Growing Neural Gas with Different Topologies for 3D Space Perception
Three-dimensional space perception is one of the most important capabilities for an autonomous mobile robot in order to operate a task in an unknown environment adaptively since the autonomous robot needs to detect the target object and estimate the 3D pose of the target object for performing given...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/3/1705 |
| _version_ | 1797488932369203200 |
|---|---|
| author | Yuichiro Toda Akimasa Wada Hikari Miyase Koki Ozasa Takayuki Matsuno Mamoru Minami |
| author_facet | Yuichiro Toda Akimasa Wada Hikari Miyase Koki Ozasa Takayuki Matsuno Mamoru Minami |
| author_sort | Yuichiro Toda |
| collection | DOAJ |
| description | Three-dimensional space perception is one of the most important capabilities for an autonomous mobile robot in order to operate a task in an unknown environment adaptively since the autonomous robot needs to detect the target object and estimate the 3D pose of the target object for performing given tasks efficiently. After the 3D point cloud is measured by an RGB-D camera, the autonomous robot needs to reconstruct a structure from the 3D point cloud with color information according to the given tasks since the point cloud is unstructured data. For reconstructing the unstructured point cloud, growing neural gas (GNG) based methods have been utilized in many research studies since GNG can learn the data distribution of the point cloud appropriately. However, the conventional GNG based methods have unsolved problems about the scalability and multi-viewpoint clustering. In this paper, therefore, we propose growing neural gas with different topologies (GNG-DT) as a new topological structure learning method for solving the problems. GNG-DT has multiple topologies of each property, while the conventional GNG method has a single topology of the input vector. In addition, the distance measurement in the winner node selection uses only the position information for preserving the environmental space of the point cloud. Next, we show several experimental results of the proposed method using simulation and RGB-D datasets measured by Kinect. In these experiments, we verified that our proposed method almost outperforms the other methods from the viewpoint of the quantization and clustering errors. Finally, we summarize our proposed method and discuss the future direction on this research. |
| first_indexed | 2024-03-10T00:10:18Z |
| format | Article |
| id | doaj.art-9e491c16106242868af14cbc0f1f22bc |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:10:18Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-9e491c16106242868af14cbc0f1f22bc2023-11-23T16:02:07ZengMDPI AGApplied Sciences2076-34172022-02-01123170510.3390/app12031705Growing Neural Gas with Different Topologies for 3D Space PerceptionYuichiro Toda0Akimasa Wada1Hikari Miyase2Koki Ozasa3Takayuki Matsuno4Mamoru Minami5Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanGraduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanGraduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanGraduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanGraduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanGraduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Kita-Ku, Okayama 700-8530, JapanThree-dimensional space perception is one of the most important capabilities for an autonomous mobile robot in order to operate a task in an unknown environment adaptively since the autonomous robot needs to detect the target object and estimate the 3D pose of the target object for performing given tasks efficiently. After the 3D point cloud is measured by an RGB-D camera, the autonomous robot needs to reconstruct a structure from the 3D point cloud with color information according to the given tasks since the point cloud is unstructured data. For reconstructing the unstructured point cloud, growing neural gas (GNG) based methods have been utilized in many research studies since GNG can learn the data distribution of the point cloud appropriately. However, the conventional GNG based methods have unsolved problems about the scalability and multi-viewpoint clustering. In this paper, therefore, we propose growing neural gas with different topologies (GNG-DT) as a new topological structure learning method for solving the problems. GNG-DT has multiple topologies of each property, while the conventional GNG method has a single topology of the input vector. In addition, the distance measurement in the winner node selection uses only the position information for preserving the environmental space of the point cloud. Next, we show several experimental results of the proposed method using simulation and RGB-D datasets measured by Kinect. In these experiments, we verified that our proposed method almost outperforms the other methods from the viewpoint of the quantization and clustering errors. Finally, we summarize our proposed method and discuss the future direction on this research.https://www.mdpi.com/2076-3417/12/3/17053D space perceptiongrowing neural gastopological structure learning method |
| spellingShingle | Yuichiro Toda Akimasa Wada Hikari Miyase Koki Ozasa Takayuki Matsuno Mamoru Minami Growing Neural Gas with Different Topologies for 3D Space Perception Applied Sciences 3D space perception growing neural gas topological structure learning method |
| title | Growing Neural Gas with Different Topologies for 3D Space Perception |
| title_full | Growing Neural Gas with Different Topologies for 3D Space Perception |
| title_fullStr | Growing Neural Gas with Different Topologies for 3D Space Perception |
| title_full_unstemmed | Growing Neural Gas with Different Topologies for 3D Space Perception |
| title_short | Growing Neural Gas with Different Topologies for 3D Space Perception |
| title_sort | growing neural gas with different topologies for 3d space perception |
| topic | 3D space perception growing neural gas topological structure learning method |
| url | https://www.mdpi.com/2076-3417/12/3/1705 |
| work_keys_str_mv | AT yuichirotoda growingneuralgaswithdifferenttopologiesfor3dspaceperception AT akimasawada growingneuralgaswithdifferenttopologiesfor3dspaceperception AT hikarimiyase growingneuralgaswithdifferenttopologiesfor3dspaceperception AT kokiozasa growingneuralgaswithdifferenttopologiesfor3dspaceperception AT takayukimatsuno growingneuralgaswithdifferenttopologiesfor3dspaceperception AT mamoruminami growingneuralgaswithdifferenttopologiesfor3dspaceperception |