WO2022219875A1 - Information processing system, information processing device, and information processing method - Google Patents
Information processing system, information processing device, and information processing method Download PDFInfo
- Publication number
- WO2022219875A1 WO2022219875A1 PCT/JP2022/002935 JP2022002935W WO2022219875A1 WO 2022219875 A1 WO2022219875 A1 WO 2022219875A1 JP 2022002935 W JP2022002935 W JP 2022002935W WO 2022219875 A1 WO2022219875 A1 WO 2022219875A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- map
- data
- information processing
- unit
- processing system
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 46
- 238000003672 processing method Methods 0.000 title claims description 8
- 230000007613 environmental effect Effects 0.000 claims abstract description 56
- 230000009471 action Effects 0.000 claims description 20
- 230000010391 action planning Effects 0.000 claims description 16
- 238000010586 diagram Methods 0.000 description 9
- 230000004048 modification Effects 0.000 description 9
- 238000012986 modification Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 238000000034 method Methods 0.000 description 2
- 101100400452 Caenorhabditis elegans map-2 gene Proteins 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000010187 selection method 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
- G05D1/02—Control of position or course in two dimensions
Definitions
- the present disclosure relates to an information processing system, an information processing device, and an information processing method.
- Patent Document 1 technologies related to mobile objects such as robots that recognize the external environment and move autonomously according to the recognized environment.
- a mobile body is equipped with various sensors to recognize the external environment, and an environmental map corresponding to the external environment is constructed based on the sensor data obtained from the various sensors. Real-time processing of environmental maps is required for autonomous mobile objects. Therefore, it is desirable to provide an information processing system, an information processing apparatus, and an information processing method that can reduce the amount of data of an environment map.
- An information processing system includes a map generation unit and a setting unit that sets the data format of map data generated by the map generation unit according to estimated environmental characteristics.
- An information processing apparatus includes a setting unit that sets the data format of map data generated by the map generation unit according to estimated environmental characteristics.
- An information processing method includes setting a data format of map data generated by a map generation unit according to estimated environmental characteristics.
- the data format of the map data generated by the map generation unit is set according to the estimated environmental characteristics. As a result, for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics.
- FIG. 1 is a diagram illustrating an example of functional blocks of an information processing system according to an embodiment of the present disclosure
- FIG. It is a figure showing the example which switches map data alternatively.
- FIG. 10 is a diagram showing an example of temporarily using a plurality of map data when switching map data;
- FIG. 4 is a diagram summarizing compatibility between map data formats and aircraft types;
- FIG. 3 is a diagram summarizing compatibility between a map data format and an operation location;
- 2 is a diagram showing an example of a procedure for selecting a format of map data in the information processing system of FIG. 1;
- FIG. FIG. 3 is a diagram showing a modified example of the functional blocks of the information processing system of FIG. 1;
- FIG. 1 and FIG. 5 are diagrams showing a modified example of the schematic configuration of the parameter generator.
- FIG. 1 shows an example of functional blocks of an information processing system 1.
- the information processing system 1 includes, for example, a sensor device section 10, a parameter generation section 20, a storage section 30, an action planning section 40, a control section 50, and an actuator 60, as shown in FIG.
- the sensor device unit 10 has, for example, a sensor element 11 that recognizes the external environment and acquires environmental data 11A corresponding to the recognized external environment.
- the sensor element 11 outputs the acquired environmental data 11A to the parameter generator 20 .
- the sensor element 11 is, for example, an RGB camera, an RGB-D camera, a depth sensor, an infrared sensor, an event camera, or a sound collection sensor.
- the RGB camera is, for example, a single visible light image sensor that outputs RGB image data obtained by receiving visible light and converting it into an electrical signal.
- the RGB-D camera is, for example, a binocular visible light image sensor, and outputs RGB-D image data (RGB image data and distance image data obtained from parallax).
- the depth sensor is, for example, a ToF (Time of Flight) sensor or a Lider (Laser Imaging Detection and Ranging), and outputs range image data obtained by measuring scattered light from pulsed laser irradiation.
- the infrared sensor outputs infrared image data obtained by, for example, receiving infrared rays and converting them into electric signals.
- the event camera is, for example, a single visible light image sensor, and outputs a difference between RGB image data (difference image data) between frames.
- the sound collection sensor outputs, for example, sound data obtained from the external environment.
- the sensor device unit 10 outputs, for example, various data (eg, RGB image data, RGB-D image data, distance image data, infrared image data, differential image data, or audio data) obtained from the external environment as environment data 11A. do.
- the parameter generating unit 20 has an environmental characteristic estimating unit 21 , a map selecting unit 22 , a plurality of map generating units 23 and 24 and a map IF providing unit 25 .
- the environmental characteristic estimation unit 21 estimates the environmental characteristic 21A based on the environmental data 11A obtained from the sensor device unit 10.
- the environmental characteristic 21A is an estimated value of the environmental characteristic generated based on the environmental data 11A, and includes, for example, any one of shape feature amount, image feature amount, and sound feature amount.
- the shape feature amount indicates, for example, the flatness, the dispersion of the point group, or the average height of the point group in the environment data 11A.
- the image feature amount indicates, for example, the number of edges, the number of feature points, or the average or variance of hue, brightness, and saturation in the environmental data 11A.
- the audio feature amount indicates volume, for example.
- the map selection unit 22 sets the data format of the map data generated by the map generation units 23 and 24 according to the estimated environmental characteristics 21A.
- the map selection unit 22 selects at least one of the map generation units 23 and 24 as the map generation unit to be used, for example, according to the estimated environmental characteristics 21A.
- the map selection unit 22 instructs the selected map generation unit to generate a map.
- the map selection unit 22, for example, provides an array of identifiers ID (identification) of the map generation units to be used, or an array indicating the use of the plurality of map generation units 23 and 24, to the plurality of map generation units 23 and 24. may be output.
- the map selection unit 22 may output a command to instruct map generation or a command to end map generation only to the map generation unit to be used.
- the map generation unit 23 uses the environment data 11A to construct the first map data 23A in the first data format.
- the map generation unit 23 generates the environment map 31 at the current time by superimposing the constructed first map data 23A on the environment map 31 read from the storage unit 30 .
- the map generation unit 23 stores the generated environment map 31 at the current time in the storage unit 30 .
- the map generation unit 23 outputs all or part of the generated environment map 31 at the current time to the map IF provision unit 25 .
- the map generation unit 24 uses the environment data 11A to construct second map data 24A in a second data format different from the first data format.
- the map generation unit 24 generates the environment map 32 at the current time by superimposing the constructed second map data 24A on the environment map 32 read from the storage unit 30 .
- the map generation unit 24 stores the generated environment map 32 at the current time in the storage unit 30 .
- the map generation unit 24 outputs all or part of the generated environment map 32 at the current time to the map IF provision unit 25 .
- the map generating unit 23 may output the first map data 23A to the map IF providing unit 25 when selected by the map selecting unit 22 . Further, the map generating unit 24 may output the second map data 24A to the map IF providing unit 25 when selected by the map selecting unit 22 . In these cases, the storage unit 30 may be omitted.
- the map generation unit 23 may delete past map data that has passed a certain period of time from the environment map 31 .
- the map generation unit 24 may delete past map data after a certain period of time has passed from the environment map 32 .
- the data format of the first map data 23A used in the map generator 23 is "map data format A”
- the data format of the second map data 24A used in the map generator 24 is " It conceptually shows the switching of the map data when the map data format is "B”.
- arrows show how a moving body equipped with the sensor device unit 10 moves from an area suitable for "map data format A” to an area suitable for "map data format B.” .
- the map selection unit 22 switches from the map generation unit 23 to the map generation unit 24 according to the estimated environmental characteristics 21A.
- the map selection unit 22 may immediately switch the data format of the map data from "map data format A” to "map data format B", for example, as shown in FIG.
- the map selection unit 22 selects "map data format A” and "map data format B” simultaneously.
- the data format of the map data may be switched from “map data format A” to "map data format B" after a period of use.
- the map IF providing unit 25 uses the environmental map (hereinafter referred to as "environmental map 25A") input from at least one of the map generating units 23 and 24 to generate parameters 25B used in the action plan.
- the map IF providing unit 25 for example, uses the environment map 25A and the information about the self-position and self-velocity to derive an information map regarding the presence or absence of a collision or the probability of collision, and uses the derived information map as a parameter 25B to create an action plan. You may output to the part 40.
- the map IF providing unit 25 uses, for example, the environment map 25A and information about a certain spatial area (for example, a predetermined spatial area including the self-position) to derive an information map about the obstacle density, and derives the information map.
- the resulting information map may be output to the action planning section 40 as the parameter 25B.
- the map IF providing unit 25, for example, generates an obstacle density gradient based on the generated information map about the obstacle density, and outputs information about the generated obstacle density gradient to the action planning unit 40 as a parameter 25B.
- the map IF providing unit 25 generates information about the distance from a certain position to obstacles in its surroundings and the direction of obstacles from a certain position, for example, based on the generated information map about the density of obstacles. , the generated information may be output to the action planning unit 40 as the parameter 25B.
- the storage unit 30 is a database containing environment maps 31 and 32, for example.
- the storage unit 30 is configured by, for example, a volatile memory such as a DRAM (Dynamic Random Access Memory), or a nonvolatile memory such as an EEPROM (Electrically Erasable Programmable Read-Only Memory) or flash memory.
- a volatile memory such as a DRAM (Dynamic Random Access Memory)
- a nonvolatile memory such as an EEPROM (Electrically Erasable Programmable Read-Only Memory) or flash memory.
- the action planning unit 40 creates an action plan based on the parameters 25B input from the map IF providing unit 25. For example, based on the parameter 25B, the action planning unit 40 determines what route, direction, and posture to move from the self position to the target position, and controls the result of the determination as the action plan 40A. Output to unit 50 .
- the control unit 50 generates a drive signal 50A for driving the actuator 60 based on the action plan 40A input from the action planning unit 40 and outputs the drive signal 50A to the actuator 60 .
