SBIR Fast-Track: PIPELINE - Precision Inspection of Pressurized Environments using Long-term, Intelligent Navigation and Evaluation

The broader/commercial impact of this Small Business Innovation Research Phase I project will be to support more effective and accessible inspection of public drinking water infrastructure. Many water utilities across the United States struggle to monitor the condition of buried pipelines due to high inspection costs, aging systems, and the operational difficulty of accessing live, pressurized water mains. This project will develop and test a robotic inspection method that can move through water main without requiring shutdowns or excavation. The robot will collect sensor data including video, and location data that can help identify leaks, damage, and areas at risk of failure. By making detailed inspection data easier to obtain and interpret, the project will assist public utilities to plan repairs more efficiently, reduce water loss, and extend the lifespan of existing infrastructure. This work will contribute to national efforts to modernize aging public works and improve the resilience and sustainability of drinking water systems. This Small Business Innovation Research Phase I project will develop a fully autonomous, untethered robotic system capable of long-duration inspection within pressurized drinking water pipelines. The high-risk element of this work lies in enabling in-pipe navigation, localization, and defect detection in environments that are highly constrained, without reliance on real-time communication or human intervention. The core innovation involves adapting and integrating robotic methods to operate in a buried, water-filled environment that imposes sensing, energy, and control limitations. The intellectual merit of this project includes contribution to infrastructure inspection technologies by advancing visual-inertial simultaneous localization and mapping (SLAM) techniques specifically adapted for piped fluid-filled environments. The work will also develop new multi-sensor data fusion approaches for classification of defects based on video, acoustic, and pressure data, contributing to the broader field of autonomous inspection in constrained domains. The research will begin with algorithmic development and adaptation of SLAM and planning frameworks suitable for branched pipe systems. Sensor fusion methods will be tested using controlled data collected in testbeds simulating real-world pipe conditions. The robot's control and planning capabilities will be validated in an above-ground test network constructed to evaluate its ability to navigate, avoid obstacles, and map infrastructure geometry. The final phase will involve the design of an energy harvesting segment that uses in-flow hydroelectric generation and passive anchoring to facilitate long-duration deployments. Together, these efforts aim to produce a self-directed, persistent inspection platform capable of operating in live drinking water systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.