Position estimation device, robot system including same, and position estimation method

Abstract

To provide a position estimation device capable of efficiently estimating a position, a robot system including the same, and a position estimation method thereof.SOLUTION: A position estimation device according to the present invention includes: a particle generation unit that generates a plurality of particles on a map; a particle election unit that calculates positional accuracy of each of the plurality of particles based on sensing data regarding a position of a target, and selects at least one particle among the plurality of particles as a reference particle based on the positional accuracy; and a position determination unit that determines the position of the target based on the reference particle.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a position estimation device, a robot system including the same, and a position estimation method thereof. [Background technology]

[0002] With the advancement of robot-related technology, various types of robots are now being used in homes and industrial sites. These robots are recognized as having great utility in a wide range of fields, from household robots that move around the house and perform household chores such as cleaning, to industrial robots that perform mechanical tasks in industrial sites such as manufacturing.

[0003] In particular, these robots perform tasks on behalf of humans while moving across multiple areas, making it essential for these mobile robots to be able to recognize their current location. Various methods have been tried to estimate the location of mobile robots, such as using separate wireless devices or comparing location information from multiple viewpoints. However, these methods require separate communication devices or complex calculation algorithms, resulting in high-spec, high-cost robot systems. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-190164 Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention has been made in consideration of the above-mentioned conventional technology, and an object of the present invention is to provide a position estimation device that can efficiently perform position estimation using only sensors inside the robot, a robot system including the same, and a position estimation method thereof. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, one aspect of the present invention provides a position estimation device, which includes a particle generation unit that generates a plurality of particles on a map, a particle selection unit that calculates the position accuracy of each of the plurality of particles based on sensing data regarding the position of an object and selects at least one particle of the plurality of particles as a reference particle based on the position accuracy, and a position determination unit that determines the position of the object based on the reference particle.

[0007] The particle selection unit may select a particle having a position accuracy equal to or greater than a reference value as the reference particle.

[0008] The particle selection unit may compare the map with the sensing data to calculate the position accuracy of each of the plurality of particles.

[0009] The particle selection unit may calculate the position accuracy of each of the plurality of particles by convolving a cost map obtained by distance transforming the map with the sensing data.

[0010] The particle selector may redistribute the remaining particles, excluding the reference particle, around the reference particle.

[0011] The particle selector may redistribute the remaining particles, excluding the reference particles, based on the positional accuracy of each of the reference particles.

[0012] When there are a plurality of reference particles, the particle selection unit may recalculate the positional accuracy of each of the reference particles after moving the object by more than a certain distance, and reselect the reference particles based on the recalculated positional accuracy of the reference particles.

[0013] The particle selector may repeatedly select the reference particles until the reference particles converge to one.

[0014] The position determination unit can determine the position of the converged reference particle as the position of the target.

[0015] When there are a plurality of reference particles, the position determination unit may determine the position of the reference particle having the highest positional accuracy as the position of the object.

[0016] In order to achieve the above-mentioned object, one aspect of the present invention provides a robot system including a sensor unit that senses topographical information around the robot, and a position estimation unit that calculates the positional accuracy of each of a plurality of particles generated on a map based on sensing data from the sensor unit, selects at least one particle of the plurality of particles as a reference particle based on the positional accuracy, and estimates the position of an object based on the reference particle.

[0017] The sensor unit preferably includes a LiDAR that detects the distance between the robot and a surrounding object.

[0018] It is preferable that the robot further includes a driving unit that moves the robot and calculates the distance traveled by the object based on a value measured by a wheel encoder.

[0019] A position estimation method according to one aspect of the present invention, which has been made to achieve the above-mentioned object, is characterized by including the steps of generating a plurality of particles on a map, calculating the position accuracy of each of the plurality of particles based on sensing data regarding the position of an object, selecting at least one particle from the plurality of particles as a reference particle based on the position accuracy, and determining the position of the object based on the reference particle.

[0020] Preferably, the step of selecting at least one particle from the plurality of particles as a reference particle comprises selecting a particle having a positional accuracy equal to or greater than a reference value as the reference particle.

[0021] Preferably, the step of selecting at least one particle of the plurality of particles as a reference particle includes calculating the positional accuracy of each of the plurality of particles by comparing the map with the sensing data.

[0022] The step of selecting at least one particle of the plurality of particles as a reference particle may include calculating the positional accuracy of each of the plurality of particles by convolving a cost map obtained by distance transforming the map with the sensing data.

