Predicting object properties based on movement kinematics
Abstract In order to grasp and transport an object, grip and load forces must be scaled according to the object’s properties (such as weight). To select the appropriate grip and load forces, the object weight is estimated based on experience or, in the case of robots, usually by use of image recogni...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-11-01
|
| Series: | Brain Informatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40708-023-00209-4 |
| _version_ | 1797636619940921344 |
|---|---|
| author | Lena Kopnarski Laura Lippert Julian Rudisch Claudia Voelcker-Rehage |
| author_facet | Lena Kopnarski Laura Lippert Julian Rudisch Claudia Voelcker-Rehage |
| author_sort | Lena Kopnarski |
| collection | DOAJ |
| description | Abstract In order to grasp and transport an object, grip and load forces must be scaled according to the object’s properties (such as weight). To select the appropriate grip and load forces, the object weight is estimated based on experience or, in the case of robots, usually by use of image recognition. We propose a new approach that makes a robot’s weight estimation less dependent on prior learning and, thereby, allows it to successfully grasp a wider variety of objects. This study evaluates whether it is feasible to predict an object’s weight class in a replacement task based on the time series of upper body angles of the active arm or on object velocity profiles. Furthermore, we wanted to investigate how prediction accuracy is affected by (i) the length of the time series and (ii) different cross-validation (CV) procedures. To this end, we recorded and analyzed the movement kinematics of 12 participants during a replacement task. The participants’ kinematics were recorded by an optical motion tracking system while transporting an object, 80 times in total from varying starting positions to a predefined end position on a table. The object’s weight was modified (made lighter and heavier) without changing the object’s visual appearance. Throughout the experiment, the object’s weight (light/heavy) was randomly changed without the participant’s knowledge. To predict the object’s weight class, we used a discrete cosine transform to smooth and compress the time series and a support vector machine for supervised learning from the achieved discrete cosine transform parameters. Results showed good prediction accuracy (up to $$95\%$$ 95 % , depending on the CV procedure and the length of the time series). Even at the beginning of a movement (after only 300 ms), we were able to predict the object weight reliably (within a classification rate of $$88-94\%$$ 88 - 94 % ). |
| first_indexed | 2024-03-11T12:36:37Z |
| format | Article |
| id | doaj.art-6f3574791b3f429698ed70e6d82fa45b |
| institution | Directory Open Access Journal |
| issn | 2198-4018 2198-4026 |
| language | English |
| last_indexed | 2024-03-11T12:36:37Z |
| publishDate | 2023-11-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Brain Informatics |
| spelling | doaj.art-6f3574791b3f429698ed70e6d82fa45b2023-11-05T12:32:53ZengSpringerOpenBrain Informatics2198-40182198-40262023-11-0110111210.1186/s40708-023-00209-4Predicting object properties based on movement kinematicsLena Kopnarski0Laura Lippert1Julian Rudisch2Claudia Voelcker-Rehage3Department of Neuromotor Behavior and Exercise, Institute of Sport and Exercise Sciences, University of MünsterApplied Functional Analysis, Chemnitz University of TechnologyDepartment of Neuromotor Behavior and Exercise, Institute of Sport and Exercise Sciences, University of MünsterDepartment of Neuromotor Behavior and Exercise, Institute of Sport and Exercise Sciences, University of MünsterAbstract In order to grasp and transport an object, grip and load forces must be scaled according to the object’s properties (such as weight). To select the appropriate grip and load forces, the object weight is estimated based on experience or, in the case of robots, usually by use of image recognition. We propose a new approach that makes a robot’s weight estimation less dependent on prior learning and, thereby, allows it to successfully grasp a wider variety of objects. This study evaluates whether it is feasible to predict an object’s weight class in a replacement task based on the time series of upper body angles of the active arm or on object velocity profiles. Furthermore, we wanted to investigate how prediction accuracy is affected by (i) the length of the time series and (ii) different cross-validation (CV) procedures. To this end, we recorded and analyzed the movement kinematics of 12 participants during a replacement task. The participants’ kinematics were recorded by an optical motion tracking system while transporting an object, 80 times in total from varying starting positions to a predefined end position on a table. The object’s weight was modified (made lighter and heavier) without changing the object’s visual appearance. Throughout the experiment, the object’s weight (light/heavy) was randomly changed without the participant’s knowledge. To predict the object’s weight class, we used a discrete cosine transform to smooth and compress the time series and a support vector machine for supervised learning from the achieved discrete cosine transform parameters. Results showed good prediction accuracy (up to $$95\%$$ 95 % , depending on the CV procedure and the length of the time series). Even at the beginning of a movement (after only 300 ms), we were able to predict the object weight reliably (within a classification rate of $$88-94\%$$ 88 - 94 % ).https://doi.org/10.1186/s40708-023-00209-4KinematicsArm movementObject replacementPredictionClassificationPattern recognition |
| spellingShingle | Lena Kopnarski Laura Lippert Julian Rudisch Claudia Voelcker-Rehage Predicting object properties based on movement kinematics Brain Informatics Kinematics Arm movement Object replacement Prediction Classification Pattern recognition |
| title | Predicting object properties based on movement kinematics |
| title_full | Predicting object properties based on movement kinematics |
| title_fullStr | Predicting object properties based on movement kinematics |
| title_full_unstemmed | Predicting object properties based on movement kinematics |
| title_short | Predicting object properties based on movement kinematics |
| title_sort | predicting object properties based on movement kinematics |
| topic | Kinematics Arm movement Object replacement Prediction Classification Pattern recognition |
| url | https://doi.org/10.1186/s40708-023-00209-4 |
| work_keys_str_mv | AT lenakopnarski predictingobjectpropertiesbasedonmovementkinematics AT lauralippert predictingobjectpropertiesbasedonmovementkinematics AT julianrudisch predictingobjectpropertiesbasedonmovementkinematics AT claudiavoelckerrehage predictingobjectpropertiesbasedonmovementkinematics |