Robot submarines are pretty good at finding their own way around but they still need a helpful human to guide them through twisting underwater ravines and channels, the poor things.
Not for much longer. A new guidance system will let autonomous underwater vehicles (AUVs) snap photos along even the most treacherous reaches of sea floor.
AUVs can already navigate in open water by matching their depth to an existing terrain map. Sarah Houts of the Aerospace Robotics Laboratory, at Stanford University, California, worked with researchers at the Monterey Bay Aquarium Research Institute in Moss Landing, also in California, to take the current technology one step further by computing flight paths around and through underwater obstacles.
The software tweak will allow underwater robots to autonomously take pictures of hazardous locations where only remote-controlled robots have gone before. The team ran a successful field test in Monterey Bay earlier this month and expects the system will be ready to be tried for real sometime next year.
Across the Pacific, Australian scientists are teaching their autonomous underwater robot Sirius a new trick as well.
Researchers at the University of Sydney have developed an system to tell the difference between kelp and other species of plant life on the ocean floor off the south-east coast of Tasmania. Their results mean that future AUVs trained on a number of dive missions could be used to document the variety and abundance of plant and animal life in the world's oceans - all without humans needing to be in the loop. The research will be presented at theAustralian Conference on Robotics and Automation next month.