Engineers Develop Optimal Path Predictor for Underwater Robots system

Engineers at Massachusetts Institute of Technology developed a software to predict the fastest way for underwater vehicles to get to one point to another, and it might not always be through a straight line.

The idea for a path predictor for AUVs is not new, but what the older systems lack were “methodology and algorithm,” says Pierre Lermusiaux, project leader. “Because ocean environments are so complex, what was missing was the integration of ocean prediction, ocean estimation, control and optimization.”

Tests done to measure the software’s accuracy has been done in challenging ocean environments including the Philippines, a country with thousands of islands, discontinuing shorelines, and restless currents. The test was done on a fleet of 1,000 AUVs that were to accomplish various goals.The breakthrough development, to be presented this May 2012 at the International Conference on Robotics and Animation, lets controllers of automated underwater vehicles (AUVs) to figure out the best route to take, considering factors like underwater terrain, ocean currents, and shorelines as well as travel goals like minimum use of power and energy and extensive data collection.