing the path 42 using methods such as Monte Carlo, local optima are often obtained. Therefore, by selecting the optimal movement path from multiple candidates, it becomes possible to appropriately set the movement path of the bulldozer 10. In addition, in S11, the output processing unit 38 may display multiple candidate travel paths on the display device 29, and the search processing unit 37 may determine the optimal travel path based on the path selection instruction received from the operator.

[0058] This concludes the explanation of Figure 3. The determined movement path information is then read from the information processing device and input to the bulldozer 10. Note that the actual trajectory of the movement path may be interpolated by the bulldozer 10.

[0059] As described above, in the first embodiment, the condition acquisition unit 31 determines the starting point P of the dozer 10 within the field F. S and the terminus P E The system acquires information on impassable areas 41 within field F (S1). The route setting unit 33 sets multiple intermediate points P within field F. n Set the starting point P S From multiple intermediate points P n Via the final stop P E The evaluation target path 42 is set (S3). The first calculation unit 34 calculates a risk index that increases according to the proximity of path 42 to the impassable area 41 (S4). The second calculation unit 35 calculates a first index and a second index that indicate the smoothness of the trajectory of path 42 (S5). Then, the search processing unit 37, based on the calculated index, sets an intermediate point P in field F so that a low-risk and smooth path is generated. n Optimize the position (S6-S8).

[0060] According to the first embodiment, as described above, the midpoint of the path 42 is optimized based on an index of the degree of danger of the path 42 and the smoothness of the trajectory, so that a smooth trajectory can be efficiently determined while ensuring the safety of the bulldozer 10.

[0061] <Second Embodiment> The bulldozer 10 has a function that, when it detects an object to be avoided, such as a person or a vehicle (such as a dump truck transporting soil), stops (or slows down) until the object to be avoided moves out of the path 42 in order to ensure safety. In the example of the second embodiment, considering the case in which the object to be avoided appears in the path 42, the path 42 is further optimized based on the expected travel time of the bulldozer 10.

[0062] The second embodiment is a modified version of the first embodiment, and the hardware configuration in the second embodiment is the same as in the first embodiment. Therefore, in the description of the second embodiment, the same reference numerals are used for components that are the same as in the first embodiment, and redundant explanations are omitted.

[0063] Figure 8 shows an example of the software configuration of the information processing device 20 of the second embodiment. The information processing device 20 further includes a software module for the third calculation unit 36 ​​in addition to the software modules of the first embodiment. The third calculation unit 36 ​​calculates the expected travel time of the dozer 10 along the path 42 based on the frequency of collision avoidance objects set in the field F.

[0064] Figure 9 is a flowchart showing an example of the operation of the information processing device 20 of the second embodiment. The operation example in Figure 9 differs from Figure 3 in that the process of S12 is performed between the processes of S5 and S6. The processes of S1 and S7 also differ as follows.

[0065] In S1 of Figure 9, the condition acquisition unit 31 accepts inputs to set the frequency of collision avoidance objects appearing on the field F, in addition to the initial conditions of the first embodiment. For example, the operator can set any area of ​​the field F as an area with a high frequency of collision avoidance objects. For example, areas that overlap with vehicle paths at construction sites or areas where pedestrian crossings are set are set on the field map 40 as areas with a high frequency of collision avoidance objects. Figure 10 shows an example where area 44, where collision avoidance objects frequently appear, is set as field F.

[0066] In Figure 9, the processes from S2 to S5 are the same as in the first embodiment. After the process in S5, in S12, the third calculation unit 36 ​​calculates the expected travel time of the dozer 10 along the path 42. In S12, the third calculation unit 36 ​​calculates the expected travel time of the path 42 from the speed of the dozer 10 and the total length of the path 42. However, when calculating the travel time in region 44, the travel time is calculated as follows, taking into account the appearance of collision avoidance objects.

[0067] In region 44, collision avoidance objects appear at a relatively higher frequency than in other regions, causing the dozer 10 to stop. Therefore, when the path 42 passes through region 44, the third calculation unit 36 ​​adds the expected stopping time of the dozer 10, based on the frequency of collision avoidance object appearances in region 44, to the expected travel time. After that, the process proceeds to S6. The process of S6 is the same as in the first embodiment.

[0068] In S7 of Figure 9, the search processing unit 37 normalizes the first and second indicators calculated in S5 and the predicted travel time calculated in S12, and performs a linear combination to calculate the evaluation value of the path 42. Then, based on the calculated evaluation value, the search processing unit 37 determines whether the path 42 satisfies the optimization termination conditions.

[0069] If the optimization termination condition is not met (No), the process proceeds to S8. On the other hand, if the optimization termination condition is met (Yes), the process proceeds to S9. The process from S8 onwards in Figure 9 is the same as in the first embodiment.

