Purdue underwater prototype demonstrates three modes in pool tests

During underwater pool tests, Purdue University students and researcher Yu She watched a compact robot demonstrate three ways of moving: drifting with the water, gliding through buoyancy-driven motion and traveling under thruster power. The patent-pending prototype is designed to switch between those modes during a single deployment, rather than remain tied to one form of locomotion.

The mechanism combines buoyancy regulation, an internal mass that shifts to control pitch, foldable hydrodynamic wings and a propulsion-and-steering system. In glider mode, the robot changes its buoyancy, shifts the internal mass and deploys its wings so vertical movement becomes forward motion. When direct control is needed, the thruster takes over.

Each mode addresses a different weakness. A drifting vehicle can conserve energy and track currents but may be unable to reach a target or recover from a bad path. A glider can travel with low power but may struggle in confined environments. A thruster-driven vehicle can maneuver directly, while consuming power quickly enough to shorten a mission. Purdue says those trade-offs can mean shorter missions, less data, greater risk of losing or trapping a robot and the need for several specialized vehicles.

The practical payoff is flexibility for long underwater missions: researchers could use drifting for energy-efficient environmental tracking, gliding for longer-range travel and propulsion for repositioning or escape. Potential applications include ocean and lake monitoring, current and flow mapping, inspection of underwater structures, search and rescue, and distributed data collection. No measurements of endurance, energy savings or operating range are reported yet, so those benefits remain targets rather than demonstrated results.

The tests confirmed that the main subsystems worked together in water, including wing folding, buoyancy-regulated ascent and descent, thruster-assisted translation and glider motion. The next stage is quantitative performance testing and field readiness. She’s team also plans closed-loop controls—software that uses mission goals and environmental feedback—to decide when the robot should drift, glide or propel itself; for now, switching remains manual.