Robust tactile object recognition in open-set scenarios using Gaussian prototype learning
Tactile object recognition is crucial for effective grasping and manipulation. Recently, it has started to attract increasing attention in robotic applications. While there are many works on tactile object recognition and they also achieved promising performances in some applications, most of them a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-12-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2022.1070645/full |
| _version_ | 1797975563352145920 |
|---|---|
| author | Wendong Zheng Wendong Zheng Huaping Liu Huaping Liu Di Guo Di Guo Fuchun Sun Fuchun Sun |
| author_facet | Wendong Zheng Wendong Zheng Huaping Liu Huaping Liu Di Guo Di Guo Fuchun Sun Fuchun Sun |
| author_sort | Wendong Zheng |
| collection | DOAJ |
| description | Tactile object recognition is crucial for effective grasping and manipulation. Recently, it has started to attract increasing attention in robotic applications. While there are many works on tactile object recognition and they also achieved promising performances in some applications, most of them are usually limited to closed world scenarios, where the object instances to be recognition in deployment are known and the same as that of during training. Since robots usually operate in realistic open-set scenarios, they inevitably encounter unknown objects. If automation systems falsely recognize unknown objects as one of the known classes based on the pre-trained model, it can lead to potentially catastrophic consequences. It motivates us to break the closed world assumption and to study tactile object recognition in realistic open-set conditions. Although several open-set recognition methods have been proposed, they focused on visual tasks and may not be suitable for tactile recognition. It is mainly due to that these methods do not take into account the special characteristic of tactile data in their models. To this end, we develop a novel Gaussian Prototype Learning method for robust tactile object recognition. Particularly, the proposed method converts feature distributions to probabilistic representations, and exploit uncertainty for tactile recognition in open-set scenarios. Experiments on the two tactile recognition benchmarks demonstrate the effectiveness of the proposed method on open-set tasks. |
| first_indexed | 2024-04-11T04:37:32Z |
| format | Article |
| id | doaj.art-fffe0dc71df2456fb9f91e11f77922df |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-11T04:37:32Z |
| publishDate | 2022-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-fffe0dc71df2456fb9f91e11f77922df2022-12-28T11:07:43ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2022-12-011610.3389/fnins.2022.10706451070645Robust tactile object recognition in open-set scenarios using Gaussian prototype learningWendong Zheng0Wendong Zheng1Huaping Liu2Huaping Liu3Di Guo4Di Guo5Fuchun Sun6Fuchun Sun7Department of Computer Science and Technology, Tsinghua University, Beijing, ChinaState Key Laboratory of Intelligent Technology and Systems, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, ChinaDepartment of Computer Science and Technology, Tsinghua University, Beijing, ChinaState Key Laboratory of Intelligent Technology and Systems, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, ChinaDepartment of Computer Science and Technology, Tsinghua University, Beijing, ChinaState Key Laboratory of Intelligent Technology and Systems, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, ChinaDepartment of Computer Science and Technology, Tsinghua University, Beijing, ChinaState Key Laboratory of Intelligent Technology and Systems, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, ChinaTactile object recognition is crucial for effective grasping and manipulation. Recently, it has started to attract increasing attention in robotic applications. While there are many works on tactile object recognition and they also achieved promising performances in some applications, most of them are usually limited to closed world scenarios, where the object instances to be recognition in deployment are known and the same as that of during training. Since robots usually operate in realistic open-set scenarios, they inevitably encounter unknown objects. If automation systems falsely recognize unknown objects as one of the known classes based on the pre-trained model, it can lead to potentially catastrophic consequences. It motivates us to break the closed world assumption and to study tactile object recognition in realistic open-set conditions. Although several open-set recognition methods have been proposed, they focused on visual tasks and may not be suitable for tactile recognition. It is mainly due to that these methods do not take into account the special characteristic of tactile data in their models. To this end, we develop a novel Gaussian Prototype Learning method for robust tactile object recognition. Particularly, the proposed method converts feature distributions to probabilistic representations, and exploit uncertainty for tactile recognition in open-set scenarios. Experiments on the two tactile recognition benchmarks demonstrate the effectiveness of the proposed method on open-set tasks.https://www.frontiersin.org/articles/10.3389/fnins.2022.1070645/fulltactile perceptionobject recognitionopen-set recognitionGaussian prototype learningtactile object recognition |
| spellingShingle | Wendong Zheng Wendong Zheng Huaping Liu Huaping Liu Di Guo Di Guo Fuchun Sun Fuchun Sun Robust tactile object recognition in open-set scenarios using Gaussian prototype learning Frontiers in Neuroscience tactile perception object recognition open-set recognition Gaussian prototype learning tactile object recognition |
| title | Robust tactile object recognition in open-set scenarios using Gaussian prototype learning |
| title_full | Robust tactile object recognition in open-set scenarios using Gaussian prototype learning |
| title_fullStr | Robust tactile object recognition in open-set scenarios using Gaussian prototype learning |
| title_full_unstemmed | Robust tactile object recognition in open-set scenarios using Gaussian prototype learning |
| title_short | Robust tactile object recognition in open-set scenarios using Gaussian prototype learning |
| title_sort | robust tactile object recognition in open set scenarios using gaussian prototype learning |
| topic | tactile perception object recognition open-set recognition Gaussian prototype learning tactile object recognition |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2022.1070645/full |
| work_keys_str_mv | AT wendongzheng robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT wendongzheng robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT huapingliu robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT huapingliu robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT diguo robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT diguo robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT fuchunsun robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning AT fuchunsun robusttactileobjectrecognitioninopensetscenariosusinggaussianprototypelearning |