[0070] As described above, in the second embodiment, the decision to terminate the optimization is made by considering the expected travel time of the path. Then, in the loop S8 in Figure 9, the intermediate point P in field F is shortened in order to reduce the expected travel time. n The position is optimized.

[0071] As a result, in the second embodiment, as shown in Figure 10, the path 42b that bypasses region 44 via intermediate point P3' (i.e., the path with a lower frequency of collision avoidance objects) is more likely to be selected than the path 42a that passes through region 44 via intermediate point P3 and has a high frequency of collision avoidance objects.

[0072] <Modified examples of embodiments> A program that implements one or more of the above embodiments may be supplied to the device via a network. Furthermore, the configuration of the above embodiments may be implemented using hardware circuits (e.g., ASICs) that implement one or more of the above functions.

[0073] Furthermore, although the above embodiment described an example of an information processing device 20 for designing the movement path of a bulldozer 10 having an autonomous movement function, the information processing device of the present invention may also be used to design the movement path of other mobile bodies other than bulldozers 10. For example, examples of mobile bodies having an autonomous movement function include construction machinery other than bulldozers, vehicles, transport robots that transport workpieces, drones, and the like.

[0074] Although preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of its essence. [Explanation of Symbols]

[0075] 10...Dozer, 20...Information processing device, 21...CPU, 24...Storage device, 29...Display device, 31...Condition acquisition unit, 32...Map generation unit, 33...Route setting unit, 34...First calculation unit, 35...Second calculation unit, 36...Third calculation unit, 37...Search processing unit, 38...Output processing unit, 41...Impassable area, 42...Route, 43...Link, F...Field, M...Risk level map, P S ...Starting point, P E ...Terminal stop, P n ...midpoint

Claims

1. In an information processing device (20) that sets a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), The starting point (P) of the moving body within the field Ｓ ) and the terminal (P Ｅ ), a condition acquisition unit (31) that acquires information on impassable areas (41) within the field, Multiple intermediate points (P) within the aforementioned field ｎ A route setting unit (33) sets a route (42) to be evaluated, which goes from the starting point through a plurality of intermediate points to the endpoint, A first calculation unit (34) calculates an index of risk that increases according to the proximity of the aforementioned path to the impassable area, A second calculation unit (35) calculates an index indicating the smoothness of the trajectory of the aforementioned path, The system includes a search processing unit (37) that optimizes the position of the intermediate points within the field so as to generate a low-risk and smooth path based on the calculated indicators, The aforementioned path is formed by using the starting point, the intermediate point, and the ending point as nodes, and connecting the nodes with links (43). The second calculation unit (35) calculates a first index corresponding to the sum of the curvatures of the curves at each intermediate point using the angle (θ n) between the links at the intermediate point, and also calculates a second index indicating the variation in the distance between the nodes. The search processing unit (37) uses an evaluation value obtained by normalizing and combining the first and second indicators to optimize the system so that the sum of the curvatures of the curves at each intermediate point is small and the variation in the distances between nodes is small. Information processing device.

2. An information processing device (20) for setting a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), A condition acquisition unit (31) acquires information on the starting point (PS) and ending point (PE) of the moving object within the field, and on the impassable area (41) within the field, A route setting unit (33) sets up a route (42) to be evaluated, which is set up within the field by setting up multiple intermediate points (P n) and going from the starting point to the endpoint via the multiple intermediate points, A first calculation unit (34) calculates an index of risk that increases according to the proximity of the aforementioned path to the impassable area, A second calculation unit (35) calculates an index indicating the smoothness of the trajectory of the aforementioned path, The system includes a search processing unit (37) that optimizes the position of the intermediate points within the field so as to generate a low-risk and smooth path based on the calculated indicators, The system further includes a third calculation unit (36) that calculates the expected travel time of the moving body along the path based on the frequency of collision avoidance objects set in the field, The search processing unit (37) optimizes the position of the intermediate point within the field to shorten the predicted travel time. Information processing device.

3. The first calculation unit (34) calculates at least one of the sum of the risk levels along the route and the maximum value of the risk level along the route as the risk level index. The information processing apparatus according to claim 1 or claim 2.

4. The system further comprises a map generation unit (32) that generates a risk map of the field by smoothing the risk level set at each position within the field with the risk level set at the surrounding positions. The first calculation unit (34) calculates the risk index using the risk map. The information processing apparatus according to any one of claims 1 to 3.

5. The system further includes an output processing unit (38) that causes the search processing unit (37) to display a plurality of candidate movement paths on a display device (29). The information processing apparatus according to any one of claims 1 to 4.

6. The aforementioned mobile unit is a construction machine having an autonomous movement function. The information processing apparatus according to any one of claims 1 to 5.