- the actuator 60 drives, for example, a motor of a moving body based on the drive signal 50A input from the control section 50 .
- the action plan unit 40 may output the generated action plan 40A to the map IF provision unit 25.
- the map IF providing unit 25 uses the environment map 25A and the acquired action plan 40A to derive an information map about the density of obstacles on the route from the current position to the target position, and the derived information
- the map may be output to the action planning section 40 as the parameter 25B.
- "the route from the current position to the target position" is referred to as a route RT.
- the map IF providing unit 25 generates an obstacle density gradient based on the information map about the obstacle density generated in this way, and uses the generated information about the obstacle density gradient as a parameter 25B for the action planning unit 40 may be output.
- the map IF providing unit 25 generates, for example, information about the distance to the obstacle on the route RT and the direction of the obstacle from the route RT based on the generated information map about the density of obstacles.
- the information may be output to the action planner 40 as parameters 25B.
- FIG. 4 is a diagram summarizing the compatibility between the map data format and the aircraft type.
- FIG. 5 is a diagram summarizing compatibility between the map data format and the operating location. Examples of the data format of the map data used by the map generators 23 and 24 include the data formats listed below. 1. grid map 2. voxel map 3. Octo Map 4. Height map 5. Hilbert Map 6. Multiplanar Map 7. Polygon Map
- Grid Map is a map data format that expresses the presence or absence of obstacles as a two-dimensional image. This is suitable when the moving body moves two-dimensionally.
- Grid Map the amount of data is small when the road surface is less uneven. Therefore, the Grid Map is suitable for small vehicles with a small amount of data, and is not very suitable for self-driving vehicles due to the large amount of calculation.
- Grid Map does not support three-dimensional space, so it cannot be applied to drones, multi-legged robots, and manipulators that handle three-dimensional space.
- Grid Map is suitable for home environments, factories, and hospitals where the floors are less uneven. However, it is not suitable for roads because the road surface is uneven and the amount of calculation is large.
- Grid Map does not support unevenness, it is not very suitable for forests and tunnels, and cannot be applied to hills, unevenness, and object manipulation that handle three-dimensional space.
- Voxel Map is a map data format that expresses space as a collection of three-dimensional rectangular parallelepipeds.
- the two horizontal axes can have the same resolution, while the vertical resolution can be set to be different from the two horizontal axes. Further, it is possible to set limits on the range of values.
- Voxel Map can express any shape, but the amount of data is large. Therefore, Voxel Map is not very suitable for small vehicles, multi-legged robots, and manipulators due to the large amount of calculation, and cannot be applied to self-driving cars and drones due to the enormous amount of calculation.
- Voxel Map is suitable for a home environment with a small amount of data, and is not so suitable for factories/hospitals, hills/unevenness, forests/tunnels, and object manipulation in that the amount of calculation is large, and the amount of calculation is enormous. Therefore, it cannot be applied to roads.
- Octo Map is a map data format that uses a tree structure called Octree. Good compatibility with man-made objects such as buildings. If the environment is a set of rectangular parallelepipeds, or if the Cartesian coordinate system for creating the map matches the planar direction of the environment, the amount of data will be small. Therefore, OctoMap is suitable for small vehicles, drones, multi-legged robots, and manipulators, where the amount of data is small. Octo Map is suitable for home environments, factories/hospitals, hills/unevenness, forests/tunnels, and object manipulation where the amount of data is small. .
- Height Map is a map data format that expresses altitude information as a two-dimensional image. Good compatibility with uneven road surfaces. When representing an uneven road surface, the amount of data is smaller than that of Voxel Map and Octo Map. However, when the road surface is a horizontal plane, the amount of data is larger than that of Voxel Map and Octo Map. In addition, when there is something flying in the air, it cannot be expressed appropriately. Therefore, Height Map is suitable for small vehicles, self-driving cars, and multi-legged robots with small amount of data, but it may not be suitable for drones depending on the environment, and it cannot represent objects for manipulators. Not applicable. In addition, Height Map is suitable for roads, hills, and irregularities, but it cannot be applied to home environments, factories, hospitals, forests, and tunnels because it cannot represent the bottom of objects. Also not applicable to operations.
- Hilbert Map is a map data format that expresses shapes as a set of three-dimensional Gaussian distributions (elliptical spheres). Good compatibility with natural objects. Arbitrary shapes can be approximated and represented, and when the environment is composed of various shapes and angles, the amount of data is smaller than Voxel Map and Octo Map. Therefore, Hilbert Map is suitable for small vehicles, drones, and multi-legged robots because the amount of data is small, but it is not so suitable for self-driving cars because the amount of calculation is large. It is not applicable to manipulators in that it is inferior. In addition, Hilbert Map is suitable for home environments, hills/unevenness, forests/tunnels in that the amount of data is small. However, it is not very suitable for factories, hospitals, and object manipulation because of its poor shape accuracy, and it is not very suitable for roads because of its large amount of calculations.
- MultiPlanar Map is a map data format that expresses shapes as a collection of multiple planes. It works well indoors, especially in buildings with little unevenness. If the environment consists of a set of planar shapes, the amount of data will be small. Therefore, the MultiPlanar Map is suitable for small vehicles in that the amount of data is small. However, its poor shape accuracy makes it unsuitable for self-driving cars and drones, and its extremely poor shape accuracy makes it unsuitable for manipulators. is also not applicable. In addition, MultiPlanar Map is suitable for factories and hospitals in that the amount of data is small, but it is not so suitable for home environments and roads in terms of shape accuracy. Not applicable to unevenness, forest/tunnel, and object manipulation.
- Polygon Map is a map data format that expresses shapes as a set of multiple convex polyhedrons (including a set of prisms and cylinders). It is compatible with the road environment of self-driving cars. When the environment is such that structures are lined up on the road surface, the amount of data becomes small. Therefore, Polygon Map is suitable for small vehicles and self-driving cars in that the amount of data is small, but it is not so suitable for drones and manipulators where it is difficult to divide the environment into multiple convex polyhedra. It cannot be applied to manipulators because the shape accuracy is significantly inferior. Also, Polygon Map is suitable for factories, hospitals, and roads in that the amount of data is small. However, it is not very suitable for home environments, forests, or tunnels, where it is difficult to divide the environment into multiple convex polyhedrons. It cannot be applied to hills and irregularities that are extremely difficult to divide into polyhedrons.
- FIG. 6 shows an example of the map data format selection procedure in the information processing system 1 .
- the map selection unit 22 first searches for planes included in the environmental characteristics 21A (step S101). When a plane is detected in the environmental characteristics 21A (step S102; Y), the map selection unit 22 clusters points above the detected plane in the environmental characteristics 21A (step S103). As a result, when a large number of objects that can be planarly approximated exist (step S104; Y), the map selection unit 22 selects MultiPlanar Map (step S105).
- step S104 determines whether there are many convex objects. As a result, when a large number of convex objects exist (step S106; Y), the map selection unit 22 selects Polygon Map (step S107). If the number of convex objects is small (step S106; N), the map selection unit 22 selects Grid Map (step S108).
- the map selection unit 22 When a plane is not detected in the environmental characteristics 21A (step S102; N), the map selection unit 22 creates a height histogram using the environmental characteristics 21A (step S109). At this time, when the histogram has one peak (step S110; Y), the map selection unit 22 selects Height Map (step S111). On the other hand, if the histogram has a plurality of peaks (step S110; N), the map selection unit 22 selects Octo Map or Hilbert Map (step S112).
- the data format of the map data generated by the map generation unit 23 is set according to the environmental characteristics 21A estimated by the environmental characteristics estimation unit 21.
- the environmental characteristics 21A estimated by the environmental characteristics estimation unit 21 For example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics.
- the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
- any one of the shape feature amount, the image feature amount, and the sound feature amount is used as the environmental characteristic 21A.
- the data format of the map data suitable for the environment As a result, the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
- the environmental characteristics 21A are estimated based on the environmental data 11A acquired by the sensor device unit 10.
- the data format of map data can be selected in real time. Since the data volume of the environment map can be kept low in real time, the environment map can be processed in real time, and the moving body can be moved autonomously.
- map generation is instructed to at least one of the map generation unit 23 and the map generation unit 24 according to the environmental characteristics 21A.
- the environmental characteristics 21A for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics.
- the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
- the map data generated by the map generation unit 23 is used to generate the parameters 25B used for the action plan.
- the parameters 25B used for the action plan for example, it is possible to accurately determine what path, direction, and posture to move from the self position to the target position. Therefore, it is possible to accurately generate an action plan.
- an action plan is created using the generated parameters 25B. This makes it possible to accurately generate an action plan. Further, in the present embodiment, the map data generated by the map generating section 23 and the action plan created by the action planning section 40 are used to generate the parameter 25B. As a result, not only the current position but also the route from the current position to the target position can be judged where there is a risk of collision with an obstacle, and the route can be reconstructed to avoid obstacles. becomes.
- the sensor device section 10 may have the signal processing section 12 as shown in FIG. 7, for example.
- the signal processing unit 12 processes environmental data 11A output from the sensor element 11 .
- the signal processing unit 12 detects, for example, an ROI included in the environment data 11A, and outputs data of the location corresponding to the detected ROI in the environment data 11A to the parameter generation unit 20 as the environment data 12A.
- the environment map can be processed in real time, and the moving object can be moved autonomously.
- the parameter generation unit 20 may be configured by, for example, a calculation unit 26 and a storage unit 27 as shown in FIG.
- the storage unit 27 stores a parameter generation program 27a that implements various functions included in the parameter generation unit 20 .
- the computing unit 26 includes, for example, a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). By loading the parameter generation program 27 a into the calculation unit 26 , the calculation unit 26 can execute various functions included in the parameter generation unit 20 .
- the present disclosure has been described above with reference to the embodiment and modifications thereof, the present disclosure is not limited to the above-described embodiment and the like, and various modifications are possible.
- the sensor device section 10, the parameter generation section 20, the storage section 30, the action planning section 40, the control section 50 and the actuator 60 may all be mounted on a moving body.
