Energy and Environment Seminar Series: Conserving energy when flying robots

Graduate student Haijie Zhang recently presented his research on “Vision-based aggressive perching for aerial robots” at the Energy Institute’s Energy and Environment Seminar Series.

Haijie Zhang poses with one of his aerial robot’s in his work space at the Energy Institute.

Right now, the longest time a robot can fly is around a half hour, so training robots to land and “perch” would allow the robot to use less energy and stay out in flight longer. Haijie’s aerial robots use video cameras to do minimum time perching, which has distinct advantages over other ways of programing a robot to fly. For instance, vision-based robots can better navigate a busy environment, like a shopping mall, and are able to detect objects like a package.

This research has real work applications and could be used to prolong aerial robots security surveillance or to help deliver packages.

If you’d like to learn more about Haijie’s work please email him at

An example of one of the aerial robots Haijie is working with for his research.