6D Pose Estimation for Subsea Intervention in Turbid Waters
Manipulation tasks on subsea instalments require extremely precise detection and localization of objects of interest. This problem is referred to as “pose estimation”. In this work, we present a framework for detecting and predicting 6DoF pose for relevant objects (fish-tail, gauges, and valves) on...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/19/2369 |
| _version_ | 1797516629663285248 |
|---|---|
| author | Ahmed Mohammed Johannes Kvam Jens T. Thielemann Karl H. Haugholt Petter Risholm |
| author_facet | Ahmed Mohammed Johannes Kvam Jens T. Thielemann Karl H. Haugholt Petter Risholm |
| author_sort | Ahmed Mohammed |
| collection | DOAJ |
| description | Manipulation tasks on subsea instalments require extremely precise detection and localization of objects of interest. This problem is referred to as “pose estimation”. In this work, we present a framework for detecting and predicting 6DoF pose for relevant objects (fish-tail, gauges, and valves) on a subsea panel under varying water turbidity. A deep learning model that takes 3D vision data as an input is developed, providing a more robust 6D pose estimate. Compared to the 2D vision deep learning model, the proposed method reduces rotation and translation prediction error by (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>−</mo><mo>Δ</mo><msup><mn>0.39</mn><mo>∘</mo></msup></mrow></semantics></math></inline-formula>) and translation (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>−</mo><mo>Δ</mo><mn>6.5</mn></mrow></semantics></math></inline-formula> mm), respectively, in high turbid waters. The proposed approach is able to provide object detection as well as 6D pose estimation with an average precision of 91%. The 6D pose estimation results show <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mn>2.59</mn><mo>∘</mo></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>6.49</mn></mrow></semantics></math></inline-formula> cm total average deviation in rotation and translation as compared to the ground truth data on varying unseen turbidity levels. Furthermore, our approach runs at over 16 frames per second and does not require pose refinement steps. Finally, to facilitate the training of such model we also collected and automatically annotated a new underwater 6D pose estimation dataset spanning seven levels of turbidity. |
| first_indexed | 2024-03-10T07:03:32Z |
| format | Article |
| id | doaj.art-230cb0692cb742bb9a60f5685b217678 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T07:03:32Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-230cb0692cb742bb9a60f5685b2176782023-11-22T15:56:43ZengMDPI AGElectronics2079-92922021-09-011019236910.3390/electronics101923696D Pose Estimation for Subsea Intervention in Turbid WatersAhmed Mohammed0Johannes Kvam1Jens T. Thielemann2Karl H. Haugholt3Petter Risholm4SINTEF Digital, Smart Sensor Systems, 0373 Oslo, NorwaySINTEF Digital, Smart Sensor Systems, 0373 Oslo, NorwaySINTEF Digital, Smart Sensor Systems, 0373 Oslo, NorwaySINTEF Digital, Smart Sensor Systems, 0373 Oslo, NorwaySINTEF Digital, Smart Sensor Systems, 0373 Oslo, NorwayManipulation tasks on subsea instalments require extremely precise detection and localization of objects of interest. This problem is referred to as “pose estimation”. In this work, we present a framework for detecting and predicting 6DoF pose for relevant objects (fish-tail, gauges, and valves) on a subsea panel under varying water turbidity. A deep learning model that takes 3D vision data as an input is developed, providing a more robust 6D pose estimate. Compared to the 2D vision deep learning model, the proposed method reduces rotation and translation prediction error by (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>−</mo><mo>Δ</mo><msup><mn>0.39</mn><mo>∘</mo></msup></mrow></semantics></math></inline-formula>) and translation (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>−</mo><mo>Δ</mo><mn>6.5</mn></mrow></semantics></math></inline-formula> mm), respectively, in high turbid waters. The proposed approach is able to provide object detection as well as 6D pose estimation with an average precision of 91%. The 6D pose estimation results show <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mn>2.59</mn><mo>∘</mo></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>6.49</mn></mrow></semantics></math></inline-formula> cm total average deviation in rotation and translation as compared to the ground truth data on varying unseen turbidity levels. Furthermore, our approach runs at over 16 frames per second and does not require pose refinement steps. Finally, to facilitate the training of such model we also collected and automatically annotated a new underwater 6D pose estimation dataset spanning seven levels of turbidity.https://www.mdpi.com/2079-9292/10/19/2369subseapose estimationobject detection3D visionAUVROV |
| spellingShingle | Ahmed Mohammed Johannes Kvam Jens T. Thielemann Karl H. Haugholt Petter Risholm 6D Pose Estimation for Subsea Intervention in Turbid Waters Electronics subsea pose estimation object detection 3D vision AUV ROV |
| title | 6D Pose Estimation for Subsea Intervention in Turbid Waters |
| title_full | 6D Pose Estimation for Subsea Intervention in Turbid Waters |
| title_fullStr | 6D Pose Estimation for Subsea Intervention in Turbid Waters |
| title_full_unstemmed | 6D Pose Estimation for Subsea Intervention in Turbid Waters |
| title_short | 6D Pose Estimation for Subsea Intervention in Turbid Waters |
| title_sort | 6d pose estimation for subsea intervention in turbid waters |
| topic | subsea pose estimation object detection 3D vision AUV ROV |
| url | https://www.mdpi.com/2079-9292/10/19/2369 |
| work_keys_str_mv | AT ahmedmohammed 6dposeestimationforsubseainterventioninturbidwaters AT johanneskvam 6dposeestimationforsubseainterventioninturbidwaters AT jenstthielemann 6dposeestimationforsubseainterventioninturbidwaters AT karlhhaugholt 6dposeestimationforsubseainterventioninturbidwaters AT petterrisholm 6dposeestimationforsubseainterventioninturbidwaters |