- the sensor device unit 10, the control unit 50, and the actuator 60 are mounted on the moving object, and the other configurations (for example, the parameter generation unit 20, the storage unit 30, the action planning unit 40).
- the server device configured to be able to communicate with a mobile unit.
- the present disclosure can have the following configuration.
- a map generator An information processing system comprising: a setting unit that sets a data format of map data generated by the map generation unit according to the estimated environmental characteristics.
- the environmental characteristics include any one of a shape feature amount, an image feature amount, and an audio feature amount.
- a sensor unit that acquires environmental data;
- the map generation unit a first map generator that generates first map data in a first data format; a second map generator that generates second map data in a second data format different from the first data format,
- the setting unit instructs at least one of the first map generation unit and the second map generation unit to generate a map according to the environmental characteristics.
- information processing system (5) The information processing system according to any one of (1) to (4), further comprising a parameter generator that generates parameters used for an action plan using the map data generated by the map generator. (6) (5) The information processing system according to (5), further comprising an action planning unit that creates the action plan using the parameters generated by the parameter generation unit.
- An information processing apparatus comprising a setting unit for setting a data format of map data generated by a map generation unit according to estimated environmental characteristics.
- An information processing method including setting a data format of map data generated by a map generation unit according to the estimated environmental characteristics.
- the data format of the map data generated by the map generation unit is set according to the estimated environmental characteristics.
- the data format of the map data generated by the map generation unit is set according to the estimated environmental characteristics.
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)
- Processing Or Creating Images (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
An information processing system according to one embodiment of the present disclosure comprises a map generating unit, and a setting unit that sets, in accordance with estimated environmental characteristics, a data format for map data generated by the map generating unit.
Description
本開示は、情報処理システム、情報処理装置および情報処理方法に関する。
The present disclosure relates to an information processing system, an information processing device, and an information processing method.
近年、外部環境を認識し、認識された環境に応じて自律的に移動するロボットなどの移動体に関する技術が開示されている(例えば、特許文献1参照)。
In recent years, technologies related to mobile objects such as robots that recognize the external environment and move autonomously according to the recognized environment have been disclosed (see Patent Document 1, for example).
移動体では、外部環境を認識するために各種センサが設けられており、各種センサから得られたセンサデータに基づいて、外部環境に対応する環境地図が構築される。自律的に移動体を移動させるためには、環境地図をリアルタイムに処理することが求められる。従って、環境地図のデータ量を低く抑えることの可能な情報処理システム、情報処理装置および情報処理方法を提供することが望ましい。
A mobile body is equipped with various sensors to recognize the external environment, and an environmental map corresponding to the external environment is constructed based on the sensor data obtained from the various sensors. Real-time processing of environmental maps is required for autonomous mobile objects. Therefore, it is desirable to provide an information processing system, an information processing apparatus, and an information processing method that can reduce the amount of data of an environment map.
本開示の一実施の形態に係る情報処理システムは、地図生成部と、推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定する設定部とを備えている。
An information processing system according to an embodiment of the present disclosure includes a map generation unit and a setting unit that sets the data format of map data generated by the map generation unit according to estimated environmental characteristics.
本開示の一実施の形態に係る情報処理装置は、推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定する設定部を備えている。
An information processing apparatus according to an embodiment of the present disclosure includes a setting unit that sets the data format of map data generated by the map generation unit according to estimated environmental characteristics.
本開示の一実施の形態に係る情報処理方法は、推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定することを含む。
An information processing method according to an embodiment of the present disclosure includes setting a data format of map data generated by a map generation unit according to estimated environmental characteristics.
本開示の一実施の形態に係る情報処理システム、情報処理装置および情報処理方法では、推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式が設定される。これにより、例えば、外部環境によって相性がある環境地図を用いた場合であっても、環境特性に応じた環境地図を選択することができる。
In the information processing system, information processing apparatus, and information processing method according to the embodiment of the present disclosure, the data format of the map data generated by the map generation unit is set according to the estimated environmental characteristics. As a result, for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics.
以下、本開示を実施するための形態について、図面を参照して詳細に説明する。なお、説明は以下の順序で行う。
1.実施の形態(図1~図6)
2.変形例(図7、図8)
EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing this disclosure is demonstrated in detail with reference to drawings. The description will be given in the following order.
1. Embodiment (Figs. 1 to 6)
2. Modified example (Fig. 7, Fig. 8)
1.実施の形態(図1~図6)
2.変形例(図7、図8)
EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing this disclosure is demonstrated in detail with reference to drawings. The description will be given in the following order.
1. Embodiment (Figs. 1 to 6)
2. Modified example (Fig. 7, Fig. 8)
<1.実施の形態>
[構成]
本開示の一実施の形態に係る情報処理システム1について説明する。図1は、情報処理システム1の機能ブロック例を表したものである。情報処理システム1は、例えば、図1に示したように、センサデバイス部10、パラメータ生成部20、記憶部30、行動計画部40、制御部50およびアクチュエータ60を備えている。 <1. Embodiment>
[Constitution]
An information processing system 1 according to an embodiment of the present disclosure will be described. FIG. 1 shows an example of functional blocks of an information processing system 1. As shown in FIG. The information processing system 1 includes, for example, asensor device section 10, a parameter generation section 20, a storage section 30, an action planning section 40, a control section 50, and an actuator 60, as shown in FIG.
[構成]
本開示の一実施の形態に係る情報処理システム1について説明する。図1は、情報処理システム1の機能ブロック例を表したものである。情報処理システム1は、例えば、図1に示したように、センサデバイス部10、パラメータ生成部20、記憶部30、行動計画部40、制御部50およびアクチュエータ60を備えている。 <1. Embodiment>
[Constitution]
An information processing system 1 according to an embodiment of the present disclosure will be described. FIG. 1 shows an example of functional blocks of an information processing system 1. As shown in FIG. The information processing system 1 includes, for example, a
センサデバイス部10は、例えば、外部環境を認識し、認識した外部環境に対応する環境データ11Aを取得するセンサ素子11を有している。センサ素子11は、取得した環境データ11Aをパラメータ生成部20に出力する。センサ素子11は、例えば、RGBカメラ、RGB-Dカメラ、深度センサ、赤外線センサ、イベントカメラ、集音センサである。
The sensor device unit 10 has, for example, a sensor element 11 that recognizes the external environment and acquires environmental data 11A corresponding to the recognized external environment. The sensor element 11 outputs the acquired environmental data 11A to the parameter generator 20 . The sensor element 11 is, for example, an RGB camera, an RGB-D camera, a depth sensor, an infrared sensor, an event camera, or a sound collection sensor.
RGBカメラは、例えば、単願の可視光画像センサであり、可視光を受光し電気信号に変換することにより得られるRGB画像データを出力する。RGB-Dカメラは、例えば、双眼の可視光画像センサであり、RGB-D画像データ(RGB画像データと、視差から得られる距離画像データと)を出力する。深度センサは、例えば、ToF(Time of Flight)センサ、または、Lider(Laser Imaging Detection and Ranging)であり、パルス状のレーザー照射に対する散乱光を測定することにより得られる距離画像データを出力する。赤外線センサは、例えば、赤外線を受光し電気信号に変換することにより得られた赤外線画像データを出力する。イベントカメラは、例えば、単願の可視光画像センサであり、フレーム間のRGB画像データの差分(差分画像データ)を出力する。集音センサは、例えば、外部環境から得られた音声データを出力する。センサデバイス部10は、例えば、外部環境から得られた各種データ(例えば、RGB画像データ、RGB-D画像データ、距離画像データ、赤外線画像データ、差分画像データまたは音声データ)を環境データ11Aとして出力する。
The RGB camera is, for example, a single visible light image sensor that outputs RGB image data obtained by receiving visible light and converting it into an electrical signal. The RGB-D camera is, for example, a binocular visible light image sensor, and outputs RGB-D image data (RGB image data and distance image data obtained from parallax). The depth sensor is, for example, a ToF (Time of Flight) sensor or a Lider (Laser Imaging Detection and Ranging), and outputs range image data obtained by measuring scattered light from pulsed laser irradiation. The infrared sensor outputs infrared image data obtained by, for example, receiving infrared rays and converting them into electric signals. The event camera is, for example, a single visible light image sensor, and outputs a difference between RGB image data (difference image data) between frames. The sound collection sensor outputs, for example, sound data obtained from the external environment. The sensor device unit 10 outputs, for example, various data (eg, RGB image data, RGB-D image data, distance image data, infrared image data, differential image data, or audio data) obtained from the external environment as environment data 11A. do.
パラメータ生成部20は、環境特性推定部21、地図選択部22、複数の地図生成部23,24および地図IF提供部25を有している。
The parameter generating unit 20 has an environmental characteristic estimating unit 21 , a map selecting unit 22 , a plurality of map generating units 23 and 24 and a map IF providing unit 25 .
環境特性推定部21は、センサデバイス部10から得られた環境データ11Aに基づいて環境特性21Aを推定する。環境特性21Aは、環境データ11Aに基づいて生成された環境特性の推定値であり、例えば、形状特徴量、画像特徴量および音声特徴量のいずれかを含む。形状特徴量とは、例えば、環境データ11Aにおける平面度、点群の分散または点群の平均高度を指している。画像特徴量とは、例えば、環境データ11Aにおけるエッジ数、特徴点数または色相・明度・彩度の平均もしくは分散を指している。音声特徴量とは、例えば、音量を指している。
The environmental characteristic estimation unit 21 estimates the environmental characteristic 21A based on the environmental data 11A obtained from the sensor device unit 10. The environmental characteristic 21A is an estimated value of the environmental characteristic generated based on the environmental data 11A, and includes, for example, any one of shape feature amount, image feature amount, and sound feature amount. The shape feature amount indicates, for example, the flatness, the dispersion of the point group, or the average height of the point group in the environment data 11A. The image feature amount indicates, for example, the number of edges, the number of feature points, or the average or variance of hue, brightness, and saturation in the environmental data 11A. The audio feature amount indicates volume, for example.