[0023] In the step of selecting at least one particle of the plurality of particles as a reference particle, if there are a plurality of reference particles, it is preferable to recalculate the positional accuracy of each of the reference particles after moving the object, and reselect the reference particle based on the recalculated positional accuracy of the reference particle.

[0024] Preferably, the step of selecting at least one particle from the plurality of particles as the reference particle comprises repeatedly selecting the reference particle until the number of reference particles converges to one.

[0025] Preferably, the step of determining the position of the target determines the position of the converged reference particle as the position of the target. [Effects of the Invention]

[0026] The position estimation device, robot system including the same, and position estimation method according to the present invention can perform global positioning of a robot using only sensors inside the robot, without the need for additional devices or complex algorithms, thereby reducing the manufacturing costs of the robot and enabling efficient position estimation procedures.

[0027] In addition, various other effects are provided that can be directly or indirectly grasped from this specification. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a block diagram showing a configuration of a robot system according to an embodiment of the present invention. [Figure 2] 1 is a block diagram showing a configuration of a position estimation device according to an embodiment of the present invention. [Figure 3] 10A and 10B are diagrams for explaining a particle generation operation of the position estimation device according to the embodiment of the present invention. [Figure 4] 10A and 10B are diagrams illustrating a reference particle selection operation of the position estimation device according to the embodiment of the present invention. [Figure 5] 4A and 4B are diagrams illustrating a position estimation operation of a position estimation device according to an embodiment of the present invention. [Figure 6] 4 is a flowchart illustrating the operation of a position estimation device according to an embodiment of the present invention. [Figure 7] 2 is a flowchart illustrating a location estimation method according to an embodiment of the present invention. [Figure 8] FIG. 1 illustrates a computing system according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0029] Hereinafter, specific examples of embodiments of the present invention will be described in detail with reference to the drawings. When assigning reference numerals to components in each drawing, the same reference numerals are used to the same components even if they appear in different drawings. Furthermore, when describing the embodiments of the present invention, if a detailed description of related well-known configurations or functions is deemed to hinder understanding of the embodiments of the present invention, the detailed description will be omitted.

[0030] When describing components of embodiments of the present invention, terms such as "first," "second," "A," "B," "(a)," and "(b)" are used; however, these terms are merely used to distinguish the component from other components and do not limit the nature, order, or sequence of the components. Furthermore, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the present invention pertains. Terms defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning given in the context of the relevant art, and are not to be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0031] Hereinafter, an embodiment of the present invention will be described in detail with reference to FIGS.

[0032] FIG. 1 is a block diagram showing the configuration of a robot system according to an embodiment of the present invention.

[0033] Referring to FIG. 1, a robot system 100 according to an embodiment of the present invention includes a sensor unit 110, a drive unit 120, and a position estimation unit .

[0034] The sensor unit 110 senses topographical information around the robot. For example, the sensor unit 110 senses the distance between the robot and a surrounding object. Specifically, the sensor unit 110 measures the distance by transmitting a signal (e.g., an optical signal) to a surrounding object (e.g., a wall, an obstacle, etc.) and detecting the signal reflected from the object. In addition, the distance information measured by the sensor unit 110 is transmitted to the position estimation unit 130. For example, the sensor unit 110 includes a LiDAR.

[0035] The driving unit 120 moves the robot. For example, the driving unit 120 moves the robot to a destination using wheels. In this case, the driving unit 120 calculates the distance traveled by the robot based on values ​​measured by a wheel encoder. In addition, information about the distance traveled calculated by the driving unit 120 is transmitted to the position estimation unit 130.

[0036] The position estimation unit 130 generates a plurality of particles over the entire map and calculates the position accuracy (position precision) of each of the plurality of particles based on sensing data from the sensor unit 110. Here, the particles are placed at candidate positions of an object (e.g., a robot). In this case, the position estimation unit 130 calculates the position accuracy for each of the plurality of particles by matching data (e.g., distance information) measured by the sensor unit 110 with each of the plurality of particles.

[0037] The position estimation unit 130 selects at least one particle from the plurality of particles as a reference particle based on the position accuracy calculated for each of the plurality of particles. At this time, a particle having a position accuracy equal to or greater than a predetermined reference value is selected as the reference particle. The remaining particles excluding the reference particle are redistributed around the reference particle.

