A multiple-predictor approach to human motion prediction
The ability to accurately predict human motion is imperative for any human-robot interaction application in which the human and robot interact in close proximity to one another. Although a variety of human motion prediction approaches have already been developed, they are often designed for specific...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2018
|
| Online Access: | http://hdl.handle.net/1721.1/116050 https://orcid.org/0000-0002-1761-221X https://orcid.org/0000-0003-1338-8107 |
| _version_ | 1826196472021909504 |
|---|---|
| author | Lasota, Przemyslaw Andrzej Shah, Julie A |
| author2 | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
| author_facet | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Lasota, Przemyslaw Andrzej Shah, Julie A |
| author_sort | Lasota, Przemyslaw Andrzej |
| collection | MIT |
| description | The ability to accurately predict human motion is imperative for any human-robot interaction application in which the human and robot interact in close proximity to one another. Although a variety of human motion prediction approaches have already been developed, they are often designed for specific types of tasks or motions, and thus do not generalize well. Furthermore, it is not always obvious which of these methods is appropriate for a given task, making human motion prediction difficult to implement in practice. We address this problem by introducing a multiple-predictor system (MPS) for human motion prediction. In our approach, the system learns directly from task data in order to determine the most favorable parameters for each implemented prediction method and which combination of these predictors to use. Our implementation consists of three complementary methods: velocity-based position projection, time series classification, and sequence prediction. We describe the process of forming the MPS and our evaluation of its performance against the individual methods in terms of accuracy of predictions of human position over a range of look-ahead time values. We report that our method leads to a reduction in mean error of 18.5%, 28.9%, and 37.3% when compared with the three individual methods, respectively. |
| first_indexed | 2024-09-23T10:27:30Z |
| format | Article |
| id | mit-1721.1/116050 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T10:27:30Z |
| publishDate | 2018 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1160502022-09-30T21:17:40Z A multiple-predictor approach to human motion prediction Lasota, Przemyslaw Andrzej Shah, Julie A Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Lasota, Przemyslaw Andrzej Shah, Julie A The ability to accurately predict human motion is imperative for any human-robot interaction application in which the human and robot interact in close proximity to one another. Although a variety of human motion prediction approaches have already been developed, they are often designed for specific types of tasks or motions, and thus do not generalize well. Furthermore, it is not always obvious which of these methods is appropriate for a given task, making human motion prediction difficult to implement in practice. We address this problem by introducing a multiple-predictor system (MPS) for human motion prediction. In our approach, the system learns directly from task data in order to determine the most favorable parameters for each implemented prediction method and which combination of these predictors to use. Our implementation consists of three complementary methods: velocity-based position projection, time series classification, and sequence prediction. We describe the process of forming the MPS and our evaluation of its performance against the individual methods in terms of accuracy of predictions of human position over a range of look-ahead time values. We report that our method leads to a reduction in mean error of 18.5%, 28.9%, and 37.3% when compared with the three individual methods, respectively. United States. National Aeronautics and Space Administration. Space Technology Research Fellowship 2018-06-04T14:57:09Z 2018-06-04T14:57:09Z 2017-07 2018-04-10T17:16:01Z Article http://purl.org/eprint/type/ConferencePaper 978-1-5090-4633-1 http://hdl.handle.net/1721.1/116050 Lasota, Przemyslaw A., and Julie A. Shah. “A Multiple-Predictor Approach to Human Motion Prediction.” 2017 IEEE International Conference on Robotics and Automation (ICRA) (May 2017). https://orcid.org/0000-0002-1761-221X https://orcid.org/0000-0003-1338-8107 http://dx.doi.org/10.1109/ICRA.2017.7989265 2017 IEEE International Conference on Robotics and Automation (ICRA) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT Web Domain |
| spellingShingle | Lasota, Przemyslaw Andrzej Shah, Julie A A multiple-predictor approach to human motion prediction |
| title | A multiple-predictor approach to human motion prediction |
| title_full | A multiple-predictor approach to human motion prediction |
| title_fullStr | A multiple-predictor approach to human motion prediction |
| title_full_unstemmed | A multiple-predictor approach to human motion prediction |
| title_short | A multiple-predictor approach to human motion prediction |
| title_sort | multiple predictor approach to human motion prediction |
| url | http://hdl.handle.net/1721.1/116050 https://orcid.org/0000-0002-1761-221X https://orcid.org/0000-0003-1338-8107 |
| work_keys_str_mv | AT lasotaprzemyslawandrzej amultiplepredictorapproachtohumanmotionprediction AT shahjuliea amultiplepredictorapproachtohumanmotionprediction AT lasotaprzemyslawandrzej multiplepredictorapproachtohumanmotionprediction AT shahjuliea multiplepredictorapproachtohumanmotionprediction |