地図選択部22は、推定された環境特性21Aに応じて、複数の地図生成部23,24からなる地図生成部が生成する地図データのデータ形式を設定する。地図選択部22は、例えば、推定された環境特性21Aに応じて、地図生成部23,24の少なくとも一方を、使用する地図生成部として選択する。地図選択部22は、例えば、選択した地図生成部に対して地図生成を指示する。地図選択部22は、例えば、複数の地図生成部23,24に対して、使用する地図生成部の識別子ID(identification)の配列、または、複数の地図生成部23,24の使用を示す配列を出力してもよい。地図選択部22は、例えば、使用する地図生成部に対してだけ、地図生成を指示するコマンドや、地図生成の終了を意味するコマンドを出力してもよい。
The map selection unit 22 sets the data format of the map data generated by the map generation units 23 and 24 according to the estimated environmental characteristics 21A. The map selection unit 22 selects at least one of the map generation units 23 and 24 as the map generation unit to be used, for example, according to the estimated environmental characteristics 21A. For example, the map selection unit 22 instructs the selected map generation unit to generate a map. The map selection unit 22, for example, provides an array of identifiers ID (identification) of the map generation units to be used, or an array indicating the use of the plurality of map generation units 23 and 24, to the plurality of map generation units 23 and 24. may be output. For example, the map selection unit 22 may output a command to instruct map generation or a command to end map generation only to the map generation unit to be used.
地図生成部23は、環境データ11Aを用いて、第1データ形式の第1地図データ23Aを構築する。地図生成部23は、構築した第1地図データ23Aを、記憶部30から読み出した環境地図31に重ね合わせることにより、現時刻の環境地図31を生成する。地図生成部23は、生成した現時刻の環境地図31を記憶部30に格納する。地図生成部23は、地図選択部22によって選択された場合には、生成した現時刻の環境地図31の全体もしくは一部を地図IF提供部25に出力する。
The map generation unit 23 uses the environment data 11A to construct the first map data 23A in the first data format. The map generation unit 23 generates the environment map 31 at the current time by superimposing the constructed first map data 23A on the environment map 31 read from the storage unit 30 . The map generation unit 23 stores the generated environment map 31 at the current time in the storage unit 30 . When selected by the map selection unit 22 , the map generation unit 23 outputs all or part of the generated environment map 31 at the current time to the map IF provision unit 25 .
地図生成部24は、環境データ11Aを用いて、第1データ形式とは異なる第2データ形式の第2地図データ24Aを構築する。地図生成部24は、構築した第2地図データ24Aを、記憶部30から読み出した環境地図32に重ね合わせることにより、現時刻の環境地図32を生成する。地図生成部24は、生成した現時刻の環境地図32を記憶部30に格納する。地図生成部24は、地図選択部22によって選択された場合には、生成した現時刻の環境地図32の全体もしくは一部を地図IF提供部25に出力する。
The map generation unit 24 uses the environment data 11A to construct second map data 24A in a second data format different from the first data format. The map generation unit 24 generates the environment map 32 at the current time by superimposing the constructed second map data 24A on the environment map 32 read from the storage unit 30 . The map generation unit 24 stores the generated environment map 32 at the current time in the storage unit 30 . When selected by the map selection unit 22 , the map generation unit 24 outputs all or part of the generated environment map 32 at the current time to the map IF provision unit 25 .
なお、地図生成部23は、地図選択部22によって選択された場合に、第1地図データ23Aを地図IF提供部25に出力してもよい。さらに、地図生成部24は、地図選択部22によって選択された場合に、第2地図データ24Aを地図IF提供部25に出力してもよい。これらの場合、記憶部30を省略してもよい。
Note that the map generating unit 23 may output the first map data 23A to the map IF providing unit 25 when selected by the map selecting unit 22 . Further, the map generating unit 24 may output the second map data 24A to the map IF providing unit 25 when selected by the map selecting unit 22 . In these cases, the storage unit 30 may be omitted.
地図生成部23は、環境地図31のうち、一定期間を経過した過去の地図データを削除してもよい。また、地図生成部24は、環境地図32のうち、一定期間を経過した過去の地図データを削除してもよい。
The map generation unit 23 may delete past map data that has passed a certain period of time from the environment map 31 . In addition, the map generation unit 24 may delete past map data after a certain period of time has passed from the environment map 32 .
図2,図3は、地図生成部23で使用される第1地図データ23Aのデータ形式を「地図データ形式A」とし、地図生成部24で使用される第2地図データ24Aのデータ形式を「地図データ形式B」としたときの地図データの切り替えを概念的に示したものである。図2,図3には、センサデバイス部10を搭載した移動体が「地図データ形式A」に適した領域から「地図データ形式B」に適した領域に移動する様子が矢印で示されている。
2 and 3, the data format of the first map data 23A used in the map generator 23 is "map data format A", and the data format of the second map data 24A used in the map generator 24 is " It conceptually shows the switching of the map data when the map data format is "B". In FIGS. 2 and 3, arrows show how a moving body equipped with the sensor device unit 10 moves from an area suitable for "map data format A" to an area suitable for "map data format B." .
地図選択部22は、例えば、地図生成部23を選択しているときに、推定された環境特性21Aに応じて、地図生成部23から地図生成部24に切り替えるとする。このとき、地図選択部22は、例えば、図2に示したように、地図データのデータ形式を、「地図データ形式A」から「地図データ形式B」に直ちに切り替えてもよい。地図選択部22は、例えば、図3に示したように、「地図データ形式A」から「地図データ形式B」に切り替える際に、「地図データ形式A」と「地図データ形式B」とを同時に使用する期間を経て、地図データのデータ形式を、「地図データ形式A」から「地図データ形式B」に切り替えてもよい。
For example, when selecting the map generation unit 23, the map selection unit 22 switches from the map generation unit 23 to the map generation unit 24 according to the estimated environmental characteristics 21A. At this time, the map selection unit 22 may immediately switch the data format of the map data from "map data format A" to "map data format B", for example, as shown in FIG. For example, as shown in FIG. 3, when switching from "map data format A" to "map data format B", the map selection unit 22 selects "map data format A" and "map data format B" simultaneously. The data format of the map data may be switched from "map data format A" to "map data format B" after a period of use.
地図IF提供部25は、地図生成部23,24の少なくとも一方から入力された環境地図(以下、「環境地図25A」と称する。)を用いて、行動計画に用いるパラメータ25Bを生成する。地図IF提供部25は、例えば、環境地図25Aと、自己位置および自己速度についての情報とを用いて、衝突の有無もしくは確率についての情報地図を導出し、導出した情報地図をパラメータ25Bとして行動計画部40に出力してもよい。
The map IF providing unit 25 uses the environmental map (hereinafter referred to as "environmental map 25A") input from at least one of the map generating units 23 and 24 to generate parameters 25B used in the action plan. The map IF providing unit 25, for example, uses the environment map 25A and the information about the self-position and self-velocity to derive an information map regarding the presence or absence of a collision or the probability of collision, and uses the derived information map as a parameter 25B to create an action plan. You may output to the part 40. FIG.
地図IF提供部25は、例えば、環境地図25Aと、空間のある領域(例えば、自己位置を含む所定の空間領域)についての情報とを用いて、障害物密度についての情報地図を導出し、導出した情報地図をパラメータ25Bとして行動計画部40に出力してもよい。地図IF提供部25は、例えば、生成した障害物密度についての情報地図に基づいて、障害物密度勾配を生成し、生成した障害物密度勾配についての情報をパラメータ25Bとして行動計画部40に出力してもよい。地図IF提供部25は、例えば、生成した障害物密度についての情報地図に基づいて、ある位置からその周囲にある障害物までの距離や、ある位置からの障害物の方向についての情報を生成し、生成した情報をパラメータ25Bとして行動計画部40に出力してもよい。
The map IF providing unit 25 uses, for example, the environment map 25A and information about a certain spatial area (for example, a predetermined spatial area including the self-position) to derive an information map about the obstacle density, and derives the information map. The resulting information map may be output to the action planning section 40 as the parameter 25B. The map IF providing unit 25, for example, generates an obstacle density gradient based on the generated information map about the obstacle density, and outputs information about the generated obstacle density gradient to the action planning unit 40 as a parameter 25B. may The map IF providing unit 25 generates information about the distance from a certain position to obstacles in its surroundings and the direction of obstacles from a certain position, for example, based on the generated information map about the density of obstacles. , the generated information may be output to the action planning unit 40 as the parameter 25B.
記憶部30は、例えば、環境地図31,32を含むデータベースである。記憶部30は、例えば、DRAM(Dynamic Random Access Memory)などの揮発性メモリ、または、EEPROM(Electrically Erasable Programmable Read-Only Memory)やフラッシュメモリなどの不揮発性メモリによって構成されている。
The storage unit 30 is a database containing environment maps 31 and 32, for example. The storage unit 30 is configured by, for example, a volatile memory such as a DRAM (Dynamic Random Access Memory), or a nonvolatile memory such as an EEPROM (Electrically Erasable Programmable Read-Only Memory) or flash memory.
行動計画部40は、地図IF提供部25から入力されたパラメータ25Bに基づいて行動計画を作成する。行動計画部40は、例えば、パラメータ25Bに基づいて、自己位置から目標位置までどのような経路を、どのような向きおよび姿勢で移動するかを判断し、その判断の結果を行動計画40Aとして制御部50に出力する。制御部50は、行動計画部40から入力された行動計画40Aに基づいて、アクチュエータ60を駆動する駆動信号50Aを生成し、アクチュエータ60に出力する。アクチュエータ60は、制御部50から入力された駆動信号50Aに基づいて、例えば、移動体のモータなどを駆動する。
The action planning unit 40 creates an action plan based on the parameters 25B input from the map IF providing unit 25. For example, based on the parameter 25B, the action planning unit 40 determines what route, direction, and posture to move from the self position to the target position, and controls the result of the determination as the action plan 40A. Output to unit 50 . The control unit 50 generates a drive signal 50A for driving the actuator 60 based on the action plan 40A input from the action planning unit 40 and outputs the drive signal 50A to the actuator 60 . The actuator 60 drives, for example, a motor of a moving body based on the drive signal 50A input from the control section 50 .
行動計画部40は、生成した行動計画40Aを地図IF提供部25に出力してもよい。この場合、地図IF提供部25は、例えば、環境地図25Aと、取得した行動計画40Aとを用いて、現在位置から目標位置までの経路における障害物密度についての情報地図を導出し、導出した情報地図をパラメータ25Bとして行動計画部40に出力してもよい。以下では、「現在位置から目標位置までの経路」を経路RTと称する。地図IF提供部25は、例えば、このようにして生成した障害物密度についての情報地図に基づいて、障害物密度勾配を生成し、生成した障害物密度勾配についての情報をパラメータ25Bとして行動計画部40に出力してもよい。地図IF提供部25は、例えば、生成した障害物密度についての情報地図に基づいて、経路RTにある障害物までの距離や、経路RTからの障害物の方向についての情報を生成し、生成した情報をパラメータ25Bとして行動計画部40に出力してもよい。
The action plan unit 40 may output the generated action plan 40A to the map IF provision unit 25. In this case, the map IF providing unit 25, for example, uses the environment map 25A and the acquired action plan 40A to derive an information map about the density of obstacles on the route from the current position to the target position, and the derived information The map may be output to the action planning section 40 as the parameter 25B. Below, "the route from the current position to the target position" is referred to as a route RT. For example, the map IF providing unit 25 generates an obstacle density gradient based on the information map about the obstacle density generated in this way, and uses the generated information about the obstacle density gradient as a parameter 25B for the action planning unit 40 may be output. The map IF providing unit 25 generates, for example, information about the distance to the obstacle on the route RT and the direction of the obstacle from the route RT based on the generated information map about the density of obstacles. The information may be output to the action planner 40 as parameters 25B.
次に、図4、図5を参照しつつ、地図生成部23,24で用いられる地図データのデータ形式について説明する。図4は、地図データ形式と機体種別との相性についてまとめた図である。図5は、地図データ形式と運用場所との相性についてまとめた図である。地図生成部23,24で用いられる地図データのデータ形式としては、例えば、以下に列挙したデータ形式が挙げられる。
1.Grid Map
2.Voxel Map
3.Octo Map
4.Height Map
5.Hilbert Map
6.MultiPlanar Map
7.Polygon Map Next, the data format of the map data used in the map generators 23 and 24 will be described with reference to FIGS. 4 and 5. FIG. FIG. 4 is a diagram summarizing the compatibility between the map data format and the aircraft type. FIG. 5 is a diagram summarizing compatibility between the map data format and the operating location. Examples of the data format of the map data used by the map generators 23 and 24 include the data formats listed below.
1. grid map
2. voxel map
3. Octo Map
4. Height map
5. Hilbert Map
6. Multiplanar Map
7. Polygon Map
1.Grid Map
2.Voxel Map
3.Octo Map
4.Height Map
5.Hilbert Map
6.MultiPlanar Map
7.Polygon Map Next, the data format of the map data used in the
1. grid map
2. voxel map
3. Octo Map
4. Height map
5. Hilbert Map
6. Multiplanar Map
7. Polygon Map
Grid Mapは、2次元画像として障害物の有無などを表現する地図データ形式である。移動体が2次元で移動する場合に好適である。Grid Mapでは、路面の凹凸が少ない場合にデータ量が小さくなる。従って、Grid Mapは、データ量が少ない小型車両に適しており、計算量が多くなる点で自動運転車にはあまり適しているとはいえない。なお、Grid Mapは、3次元には未対応であるので、3次元空間を扱うドローンや、多脚のロボット、マニプレータには適用不可である。また、Grid Mapは、床の凹凸が少ない家庭環境や工場・病院には適している。しかし、路面の凹凸が多く計算量が多くなる点で道路にはあまり適していない。また、Grid Mapは凹凸に未対応であることから、森・トンネルにもあまり適しておらず、3次元空間を扱う丘陵・凹凸、物体操作には適用不可である。
Grid Map is a map data format that expresses the presence or absence of obstacles as a two-dimensional image. This is suitable when the moving body moves two-dimensionally. In Grid Map, the amount of data is small when the road surface is less uneven. Therefore, the Grid Map is suitable for small vehicles with a small amount of data, and is not very suitable for self-driving vehicles due to the large amount of calculation. Grid Map does not support three-dimensional space, so it cannot be applied to drones, multi-legged robots, and manipulators that handle three-dimensional space. In addition, Grid Map is suitable for home environments, factories, and hospitals where the floors are less uneven. However, it is not suitable for roads because the road surface is uneven and the amount of calculation is large. In addition, since Grid Map does not support unevenness, it is not very suitable for forests and tunnels, and cannot be applied to hills, unevenness, and object manipulation that handle three-dimensional space.
Voxel Mapは、空間を3次元の直方体の集合として表現する地図データ形式である。水平方向の2軸の解像度が互いに等しく、垂直方向については、水平方向の2軸の解像度とは異なる解像度に設定可能であり、さらに、値域に制限を設けることも可能である。Voxel Mapでは、任意の形状を表現することができるが、データ量が大きい。従って、Voxel Mapは、計算量が多くなる点で小型車両や多脚のロボット、マニプレータにはあまり適しておらず、計算量が膨大となる点で自動運転車やドローンには適用不可である。また、Voxel Mapは、データ量が少ない家庭環境に適しており、計算量が多くなる点で工場・病院、丘陵・凹凸、森・トンネル、物体操作にはあまり適しておらず、計算量が膨大になる点で道路には適用不可である。
Voxel Map is a map data format that expresses space as a collection of three-dimensional rectangular parallelepipeds. The two horizontal axes can have the same resolution, while the vertical resolution can be set to be different from the two horizontal axes. Further, it is possible to set limits on the range of values. Voxel Map can express any shape, but the amount of data is large. Therefore, Voxel Map is not very suitable for small vehicles, multi-legged robots, and manipulators due to the large amount of calculation, and cannot be applied to self-driving cars and drones due to the enormous amount of calculation. In addition, Voxel Map is suitable for a home environment with a small amount of data, and is not so suitable for factories/hospitals, hills/unevenness, forests/tunnels, and object manipulation in that the amount of calculation is large, and the amount of calculation is enormous. Therefore, it cannot be applied to roads.
Octo Mapは、Octreeと呼ばれる木構造を使用する地図データ形式である。建造物などの人工物との相性が良い。環境が直方体の集合形状となっている場合や、地図を作成する直交座標系が環境の平面方向と一致する場合には、データ量が小さくなる。従って、Octo Mapは、データ量が小さくなる小型車両やドローン、多脚のロボット、マニプレータには適しており、計算量が多くなる点で自動運転車にはあまり適していない。また、Octo Mapは、データ量が小さくなる家庭環境、工場・病院、丘陵・凹凸、森・トンネル、物体操作には適しているが、計算量が膨大になる点で道路には適用不可である。
Octo Map is a map data format that uses a tree structure called Octree. Good compatibility with man-made objects such as buildings. If the environment is a set of rectangular parallelepipeds, or if the Cartesian coordinate system for creating the map matches the planar direction of the environment, the amount of data will be small. Therefore, OctoMap is suitable for small vehicles, drones, multi-legged robots, and manipulators, where the amount of data is small. Octo Map is suitable for home environments, factories/hospitals, hills/unevenness, forests/tunnels, and object manipulation where the amount of data is small. .
Height Mapでは、高度情報を2次元画像として表現する地図データ形式である。凹凸のある路面との相性が良い。凹凸のある路面を表現する場合には、Voxel MapやOcto Mapと比べて、データ量が小さくなる。しかし、路面が水平な平面となっている場合には、Voxel MapやOcto Mapと比べて、データ量が大きくなる。また、空中に飛び出しているものがある場合には、それを適切に表現できない。従って、Height Mapは、データ量が小さくなる小型車両や自動運転車、多脚のロボットには適しているが、ドローンについては環境によっては適さない場合があり、マニプレータについては物体を表現できない点で適用不可である。また、Height Mapは、道路や丘陵・凹凸には適しているが、物体の下が表現できない点で家庭環境、工場・病院、森・トンネルには適用不可であり、物体が表現できない点で物体操作にも適用不可である。
Height Map is a map data format that expresses altitude information as a two-dimensional image. Good compatibility with uneven road surfaces. When representing an uneven road surface, the amount of data is smaller than that of Voxel Map and Octo Map. However, when the road surface is a horizontal plane, the amount of data is larger than that of Voxel Map and Octo Map. In addition, when there is something flying in the air, it cannot be expressed appropriately. Therefore, Height Map is suitable for small vehicles, self-driving cars, and multi-legged robots with small amount of data, but it may not be suitable for drones depending on the environment, and it cannot represent objects for manipulators. Not applicable. In addition, Height Map is suitable for roads, hills, and irregularities, but it cannot be applied to home environments, factories, hospitals, forests, and tunnels because it cannot represent the bottom of objects. Also not applicable to operations.
Hilbert Mapは、3次元のガウス分布(楕円球)の集合として形状を表現する地図データ形式である。自然物との相性が良い。任意形状を近似して表現することができ、環境が様々な形状や角度で構成されている場合に、Voxel MapやOcto Mapと比べて、データ量が小さくなる。従って、Hilbert Mapは、データ量が小さくなる点で小型車両やドローン、多脚のロボットには適しているが、計算量が多くなる点で自動運転車にはあまり適しておらず、形状精度が劣る点でマニプレータには適用不可である。また、Hilbert Mapは、データ量が小さくなる点で家庭環境、丘陵・凹凸、森・トンネルには適している。しかし、形状精度が劣る点で工場・病院や物体操作にはあまり適しておらず、計算量が多くなる点で道路にもあまり適していない。
Hilbert Map is a map data format that expresses shapes as a set of three-dimensional Gaussian distributions (elliptical spheres). Good compatibility with natural objects. Arbitrary shapes can be approximated and represented, and when the environment is composed of various shapes and angles, the amount of data is smaller than Voxel Map and Octo Map. Therefore, Hilbert Map is suitable for small vehicles, drones, and multi-legged robots because the amount of data is small, but it is not so suitable for self-driving cars because the amount of calculation is large. It is not applicable to manipulators in that it is inferior. In addition, Hilbert Map is suitable for home environments, hills/unevenness, forests/tunnels in that the amount of data is small. However, it is not very suitable for factories, hospitals, and object manipulation because of its poor shape accuracy, and it is not very suitable for roads because of its large amount of calculations.