7. A method for setting a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), The starting point (P) of the moving body within the field Ｓ ) and the terminal (P Ｅ ), the step (S1) of obtaining information on the impassable area (41) within the field, Multiple intermediate points (P) within the aforementioned field ｎ Step (S3) sets a path (42) to be evaluated, which goes from the starting point through a plurality of intermediate points to the endpoint, Step (S4) of calculating an index of risk that increases according to the proximity of the aforementioned route to the impassable area, The steps include: (S5) calculating an index that indicates the smoothness of the trajectory of the aforementioned path; The process includes the steps (S6-S8) of optimizing the position of the intermediate point in the field so that a low-risk, smooth path is generated based on the calculated indicators, The aforementioned path is formed by using the starting point, the intermediate point, and the ending point as nodes, and connecting the nodes with links (43). In step (S5), which involves calculating an index indicating the smoothness of the trajectory of the aforementioned path, a first index corresponding to the sum of the curvatures of the curves at each intermediate point is calculated using the angle (θ n) between the links at the intermediate point, and a second index indicating the variation in distance between the nodes is calculated. In the step of optimizing the position of the intermediate points (S6-S8), the evaluation value obtained by normalizing and combining the first and second indicators is used to optimize so that the sum of the curvatures of the curves for each intermediate point is small and the variation in the distances between the nodes is small. method.

8. A method for setting a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), Step (S1) is to acquire information on the starting point (PS) and ending point (PE) of the moving object within the field, and on the impassable area (41) within the field. Step (S3) is to set up a route (42) to be evaluated, which goes from the starting point to the endpoint via the multiple intermediate points, by setting up a plurality of intermediate points (P n) within the field, Step (S4) of calculating an index of risk that increases according to the proximity of the aforementioned route to the impassable area, The steps include: (S5) calculating an index that indicates the smoothness of the trajectory of the aforementioned path; The process includes the steps (S6-S8) of optimizing the position of the intermediate point in the field so that a low-risk, smooth path is generated based on the calculated indicators, The process further includes the step (S12) of calculating the expected travel time of the moving body along the path based on the frequency of occurrence of collision avoidance objects set in the field, In the step of optimizing the position of the intermediate point (S6-S8), the position of the intermediate point within the field is optimized to shorten the expected travel time. method.

9. A program that causes a computer (20) to execute a method for setting a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), wherein the method is: The starting point (P) of the moving body within the field Ｓ ) and the terminal (P Ｅ ), the step (S1) of obtaining information on the impassable area (41) within the field, Set a plurality of intermediate points (P ｎ ) within the field, and set a path (42) of an evaluation target that reaches the end point via the plurality of intermediate points from the start point (step S3); Step (S4) of calculating an index of risk that increases according to the proximity of the aforementioned route to the impassable area, The steps include: (S5) calculating an index that indicates the smoothness of the trajectory of the aforementioned path; The process includes the steps (S6-S8) of optimizing the position of the intermediate point in the field so that a low-risk, smooth path is generated based on the calculated indicators, The aforementioned path is formed by using the starting point, the intermediate point, and the ending point as nodes, and connecting the nodes with links (43). In step (S5), which involves calculating an index indicating the smoothness of the trajectory of the aforementioned path, a first index corresponding to the sum of the curvatures of the curves at each intermediate point is calculated using the angle (θ n) between the links at the intermediate point, and a second index indicating the variation in distance between the nodes is calculated. In the step of optimizing the position of the intermediate points (S6-S8), the evaluation value obtained by normalizing and combining the first and second indicators is used to optimize so that the sum of the curvatures of the curves for each intermediate point is small and the variation in the distances between the nodes is small. program.

10. A program that causes a computer (20) to execute a method for setting a movement path when moving a mobile body (10) having an autonomous movement function within a predetermined field (F), wherein the method is: Step (S1) is to acquire information on the starting point (PS) and ending point (PE) of the moving object within the field, and on the impassable area (41) within the field. Step (S3) is to set up a route (42) to be evaluated, which goes from the starting point to the endpoint via the multiple intermediate points, by setting up a plurality of intermediate points (P n) within the field, Step (S4) of calculating an index of risk that increases according to the proximity of the aforementioned route to the impassable area, The steps include: (S5) calculating an index that indicates the smoothness of the trajectory of the aforementioned path; The process includes the steps (S6-S8) of optimizing the position of the intermediate point in the field so that a low-risk, smooth path is generated based on the calculated indicators, The process further includes the step (S12) of calculating the expected travel time of the moving body along the path based on the frequency of occurrence of collision avoidance objects set in the field, In the step of optimizing the position of the intermediate point (S6-S8), the position of the intermediate point within the field is optimized to shorten the expected travel time. program.

Images