This workshop’s objective is to understand and share the current best practices and open challenges in active perception, i.e., decision-making and navigation based on perception inputs, specifically focusing in the extremely challenging marine domain. Advances towards this area are necessary to enable the study of the aquatic world – still largely unexplored – and consequently the understanding of how the environment is changing. This is recognized to be a priority by the United Nations Agenda 2020, which sets as a sustainable development goal, the oceans.
Achieving robust active perception is hard in the marine domain: underwater robots’ perception and communication are very limited. This workshop will explore the robust integration of perception tasks/inputs to decision modules, towards a holistic robotic system. This poses several open questions: e.g., how to best encode or represent the acquired data for control or decision units? How to cope with noisy perception data? Are there preferred perception sensors or methods depending on the end application?
This workshop aims to involve researchers from both academia and industry, working on different areas – spanning from mapping and sampling to intervention tasks (manipulation) – and its applications to real-world problems.
Topics of interest
We will explore how to make robust decisions based on low-quality and noisy perception inputs. We welcome novel contributions and work in progress related to active perception, including, but not limited to:
- Informative path planning and adaptive sampling.
- Environment modeling (bathymetry, 3D representation and mapping, SLAM).
- Sensor fusion and signal processing across different sensor domains (stereo vision,sonars, …).
- Human-robot cooperation.
- Machine learning practices in any of the above topics.
- Heavily tested systems in field trials and best practices for deployment and data man-agement (aquaculture, oil & gas, environmental monitoring, …).
In general, active perception is fundamental to increase the level of autonomy of roboticssystems and increase their impact for large-scale applications, e.g., ocean exploration andenvironmental monitoring. Advances in this challenging area will benefit other domains aswell – e.g., aerial and ground robots – when deployed in harsh environments.
|Workshop||June 4th, 2021 (full-day)|
Submission Link: Microsoft CMT3
This workshop is endorsed by the IEEE RAS Marine Robotics Technical Committee.