MultiPlanar Mapは、複数の平面の集合として形状を表現する地図データ形式である。屋内、特に凹凸の少ない建物の中との相性が良い。環境が平面の集合形状で構成されている場合、データ量が小さくなる。従って、MultiPlanar Mapは、データ量が小さくなる点で小型車両には適している。しかし、形状精度が劣る点で自動運転車やドローンにはあまり適しておらず、形状精度が著しく劣る点でマニプレータには適用不可であり、凹凸で構成される環境が想定されやすい多脚のロボットについても適用不可である。また、MultiPlanar Mapは、データ量が小さくなる点で工場・病院には適しているが、形状精度が劣る点で家庭環境や道路にはあまり適しておらず、形状精度が著しく劣る点で丘陵・凹凸、森・トンネル、物体操作には適用不可である。
MultiPlanar Map is a map data format that expresses shapes as a collection of multiple planes. It works well indoors, especially in buildings with little unevenness. If the environment consists of a set of planar shapes, the amount of data will be small. Therefore, the MultiPlanar Map is suitable for small vehicles in that the amount of data is small. However, its poor shape accuracy makes it unsuitable for self-driving cars and drones, and its extremely poor shape accuracy makes it unsuitable for manipulators. is also not applicable. In addition, MultiPlanar Map is suitable for factories and hospitals in that the amount of data is small, but it is not so suitable for home environments and roads in terms of shape accuracy. Not applicable to unevenness, forest/tunnel, and object manipulation.
Polygon Mapは、複数の凸多面体の集合(角柱や円柱の集合も含む)として形状を表現する地図データ形式である。自動運転車の道路環境などとの相性が良い。環境が路面の上に構造物が並ぶような場合に、データ量が小さくなる。従って、Polygon Mapは、データ量が小さくなる点で小型車両や自動運転車には適しているが、環境を複数の凸多面体に分割するのが困難なドローンやマニプレータにはあまり適しておらず、形状精度が著しく劣る点でマニプレータには適用不可である。また、Polygon Mapは、データ量が小さくなる点で工場・病院や道路には適している。しかし、環境を複数の凸多面体に分割するのが困難な家庭環境や森・トンネルにはあまり適しておらず、形状精度が劣る点で物体操作にもあまり適しておらず、環境を複数の凸多面体に分割するのが極めて困難な丘陵・凹凸には適用不可である。
Polygon Map is a map data format that expresses shapes as a set of multiple convex polyhedrons (including a set of prisms and cylinders). It is compatible with the road environment of self-driving cars. When the environment is such that structures are lined up on the road surface, the amount of data becomes small. Therefore, Polygon Map is suitable for small vehicles and self-driving cars in that the amount of data is small, but it is not so suitable for drones and manipulators where it is difficult to divide the environment into multiple convex polyhedra. It cannot be applied to manipulators because the shape accuracy is significantly inferior. Also, Polygon Map is suitable for factories, hospitals, and roads in that the amount of data is small. However, it is not very suitable for home environments, forests, or tunnels, where it is difficult to divide the environment into multiple convex polyhedrons. It cannot be applied to hills and irregularities that are extremely difficult to divide into polyhedrons.
[動作]
次に、情報処理システム1における地図データの形式の選択について説明する。 [motion]
Next, selection of the map data format in the information processing system 1 will be described.
次に、情報処理システム1における地図データの形式の選択について説明する。 [motion]
Next, selection of the map data format in the information processing system 1 will be described.
図6は、情報処理システム1における地図データの形式の選択手順の一例を表したものである。地図選択部22は、まず、環境特性21Aに含まれる平面を探索する(ステップS101)。地図選択部22は、環境特性21Aにおいて平面が検出された場合(ステップS102;Y)、環境特性21Aにおいて、検出した平面よりも上方の点群をクラスタリングする(ステップS103)。その結果、平面近似可能な物体が多数、存在する場合には(ステップS104;Y)、地図選択部22は、MultiPlanar Mapを選択する(ステップS105)。
FIG. 6 shows an example of the map data format selection procedure in the information processing system 1 . The map selection unit 22 first searches for planes included in the environmental characteristics 21A (step S101). When a plane is detected in the environmental characteristics 21A (step S102; Y), the map selection unit 22 clusters points above the detected plane in the environmental characteristics 21A (step S103). As a result, when a large number of objects that can be planarly approximated exist (step S104; Y), the map selection unit 22 selects MultiPlanar Map (step S105).
一方、平面近似可能な物体の数が少ない場合には(ステップS104;N)、地図選択部22は、凸形状の物体が多数、存在するか否か判定する。その結果、凸形状の物体が多数、存在する場合には(ステップS106;Y)、地図選択部22は、Polygon Mapを選択する(ステップS107)。凸形状の物体の数が少ない場合には(ステップS106;N)、地図選択部22は、Grid Mapを選択する(ステップS108)。
On the other hand, if the number of objects that can be planarly approximated is small (step S104; N), the map selection unit 22 determines whether there are many convex objects. As a result, when a large number of convex objects exist (step S106; Y), the map selection unit 22 selects Polygon Map (step S107). If the number of convex objects is small (step S106; N), the map selection unit 22 selects Grid Map (step S108).
地図選択部22は、環境特性21Aにおいて平面が検出されなかった場合(ステップS102;N)、環境特性21Aを用いて高さのヒストグラムを作成する(ステップS109)。このとき、ヒストグラムのピークが1つの場合には(ステップS110;Y)、地図選択部22は、Height Mapを選択する(ステップS111)。一方、ヒストグラムのピークが複数ある場合には(ステップS110;N)、地図選択部22は、Octo MapまたはHilbert Mapを選択する(ステップS112)。
When a plane is not detected in the environmental characteristics 21A (step S102; N), the map selection unit 22 creates a height histogram using the environmental characteristics 21A (step S109). At this time, when the histogram has one peak (step S110; Y), the map selection unit 22 selects Height Map (step S111). On the other hand, if the histogram has a plurality of peaks (step S110; N), the map selection unit 22 selects Octo Map or Hilbert Map (step S112).
[効果]
次に、情報処理システム1の効果について説明する。 [effect]
Next, effects of the information processing system 1 will be described.
次に、情報処理システム1の効果について説明する。 [effect]
Next, effects of the information processing system 1 will be described.
本実施の形態では、環境特性推定部21で推定された環境特性21Aに応じて、地図生成部23が生成する地図データのデータ形式が設定される。これにより、例えば、外部環境によって相性がある環境地図を用いた場合であっても、環境特性に応じた環境地図を選択することができる。その結果、環境地図のデータ量を低く抑えることができるので、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。
In the present embodiment, the data format of the map data generated by the map generation unit 23 is set according to the environmental characteristics 21A estimated by the environmental characteristics estimation unit 21. As a result, for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics. As a result, the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
また、本実施の形態では、環境特性21Aとして、形状特徴量、画像特徴量および音声特徴量のいずれかが用いられる。これにより、環境に適した地図データのデータ形式を選択することができる。その結果、環境地図のデータ量を低く抑えることができるので、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。
Also, in the present embodiment, any one of the shape feature amount, the image feature amount, and the sound feature amount is used as the environmental characteristic 21A. As a result, it is possible to select the data format of the map data suitable for the environment. As a result, the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
また、本実施の形態では、センサデバイス部10で取得した環境データ11Aに基づいて環境特性21Aが推定される。これにより、センサデバイス部10が搭載された移動体が、例えば「地図データ形式A」に適した領域から「地図データ形式B」に適した領域に移動した場合であっても、環境に適した地図データのデータ形式をリアルタイムに選択することができる。環境地図のデータ量をリアルタイムに低く抑えることができるので、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。
Also, in the present embodiment, the environmental characteristics 21A are estimated based on the environmental data 11A acquired by the sensor device unit 10. As a result, even if the moving object equipped with the sensor device unit 10 moves from an area suitable for "map data format A" to an area suitable for "map data format B", for example, The data format of map data can be selected in real time. Since the data volume of the environment map can be kept low in real time, the environment map can be processed in real time, and the moving body can be moved autonomously.
また、本実施の形態では、環境特性21Aに応じて、地図生成部23および地図生成部24の少なくとも一方に対して地図生成が指示される。これにより、例えば、外部環境によって相性がある環境地図を用いた場合であっても、環境特性に応じた環境地図を選択することができる。その結果、環境地図のデータ量を低く抑えることができるので、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。
Also, in the present embodiment, map generation is instructed to at least one of the map generation unit 23 and the map generation unit 24 according to the environmental characteristics 21A. As a result, for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics. As a result, the data amount of the environment map can be kept low, so the environment map can be processed in real time, and the moving object can be moved autonomously.
また、本実施の形態では、地図生成部23で生成された地図データを用いて、行動計画に用いるパラメータ25Bが生成される。これにより、例えば、自己位置から目標位置までどのような経路を、どのような向きおよび姿勢で移動するかを的確に判断することが可能となる。従って、行動計画を的確に生成することが可能となる。
In addition, in the present embodiment, the map data generated by the map generation unit 23 is used to generate the parameters 25B used for the action plan. As a result, for example, it is possible to accurately determine what path, direction, and posture to move from the self position to the target position. Therefore, it is possible to accurately generate an action plan.
また、本実施の形態では、生成されたパラメータ25Bを用いて行動計画が作成される。これにより、行動計画を的確に生成することが可能となる。また、本実施の形態において、地図生成部23で生成された地図データと、行動計画部40で作成した行動計画とを用いて、パラメータ25Bが生成される。これにより、現在位置だけでなく、現在位置から目標位置までの経路において、どの場所で障害物への衝突の危険性があるか判断することができ、障害物を避けた経路の再構築が可能となる。
Also, in the present embodiment, an action plan is created using the generated parameters 25B. This makes it possible to accurately generate an action plan. Further, in the present embodiment, the map data generated by the map generating section 23 and the action plan created by the action planning section 40 are used to generate the parameter 25B. As a result, not only the current position but also the route from the current position to the target position can be judged where there is a risk of collision with an obstacle, and the route can be reconstructed to avoid obstacles. becomes.
<2.変形例>
[変形例A]
上記実施の形態およびその変形例において、センサデバイス部10が、例えば、図7に示したように、信号処理部12を有していてもよい。信号処理部12は、センサ素子11から出力された環境データ11Aを処理する。信号処理部12は、例えば、環境データ11Aに含まれるROIを検出し、環境データ11Aのうち、検出したROIに対応する箇所のデータを環境データ12Aとしてパラメータ生成部20に出力する。このように、センサデバイス部10において、データ量をあらかじめ削減することにより、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。 <2. Variation>
[Modification A]
In the above embodiments and modifications thereof, thesensor device section 10 may have the signal processing section 12 as shown in FIG. 7, for example. The signal processing unit 12 processes environmental data 11A output from the sensor element 11 . The signal processing unit 12 detects, for example, an ROI included in the environment data 11A, and outputs data of the location corresponding to the detected ROI in the environment data 11A to the parameter generation unit 20 as the environment data 12A. By reducing the amount of data in advance in the sensor device unit 10 in this manner, the environment map can be processed in real time, and the moving object can be moved autonomously.
[変形例A]
上記実施の形態およびその変形例において、センサデバイス部10が、例えば、図7に示したように、信号処理部12を有していてもよい。信号処理部12は、センサ素子11から出力された環境データ11Aを処理する。信号処理部12は、例えば、環境データ11Aに含まれるROIを検出し、環境データ11Aのうち、検出したROIに対応する箇所のデータを環境データ12Aとしてパラメータ生成部20に出力する。このように、センサデバイス部10において、データ量をあらかじめ削減することにより、環境地図をリアルタイムに処理することができ、自律的に移動体を移動させることができる。 <2. Variation>
[Modification A]
In the above embodiments and modifications thereof, the
[変形例B]
上記実施の形態およびその変形例において、パラメータ生成部20に含まれる各種機能がハードウェアで実現されていてもよいし、ソフトウェアで実現されていてもよい。上記実施の形態およびその変形例において、パラメータ生成部20が、例えば、図8に示したように、演算部26と、記憶部27とにより構成されていてもよい。この場合、記憶部27には、パラメータ生成部20に含まれる各種機能を実現するパラメータ生成プログラム27aが記憶されている。演算部26は、例えば、CPU(Central Processing Unit)およびGPU(Graphics Processing Unit)を含んで構成される。パラメータ生成プログラム27aが演算部26にロードされることにより、演算部26は、パラメータ生成部20に含まれる各種機能を実行することができる。 [Modification B]
In the above embodiments and their modifications, various functions included in theparameter generator 20 may be realized by hardware or by software. In the above embodiments and modifications thereof, the parameter generation unit 20 may be configured by, for example, a calculation unit 26 and a storage unit 27 as shown in FIG. In this case, the storage unit 27 stores a parameter generation program 27a that implements various functions included in the parameter generation unit 20 . The computing unit 26 includes, for example, a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). By loading the parameter generation program 27 a into the calculation unit 26 , the calculation unit 26 can execute various functions included in the parameter generation unit 20 .
上記実施の形態およびその変形例において、パラメータ生成部20に含まれる各種機能がハードウェアで実現されていてもよいし、ソフトウェアで実現されていてもよい。上記実施の形態およびその変形例において、パラメータ生成部20が、例えば、図8に示したように、演算部26と、記憶部27とにより構成されていてもよい。この場合、記憶部27には、パラメータ生成部20に含まれる各種機能を実現するパラメータ生成プログラム27aが記憶されている。演算部26は、例えば、CPU(Central Processing Unit)およびGPU(Graphics Processing Unit)を含んで構成される。パラメータ生成プログラム27aが演算部26にロードされることにより、演算部26は、パラメータ生成部20に含まれる各種機能を実行することができる。 [Modification B]
In the above embodiments and their modifications, various functions included in the
以上、実施の形態およびその変形例を挙げて本開示を説明したが、本開示は上記実施の形態等に限定されるものではなく、種々変形が可能である。例えば、上記実施の形態等において、センサデバイス部10、パラメータ生成部20、記憶部30、行動計画部40、制御部50およびアクチュエータ60が全て、移動体に搭載されていてもよい。また、例えば、上記実施の形態等において、センサデバイス部10、制御部50およびアクチュエータ60が移動体に搭載され、それ以外の構成(例えば、パラメータ生成部20、記憶部30、行動計画部40)が移動体と通信可能に構成されたサーバ装置に設けられていてもよい。
Although the present disclosure has been described above with reference to the embodiment and modifications thereof, the present disclosure is not limited to the above-described embodiment and the like, and various modifications are possible. For example, in the above-described embodiments and the like, the sensor device section 10, the parameter generation section 20, the storage section 30, the action planning section 40, the control section 50 and the actuator 60 may all be mounted on a moving body. Further, for example, in the above embodiments and the like, the sensor device unit 10, the control unit 50, and the actuator 60 are mounted on the moving object, and the other configurations (for example, the parameter generation unit 20, the storage unit 30, the action planning unit 40). may be provided in a server device configured to be able to communicate with a mobile unit.
なお、本明細書中に記載された効果は、あくまで例示である。本開示の効果は、本明細書中に記載された効果に限定されるものではない。本開示が、本明細書中に記載された効果以外の効果を持っていてもよい。
It should be noted that the effects described in this specification are merely examples. The effects of the present disclosure are not limited to the effects described herein. The disclosure may have advantages other than those described herein.
また、例えば、本開示は以下のような構成を取ることができる。
(1)
地図生成部と、
推定された環境特性に応じて、前記地図生成部が生成する地図データのデータ形式を設定する設定部と
を備えた情報処理システム。
(2)
前記環境特性は、形状特徴量、画像特徴量および音声特徴量のいずれかを含む
(1)に記載の情報処理システム。
(3)
環境データを取得するセンサ部と、
前記センサ部で取得した前記環境データに基づいて前記環境特性を推定する推定部と
を更に備えた
(1)または(2)に記載の情報処理システム。
(4)
前記地図生成部は、
第1データ形式の第1地図データを生成する第1地図生成部と、
前記第1データ形式とは異なる第2データ形式の第2地図データを生成する第2地図生成部と
を備え、
前記設定部は、前記環境特性に応じて、前記第1地図生成部および前記第2地図生成部の少なくとも一方に対して地図生成を指示する
(1)ないし(3)のいずれか1つに記載の情報処理システム。
(5)
前記地図生成部で生成された前記地図データを用いて、行動計画に用いるパラメータを生成するパラメータ生成部を更に備えた (1)ないし(4)のいずれか1つに記載の情報処理システム。
(6)
前記パラメータ生成部で生成された前記パラメータを用いて、前記行動計画を作成する行動計画部を更に備えた
(5)に記載の情報処理システム。
(7)
前記パラメータ生成部は、前記地図生成部で生成された前記地図データと、前記行動計画部で作成した前記行動計画とを用いて、前記パラメータを生成する
(6)に記載の情報処理システム。
(8)
推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定する設定部を備えた
情報処理装置。
(9)
推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定することを含む
情報処理方法。 Further, for example, the present disclosure can have the following configuration.
(1)
a map generator;
An information processing system comprising: a setting unit that sets a data format of map data generated by the map generation unit according to the estimated environmental characteristics.
(2)
(1) The information processing system according to (1), wherein the environmental characteristics include any one of a shape feature amount, an image feature amount, and an audio feature amount.
(3)
a sensor unit that acquires environmental data;
The information processing system according to (1) or (2), further comprising: an estimation unit that estimates the environmental characteristics based on the environmental data acquired by the sensor unit.
(4)
The map generation unit
a first map generator that generates first map data in a first data format;
a second map generator that generates second map data in a second data format different from the first data format,
According to any one of (1) to (3), the setting unit instructs at least one of the first map generation unit and the second map generation unit to generate a map according to the environmental characteristics. information processing system.
(5)
The information processing system according to any one of (1) to (4), further comprising a parameter generator that generates parameters used for an action plan using the map data generated by the map generator.
(6)
(5) The information processing system according to (5), further comprising an action planning unit that creates the action plan using the parameters generated by the parameter generation unit.
(7)
The information processing system according to (6), wherein the parameter generating section generates the parameters using the map data generated by the map generating section and the action plan created by the action planning section.
(8)
An information processing apparatus comprising a setting unit for setting a data format of map data generated by a map generation unit according to estimated environmental characteristics.
(9)
An information processing method including setting a data format of map data generated by a map generation unit according to the estimated environmental characteristics.
(1)
地図生成部と、
推定された環境特性に応じて、前記地図生成部が生成する地図データのデータ形式を設定する設定部と
を備えた情報処理システム。
(2)
前記環境特性は、形状特徴量、画像特徴量および音声特徴量のいずれかを含む
(1)に記載の情報処理システム。
(3)
環境データを取得するセンサ部と、
前記センサ部で取得した前記環境データに基づいて前記環境特性を推定する推定部と
を更に備えた
(1)または(2)に記載の情報処理システム。
(4)
前記地図生成部は、
第1データ形式の第1地図データを生成する第1地図生成部と、
前記第1データ形式とは異なる第2データ形式の第2地図データを生成する第2地図生成部と
を備え、
前記設定部は、前記環境特性に応じて、前記第1地図生成部および前記第2地図生成部の少なくとも一方に対して地図生成を指示する
(1)ないし(3)のいずれか1つに記載の情報処理システム。
(5)
前記地図生成部で生成された前記地図データを用いて、行動計画に用いるパラメータを生成するパラメータ生成部を更に備えた (1)ないし(4)のいずれか1つに記載の情報処理システム。
(6)
前記パラメータ生成部で生成された前記パラメータを用いて、前記行動計画を作成する行動計画部を更に備えた
(5)に記載の情報処理システム。
(7)
前記パラメータ生成部は、前記地図生成部で生成された前記地図データと、前記行動計画部で作成した前記行動計画とを用いて、前記パラメータを生成する
(6)に記載の情報処理システム。
(8)
推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定する設定部を備えた
情報処理装置。
(9)
推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定することを含む
情報処理方法。 Further, for example, the present disclosure can have the following configuration.