[0038] The position estimation unit 130 then estimates the position of the target based on the selected reference particle. For example, the position estimation unit 130 determines the reference particle with the highest positional accuracy as the position of the target, or moves the position of the target using the driving unit 120 and then repeatedly selects reference particles until they converge to one, thereby determining the position of the target. This will be described in detail below with reference to FIGS. 2 to 6.

[0039] FIG. 2 is a block diagram showing the configuration of a position estimation device according to an embodiment of the present invention.

[0040] 2, a position estimation device 130 according to an embodiment of the present invention includes a particle generation unit 131, a particle selection unit 132, and a position determination unit 133. In this case, the position estimation device 130 of FIG. 2 has substantially the same configuration as the position estimation unit 130 of FIG. 1.

[0041] The particle generation unit 131 generates a plurality of particles on a map. At this time, each particle is placed on the map at a candidate position of an object whose position is to be estimated. Alternatively, each particle is placed uniformly at regular intervals on the map.

[0042] The particle selection unit 132 calculates the position accuracy of each of the plurality of particles based on sensing data related to the target position. In this case, the particle selection unit 132 compares the map with the sensing data to calculate the position accuracy of each of the plurality of particles. For example, the particle selection unit 132 calculates the position accuracy of each of the plurality of particles by comparing the position of each of the plurality of particles on the map with sensing information acquired from an external sensor (e.g., the sensor unit 110 of FIG. 1).

[0043] Specifically, the particle selection unit 132 calculates the position accuracy of each of the plurality of particles by convolving a cost map obtained by distance transforming a pre-stored map with the sensing data. Here, the cost map is the result of distance transforming a map of the area where the target is located, and is configured so that, for example, the closer a point on the map is to a point with a wall or obstacle, the smaller the value, and the larger the value is for a point without walls or obstacles in the vicinity.

[0044] Furthermore, the particle selection unit 132 selects at least one particle from the plurality of particles as a reference particle based on the calculated positional accuracy. Here, a particle that is highly likely to correspond to the current position of an object (e.g., a robot) is selected as the reference particle. For example, the particle selection unit 132 selects a particle whose positional accuracy is equal to or greater than a reference value as the reference particle.

[0045] The particle selection unit 132 redistributes the remaining particles excluding the reference particle around the reference particle. For example, the particle selection unit 132 redistributes the remaining particles excluding the reference particle based on the positional accuracy of each of the reference particles. In this case, if there are multiple reference particles, the particle selection unit 132 distributes the remaining particles by differentiating them according to the positional accuracy of each reference particle. For example, if the positional accuracy of three reference particles is 90%, 80%, and 60%, the remaining particles are distributed to the three reference particles in a ratio of 9:8:6.

[0046] In addition, if there are multiple reference particles, the particle selection unit 132 recalculates the positional accuracy of each reference particle after moving the object by a certain distance or more. In this case, the particle selection unit 132 reselects the reference particle based on the recalculated positional accuracy of the reference particle. In this manner, the particle selection unit 132 repeatedly selects reference particles until the number of reference particles converges to one.

[0047] The position determination unit 133 determines the position of the target based on the reference particles. For example, the position determination unit 133 determines the position of the reference particles converged to one by the particle selection unit 132 as the position of the target. Alternatively, when there are multiple reference particles, the position determination unit 133 determines the position of the reference particle with the highest positional accuracy as the position of the target.

[0048] In this way, the position estimation device 130 according to one embodiment of the present invention and the robot system 100 including the same can perform global positioning of the robot using only the sensors inside the robot, without the need for additional devices or complex algorithms, thereby reducing the manufacturing costs of the robot and enabling efficient position estimation procedures.

[0049] FIG. 3 is a diagram for explaining the particle generation operation of the position estimation device according to one embodiment of the present invention.

[0050] Referring to (a) of FIG. 3, R denotes an object (e.g., a robot) whose position is to be estimated by a position estimation device 130 according to an embodiment of the present invention, and L1 denotes sensing data measured by the sensor unit 110 (e.g., a LIDAR sensor). Here, L1 denotes a portion of a structure (wall) extracted from the sensing data measured by the LIDAR sensor. Also, as shown in (a) of FIG. 3, the position estimation device 130 sets an arbitrary position on a map as the position of the object R.

