Intelligent Propulsion System for Water-Borne Sensors, 20-R8431
Inclusive Dates: 12/13/13 – 12/13/14
Background — Original work on developing water-borne sensor (WBS) systems that utilize sensors for mapping the environment was undertaken at Southwest Research Institute in 2007. Past SwRI-funded research and development led to externally funded projects with the U.S. Army Corps of Engineers and the Federal Highway Administration. Current WBS technology consists of a self-contained sensor platform, equipped with ultrasonic sensors, that uses advanced data processing techniques to produce three-dimensional maps of cavities. The SwRI research team advanced the capabilities of the water-borne sensors by incorporating the following improvements:
- propulsion system
- improved sensor packaging
- radiation-hardened sensors
- remedial navigation system
- refined data communication capabilities
- improved data display capabilities
SwRI researchers are currently working to design and develop a next-generation WBS system with propulsion and navigation capabilities to enhance the utility of the technology. SwRI has already received expressions of interest from a broad variety of potential clients with different application requirements including routine maintenance inspections; high-risk, critical surveys (characterization of nuclear reactor containment vessels); and earth science exploratory surveys.
The overall objective of this project was to develop, demonstrate, and evaluate the use of an inexpensive propulsion and navigation system for enhanced utility of the WBS system. A propulsion system framework with navigation capabilities was planned for implementation on a WBS platform to which other environmental sensors (e.g., cameras, dosimeters, temperature probes) could be added. These combined capabilities provide for mobility and additional sensing capabilities that broaden the overall applicability of WBS. The new WBS is capable of collecting an array of data in both confined and large spaces. Adding a quasi-intelligent guidance system to the propelled sensor expands the range of applications of the WBS.
Approach — The specific research approach was to assemble and field-test a WBS with propulsion and a guidance system. The mobility platform is capable of delivering the sensor platform in a scientifically targeted manner to meet project objectives. Requirements and goals constrained by realistic needs were used to first select the mobility platform and then subsequently to evaluate its efficacy. A removable, off-the-shelf, propulsion system was added to the WBS. Initially, a tethered control system was implemented to allow an operator to remotely control the probe with real-time feedback. Current prototypes are now capable of non-tethered propulsion and navigation. Navigation algorithms couple input from ultrasonic and payload sensors (video, pressure, temperature, etc.) to make decisions about navigation. This technique allows an efficient physical search of a subject area by intelligently choosing which areas to map with greater complexity. The ability to simultaneously search a wide area and then focus sensor attention to areas that warrant more detailed evaluation through these techniques will speed reaction time, efficiency, and effectiveness. This approach balances competing constraints (e.g., size, sensitivity, cost) while remaining flexible enough to fit to a broad range of applications.
Accomplishments — Construction of the prototype propulsion platform, a remotely operated vehicle, has been completed. The prototype has been tested in a water-filled tank. An autonomy plugin to the remotely operated vehicle platform has been developed. The goal of this autonomy plugin is to generate a global map of the environment, make decisions on where to explore, and be able to return the WBS to the point of deployment. The autonomy plugin uses simulated ultrasonic sensor data to map out and navigate a virtual world. Existing ultrasonic imaging capabilities were repackaged and attached to the propulsion unit. A general user interface has been implemented and basic two-dimensional mapping and return path planning has been implemented. Programmatic control of the remotely operated vehicle based on output from navigation algorithms has been accomplished. Current prototypes have greatly improved reliability and capacity and are available for advanced field testing and demonstration.