(1)
a map generator;
An information processing system comprising: a setting unit that sets a data format of map data generated by the map generation unit according to the estimated environmental characteristics.
(2)
(1) The information processing system according to (1), wherein the environmental characteristics include any one of a shape feature amount, an image feature amount, and an audio feature amount.
(3)
a sensor unit that acquires environmental data;
The information processing system according to (1) or (2), further comprising: an estimation unit that estimates the environmental characteristics based on the environmental data acquired by the sensor unit.
(4)
The map generation unit
a first map generator that generates first map data in a first data format;
a second map generator that generates second map data in a second data format different from the first data format,
According to any one of (1) to (3), the setting unit instructs at least one of the first map generation unit and the second map generation unit to generate a map according to the environmental characteristics. information processing system.
(5)
The information processing system according to any one of (1) to (4), further comprising a parameter generator that generates parameters used for an action plan using the map data generated by the map generator.
(6)
(5) The information processing system according to (5), further comprising an action planning unit that creates the action plan using the parameters generated by the parameter generation unit.
(7)
The information processing system according to (6), wherein the parameter generating section generates the parameters using the map data generated by the map generating section and the action plan created by the action planning section.
(8)
An information processing apparatus comprising a setting unit for setting a data format of map data generated by a map generation unit according to estimated environmental characteristics.
(9)
An information processing method including setting a data format of map data generated by a map generation unit according to the estimated environmental characteristics.
本開示の一実施の形態に係る情報処理システム、情報処理装置および情報処理方法では、推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式が設定される。これにより、例えば、外部環境によって相性がある環境地図を用いた場合であっても、環境特性に応じた環境地図を選択することができる。その結果、移動体が環境特性の異なる外部環境をまたいで移動する場合であっても環境地図のデータ量を低く抑えることができる。
In the information processing system, information processing apparatus, and information processing method according to the embodiment of the present disclosure, the data format of the map data generated by the map generation unit is set according to the estimated environmental characteristics. As a result, for example, even when using an environment map that is compatible with the external environment, it is possible to select an environment map according to the environmental characteristics. As a result, it is possible to keep the data amount of the environment map low even when the mobile body moves across external environments having different environmental characteristics.
本出願は、日本国特許庁において2021年4月15日に出願された日本特許出願番号第2021-069082号を基礎として優先権を主張するものであり、この出願のすべての内容を参照によって本出願に援用する。
This application claims priority based on Japanese Patent Application No. 2021-069082 filed on April 15, 2021 at the Japan Patent Office, and the entire contents of this application are incorporated herein by reference. incorporated into the application.
当業者であれば、設計上の要件や他の要因に応じて、種々の修正、コンビネーション、サブコンビネーション、および変更を想到し得るが、それらは添付の請求の範囲やその均等物の範囲に含まれるものであることが理解される。
Depending on design requirements and other factors, those skilled in the art may conceive various modifications, combinations, subcombinations, and modifications that fall within the scope of the appended claims and their equivalents. It is understood that
Claims (9)
- 地図生成部と、
推定された環境特性に応じて、前記地図生成部が生成する地図データのデータ形式を設定する設定部と
を備えた情報処理システム。 a map generator;
An information processing system comprising: a setting unit that sets a data format of map data generated by the map generation unit according to the estimated environmental characteristics. - 前記環境特性は、形状特徴量、画像特徴量および音声特徴量のいずれかを含む
請求項1に記載の情報処理システム。 The information processing system according to claim 1, wherein the environmental characteristics include any one of a shape feature amount, an image feature amount, and an audio feature amount. - 環境データを取得するセンサ部と、
前記センサ部で取得した前記環境データに基づいて前記環境特性を推定する推定部と
を更に備えた
請求項1に記載の情報処理システム。 a sensor unit that acquires environmental data;
The information processing system according to claim 1, further comprising an estimation unit that estimates the environmental characteristics based on the environmental data acquired by the sensor unit. - 前記地図生成部は、
第1データ形式の第1地図データを生成する第1地図生成部と、
前記第1データ形式とは異なる第2データ形式の第2地図データを生成する第2地図生成部と
を備え、
前記設定部は、前記環境特性に応じて、前記第1地図生成部および前記第2地図生成部の少なくとも一方に対して地図生成を指示する
請求項1に記載の情報処理システム。 The map generation unit
a first map generator that generates first map data in a first data format;
a second map generator that generates second map data in a second data format different from the first data format,
The information processing system according to claim 1, wherein the setting unit instructs at least one of the first map generation unit and the second map generation unit to generate a map according to the environmental characteristics. - 前記地図生成部で生成された前記地図データを用いて、行動計画に用いるパラメータを生成するパラメータ生成部を更に備えた
請求項1に記載の情報処理システム。 The information processing system according to claim 1, further comprising a parameter generator that generates parameters used for an action plan using the map data generated by the map generator. - 前記パラメータ生成部で生成された前記パラメータを用いて、前記行動計画を作成する行動計画部を更に備えた
請求項5に記載の情報処理システム。 The information processing system according to claim 5, further comprising an action planning section that creates the action plan using the parameters generated by the parameter generation section. - 前記パラメータ生成部は、前記地図生成部で生成された前記地図データと、前記行動計画部で作成した前記行動計画とを用いて、前記パラメータを生成する
請求項6に記載の情報処理システム。 The information processing system according to claim 6, wherein the parameter generating section generates the parameters using the map data generated by the map generating section and the action plan created by the action planning section. - 推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定する設定部を備えた
情報処理装置。 An information processing apparatus comprising a setting unit for setting a data format of map data generated by a map generation unit according to estimated environmental characteristics. - 推定された環境特性に応じて、地図生成部が生成する地図データのデータ形式を設定することを含む
情報処理方法。 An information processing method including setting a data format of map data generated by a map generation unit according to the estimated environmental characteristics.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021069082A JP2022163931A (en) | 2021-04-15 | 2021-04-15 | Information processing system, information processing device and information processing method |
JP2021-069082 | 2021-04-15 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022219875A1 true WO2022219875A1 (en) | 2022-10-20 |
Family
ID=83639529
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/002935 WO2022219875A1 (en) | 2021-04-15 | 2022-01-26 | Information processing system, information processing device, and information processing method |
Country Status (2)
Country | Link |
---|---|
JP (1) | JP2022163931A (en) |
WO (1) | WO2022219875A1 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024134803A1 (en) * | 2022-12-21 | 2024-06-27 | オプテックス株式会社 | Door opening/closing system and sensor for door opening/closing system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013033356A (en) * | 2011-08-01 | 2013-02-14 | Toyota Central R&D Labs Inc | Autonomous mobile device |
JP2020154751A (en) * | 2019-03-20 | 2020-09-24 | 学校法人明治大学 | Mobile body control device, mobile body control method, and computer program |
-
2021
- 2021-04-15 JP JP2021069082A patent/JP2022163931A/en active Pending
-
2022
- 2022-01-26 WO PCT/JP2022/002935 patent/WO2022219875A1/en active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013033356A (en) * | 2011-08-01 | 2013-02-14 | Toyota Central R&D Labs Inc | Autonomous mobile device |
JP2020154751A (en) * | 2019-03-20 | 2020-09-24 | 学校法人明治大学 | Mobile body control device, mobile body control method, and computer program |
Also Published As
Publication number | Publication date |
---|---|
JP2022163931A (en) | 2022-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10665115B2 (en) | Controlling unmanned aerial vehicles to avoid obstacle collision | |
JP6980755B2 (en) | Estimating the dimensions of the enclosed space using a multi-directional camera | |
CN108290294B (en) | Mobile robot and control method thereof | |
Matthies et al. | Stereo vision-based obstacle avoidance for micro air vehicles using disparity space | |
Saha et al. | A real-time monocular vision-based frontal obstacle detection and avoidance for low cost UAVs in GPS denied environment | |
Costante et al. | Exploiting photometric information for planning under uncertainty | |
US8078399B2 (en) | Method and device for three-dimensional path planning to avoid obstacles using multiple planes | |
US11119498B2 (en) | Systems and methods for simulation utilizing a segmentable monolithic mesh | |
US10578453B2 (en) | Render-based trajectory planning | |
Rojas-Fernández et al. | Performance comparison of 2D SLAM techniques available in ROS using a differential drive robot | |
WO2022219875A1 (en) | Information processing system, information processing device, and information processing method | |
Sanchez-Rodriguez et al. | A survey on stereo vision-based autonomous navigation for multi-rotor MUAVs | |
Smith et al. | Real-time egocentric navigation using 3d sensing | |
Smith et al. | PiPS: Planning in perception space | |
Sheikh et al. | Stereo vision-based optimal path planning with stochastic maps for mobile robot navigation | |
WO2021131990A1 (en) | Information processing device, information processing method, and program | |
Karrer et al. | Real-time dense surface reconstruction for aerial manipulation | |
Chikhalikar et al. | An object-oriented navigation strategy for service robots leveraging semantic information | |
Sidaoui et al. | A-slam: Human in-the-loop augmented slam | |
Dinaux et al. | FAITH: Fast iterative half-plane focus of expansion estimation using optic flow | |
Spampinato et al. | An embedded stereo vision module for industrial vehicles automation | |
Umari | Multi-robot map exploration based on multiple rapidly-exploring randomized trees | |
Andert et al. | Combined grid and feature-based occupancy map building in large outdoor environments | |
Barak et al. | Two Dimensional Mapping by using Single Ultrasonic Sensor. | |
Bellone | Watch your step! terrain traversability for robot control |
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: 22787803 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 22787803 Country of ref document: EP Kind code of ref document: A1 |