[0051] 3(b), the position estimation device 130 generates a plurality of particles P at any position on the map. In this case, the position estimation device 130 generates the particles uniformly over the entire area of ​​the map, as shown in FIG. 3(b), or generates the particles only at candidate positions on the map where the object R is likely to be located. For example, coordinates are assigned to each of the plurality of particles P.

[0052] FIG. 4 is a diagram for explaining the reference particle selection operation of the position estimation device according to one embodiment of the present invention.

[0053] Referring to (a) of Figure 4, C indicates a cost map obtained by distance transforming a map using a position estimation device 130 according to one embodiment of the present invention, M indicates map data (e.g., walls), and L indicates data measured by a LIDAR sensor. As indicated by the shading in (a) of Figure 4, in the cost map C, areas close to walls or obstacles have low values ​​and are displayed dark, while areas farther away from the walls or obstacles have larger values ​​and are displayed brighter. Therefore, it is possible to determine from the cost map C whether or not a wall or obstacle exists around a specific position.

[0054] In this way, the position estimation device 130 according to an embodiment of the present invention performs convolution between the data of the cost map C calculated by the distance transformation and the sensing data received from the LIDAR sensor, and as a result, it is possible to calculate the position accuracy for each of the multiple particles generated in (a) of FIG.

[0055] 4(b), four particles out of the plurality of particles are calculated as reference particles (P1 to P4) by the position estimation device 130. In this case, particles having a position accuracy equal to or greater than a reference value (e.g., 80%) are selected as the reference particles (P1 to P4).

[0056] Furthermore, as shown in Figure 4(b), the remaining particles are redistributed to each of the reference particles (P1 to P4). In this case, the remaining particles are distributed differently depending on the positional accuracy of each of the reference particles (P1 to P4). For example, in Figure 4(b), the positional accuracy increases from P4 to P1, so the remaining particles are also distributed in greater numbers in the order P4 to P1.

[0057] FIG. 5 is a diagram for explaining the position estimation operation of the position estimation device according to one embodiment of the present invention.

[0058] Referring to FIG. 5(a), R indicates that the object in FIGS. 3 and 4 has been moved a certain distance to the right, and L2 indicates the measurement data of the lidar sensor resulting from the movement of the object R. P1' indicates that the position estimation device 130 according to an embodiment of the present invention has converged to one of the reference particles (P1 to P4) shown in FIG. 4(b), which is P1. In this case, in FIG. 5(a), after the object R has been moved a certain distance or more to the right, the reference particles are reselected using the above-described method, so that the particles converge to a single reference particle P1'. In this case, the above-described process is repeated until the reference particles converge to a single particle. For example, the process of converging the reference particles is performed according to a PSO (particle swarm optimization) algorithm.

[0059] 5(b) shows that the position estimation device 130 has determined the position of the reference particle P1' on the map shown in FIG. 5(a) as the current position of the target R'. Also, as shown in FIG. 5(b), it can be seen that the sensing data L' obtained by the lidar sensor also corresponds to the actual current position of the target R'.

[0060] As described above, conventional robots estimate their current position (t) based on their position at a previous time (t-1), which means that if the robot system loses its current position due to a restart, an error, etc., continuous position estimation is no longer possible. However, the robot system 100 including the position estimation device 130 according to an embodiment of the present invention effectively estimates the current position even when a situation occurs in which the robot loses its current position, thereby enabling continuous operation of the robot system.

[0061] FIG. 6 is a flowchart showing the operation of the position estimation device according to one embodiment of the present invention.

[0062] 6, a location estimation device 130 according to an embodiment of the present invention first generates particles on a map (S10). At this time, a predetermined number of particles are generated uniformly across the entire map or centered on candidate positions. Then, the location accuracy of each generated particle is measured (S20). In this case, the location accuracy is measured by measuring the consistency of each particle with the location on the map based on sensing data from the sensor unit 110 (e.g., a lidar sensor).

[0063] Then, a reference particle is selected from the generated particles based on the positional accuracy (S30). For example, a particle whose positional accuracy is equal to or greater than a predetermined reference value is selected as the reference particle, or n particles with the highest positional accuracy are selected as the reference particle.

[0064] Then, the remaining particles are redistributed based on the reference particles (S40). For example, if there are a plurality of reference particles, the remaining particles are distributed with differences depending on the positional accuracy of the reference particles.

