Monitoring and predicting water quality poses a significant challenge since sources of fresh water and contaminants come in from huge areas of land and waterways. Further, the source of pollution can change quickly during and after rainfall events. Characterizing large-scale and quickly changing water systems remains a critical bottleneck that inhibits understanding of transport processes and the development of effective management plans to address water quality issues. Fixed sensors tend to lack the versatility to directly detect contaminants of interest, are expensive, and only monitor a single location. Consequently, there is still a strong reliance upon manual “grab-sampling” within hydrologic and aquatic ecology applications. At best, this reliance is expensive, inconvenient and presents safety risks to personnel involved (e.g. when samples must be taken at night). At worst, manual sampling results in datasets that cannot answer many questions of interest due to limitations of temporal and spatial resolution in the sampling strategy or the inaccessibility of sites (e.g. canyons).
Goals and Objectives
This research develops an aerial water sampling system that can be quickly and safely deployed to reach varying and hard to access locations, that integrates with existing static sensors, and that is adaptable to a wide range of scientific goals. The capability to obtain quick samples over large areas will lead to a dramatic increase in the understanding of critical natural resources. This research will enable better interactions between non-expert operators and robots by using semi-autonomous systems to detect faults and unknown situations to ensure safety of the operators and environment. To achieve this we will:
- Create an aerial water sampling system that will have a dramatic impact in the field of aquatic ecology, reducing costs and making it possible to test new hypotheses as sampling and deployment time is reduced by orders of magnitude.
- Develop techniques to synthesize and optimize adaptive autonomy systems that complement engineered ones by adding robustness in the presence of unexpected events, including those caused by environmental conditions and evolving relationships between scientist and robot.
- Integrate with existing networks of ground and water sensors, minimizing the cost of operation by extending the life of sensors through UAV-based wireless recharging and providing easier access to the data through automated data muling.
- Deploy the system with our team of water scientists, incrementally incorporating the technical advances and assessing the cost-effectiveness and capabilities of this co-robot system through field studies conducted in watersheds and lakes in Nebraska and California.
- Educate scientists, students, and the public on the impact of using robotic instruments for science through a series of videos, articles, and courses.
This project is partially funded as part of the National Robotics Initiative through the USDA National Institute of Food and Agriculture (grant #2013-67021-20947). The project is lead by UNL with collaborators at UC Berkeley. At UNL, current and past people working on this project include:
- Carrick Detweiler (PI, Computer Science and Engineering)
- Sebastian Elbaum (Co-PI, Computer Science and Engineering)
- Amy Burgin (Co-PI, School of Natural Resources)
- Matt Waite (Co-PI, College of Journalism and Mass Communications)
- John-Paul Ore (Computer Science and Engineering)
- James Higgins (Mechanical and Materials Engineering)
- Dave Anthony (Computer Science and Engineering)
- Adam Taylor (Computer Science and Engineering)
- Emily Waring (School of Natural Resources)
- Valerie Schoepfer (School of Natural Resources)
- Megan Hoyt (College of Journalism and Mass Communications)
- Tony Papousek (College of Journalism and Mass Communications)
And at the University of California, Berkeley: