Researchers at Shanghai Jiaotong University recently developed a new obstacle avoidance system based on biology and computer vision that can improve the navigation of flying robots operating in a dynamic environment. It is based on how owls detect objects or other animals in their environment and avoid them.
“Although owls cannot move their eyes in any direction (similar to stereo cameras), they have a very flexible neck that can rotate 270 degrees. This allows them to quickly track objects even from behind, without moving their torso, ”the researchers write in their article.
To replicate the way owls move their eyes in different directions and detect both static and moving objects around them, the researchers installed a servo motor and a stereo camera on the quadcopter. In their design, the servo motor acts as the neck, and the stereo camera acts as the head. Due to the lightweight of the stereo camera, it can move much faster than the robot’s body, and its movements have practically no effect on the quality of the robot’s movements or the direction of flight.
The system uses a sensor planning algorithm to evaluate the usefulness of detecting objects in different directions and plans the angle by which its “head” (ie stereo camera) should rotate accordingly. Thus, the quadcopter constantly and actively senses its surroundings, quickly identifying obstacles that interfere with it.
In addition, the system monitors and predicts the trajectory of moving obstacles in its vicinity, adapting its movements to changes in the environment.
In the future, this system can be used to carry out various missions in urban to natural environments. In the future, scientists will try to create systems that reproduce the behavior of other animals.