[0065] Next, it is determined whether the reference particles converge (S50). If the reference particles converge to one (YES), the position of the reference particle is determined (determined) as the position of the object (S60). On the other hand, if the reference particles do not converge and multiple particles are calculated (NO), the object is moved a certain distance or more (S70). Furthermore, the generated particles are also moved in accordance with the movement of the object (S80). Then, the above steps (S20 to S50) are repeated until the reference particles converge to one.

[0066] FIG. 7 is a flowchart illustrating a location estimation method according to one embodiment of the present invention.

[0067] A location estimation method according to an embodiment of the present invention will now be described in detail with reference to Fig. 7. In the following, it is assumed that the location estimation device 130 of Fig. 2 performs the process of Fig. 7. In addition, in the description of Fig. 7, the operations described as being performed by the device are controlled by the processor of the location estimation device 130.

[0068] 7, in a location estimation method according to an embodiment of the present invention, first, a plurality of particles are generated on a map (S110). At this time, each particle is arranged at a candidate location of an object whose location is to be estimated on the map. Alternatively, each particle is arranged uniformly at regular intervals on the map.

[0069] Furthermore, the position accuracy of each of the plurality of particles is calculated based on the sensing data regarding the target position (S120). In this case, the map and the sensing data are compared to calculate the position accuracy of each of the plurality of particles. For example, in step S120, the position accuracy of each of the plurality of particles is calculated by comparing the position of each of the plurality of particles on the map with sensing information acquired from an external sensor (e.g., the sensor unit 110 of FIG. 1). Specifically, in step S120, a cost map acquired by distance transforming a pre-stored map is convoluted with the sensing data to calculate the position accuracy of each of the plurality of particles.

[0070] Next, based on the calculated positional accuracy, at least one particle from the plurality of particles is selected as a reference particle (S130). Here, a particle that is likely to correspond to the current position of an object (e.g., a robot) is selected as the reference particle. For example, in step S130, a particle whose positional accuracy is equal to or greater than a reference value is selected as the reference particle.

[0071] In step S130, the remaining particles excluding the reference particle are redistributed around the reference particle. For example, the remaining particles excluding the reference particle are redistributed based on the positional accuracy of each of the reference particles. If there are multiple reference particles, the remaining particles are distributed with differences depending on the positional accuracy of each reference particle.

[0072] In addition, in steps S120 and S130, if there are multiple reference particles, the object is moved a certain distance or more, and then the positional accuracy of each of the reference particles is recalculated. In this case, the reference particle is reselected based on the recalculated positional accuracy of the reference particle. In this manner, reference particles are repeatedly selected until the number of reference particles converges to one.

[0073] Then, the position of the object is determined based on the reference particles (S140). For example, in step S140, the position of the reference particles that converge to one is determined as the position of the object. Alternatively, if there are multiple reference particles, the position of the reference particle with the highest positional accuracy is determined as the position of the object.

[0074] In this way, the position estimation method according to one embodiment of the present invention performs global positioning of a robot using only sensors inside the robot, without the need for additional devices or complex algorithms, thereby reducing the manufacturing costs of the robot and enabling efficient position estimation procedures.

[0075] FIG. 8 is a diagram illustrating a computing system according to one embodiment of the present invention.

[0076] Referring to FIG. 8, computing system 1000 includes at least one processor 1100, memory 1300, user interface input device 1400, user interface output device 1500, storage 1600, and network interface 1700, all coupled via a bus 1200.

[0077] The processor 1100 is a central processing unit (CPU) or a semiconductor device that executes processing based on instructions stored in the memory 1300 and / or the storage 1600. The memory 1300 and the storage 1600 include various volatile or non-volatile storage media. For example, the memory 1300 includes a read-only memory (ROM) 1310 and a random access memory (RAM) 1320.

[0078] Thus, the steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in software modules executed by processor 1100, or in a combination of the two. The software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or other storage medium (i.e., memory 1300 and / or storage 1600).

[0079] An exemplary storage medium may be coupled to processor 1100 such that processor 1100 can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to processor 1100. The processor and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

[0080] The above description is merely an illustrative example of the technical concept of the present invention, and various modifications and variations can be made by a person having ordinary knowledge in the technical field to which the present invention pertains without departing from the essential characteristics of the present invention.

[0081] Therefore, the embodiments disclosed in this specification are not intended to limit the technical idea of ​​the present invention, but are merely for illustrative purposes, and these embodiments do not limit the scope of the technical idea of ​​the present invention. [Explanation of symbols]

[0082] 100 Robot Systems 110 Sensor unit 120 Drive unit 130 Position estimation unit (position estimation device) 131 Particle Generation Unit 132 Particle Selection Section 133 Position determination section 1000 Computing Systems 1100 processor 1200 Bus 1300 memory 1310 ROM 1320 RAM 1400 User Interface Input Device 1500 User interface output device 1600 Storage 1700 Network Interface

Claims

1. a particle generation unit that generates a plurality of particles on a map; a particle selection unit that calculates a positional accuracy of each of the plurality of particles based on sensing data relating to a position of an object, and selects at least one particle of the plurality of particles as a reference particle based on the positional accuracy; a position determination unit that determines a position of the target based on the reference particle, The particle selection unit calculates the position accuracy of each of the plurality of particles by convolving a cost map obtained by distance transforming the map with the sensing data.

2. The position estimation device according to claim 1 , wherein the particle selection unit selects a particle having a position accuracy equal to or greater than a reference value as the reference particle.

3. The position estimation device according to claim 1 , wherein the particle selection unit compares the map with the sensing data to calculate the position accuracy of each of the plurality of particles.

4. The position estimation device according to claim 1 , wherein the particle selection unit redistributes the remaining particles, excluding the reference particle, around the reference particle.

5. The position estimation device according to claim 4 , wherein the particle selection unit redistributes the remaining particles excluding the reference particles based on the position accuracy of each of the reference particles.

6. 2. The position estimation device of claim 1, wherein, when there are a plurality of reference particles, the particle selection unit recalculates the position accuracy of each of the reference particles after moving the object by a certain distance or more, and reselects the reference particles based on the recalculated position accuracy of the reference particles.

7. 7. The position estimation device according to claim 6, wherein the particle selection unit repeatedly selects the reference particles until the number of the reference particles converges to one.

8. 8. The position estimation device according to claim 7, wherein the position determination unit determines the position of the converged reference particle as the position of the target.

9. The position estimation device according to claim 1 , wherein, when there are a plurality of reference particles, the position determination unit determines the position of the reference particle having the highest positional accuracy as the position of the object.

10. a sensor unit that senses topographical information around the robot; a position estimation unit that calculates a position accuracy of each of a plurality of particles generated on a map based on sensing data from the sensor unit, selects at least one particle of the plurality of particles as a reference particle based on the position accuracy, and estimates a position of an object based on the reference particle, The position estimation unit a cost map obtained by distance transforming the map and convolving the sensing data to calculate the positional accuracy of each of the plurality of particles, in order to select at least one particle of the plurality of particles as a reference particle.

11. The robot system according to claim 10 , wherein the sensor unit includes a LiDAR (Light Detection and Ranging) device that detects a distance between the robot and a surrounding object.

12. The robot system of claim 10 , further comprising a driving unit that moves the robot and calculates the distance traveled by the object based on a value measured by a wheel encoder.

13. generating a plurality of particles on a map; calculating a position accuracy of each of the plurality of particles based on sensing data relating to the position of an object; selecting at least one particle of the plurality of particles as a reference particle based on the position accuracy; determining a position of the object based on the reference particles; The step of selecting at least one particle from the plurality of particles as a reference particle comprises convolving a cost map obtained by distance transforming the map with the sensing data to calculate the position accuracy of each of the plurality of particles.

14. The method of claim 13 , wherein the step of selecting at least one particle from the plurality of particles as a reference particle comprises selecting a particle whose position accuracy is equal to or greater than a reference value as the reference particle.

15. 14. The position estimation method of claim 13, wherein the step of selecting at least one particle from the plurality of particles as a reference particle includes comparing the map with the sensing data to calculate the position accuracy of each of the plurality of particles.

16. 14. The position estimation method of claim 13, wherein the step of selecting at least one particle among the plurality of particles as a reference particle comprises, if there are a plurality of reference particles, recalculating the position accuracy of each of the reference particles after moving the object, and reselecting the reference particle based on the recalculated position accuracy of the reference particle.

17. 17. The position estimation method according to claim 16, wherein the step of selecting at least one particle from the plurality of particles as a reference particle comprises repeatedly selecting the reference particle until the number of the reference particles converges to one.

18. The step of determining the position of the target is performed by determining the position of the converged reference particle by 18. The position estimation method according to claim 17, wherein the position is determined as the position.

Images