Scientists have developed an artificial hummingbird weighing 12 grams that can mimic the extreme maneuvers of real hummingbirds thanks to training based on machine learning.
The 'hummingbird robot' has two actuators and can therefore move both wings separately. According to the scientists, the fluttering happens at a frequency close to that of real hummingbirds: 40Hz. The artificial bird itself weighs twelve grams but can lift a weight of 27 grams. The model that the researchers at Purdue University have developed still uses power via a cable, but in the long run the model must be able to hold a battery due to its capacity.
The bird also cannot 'see' yet, but the electrical currents that are generated when touching environmental objects can be used to detect surfaces. The researchers chose the hummingbird because of the special aerodynamic properties of this animal and the possibilities it has to perform extreme maneuvers, such as being able to turn 180 degrees with a minimum number of blows with the wings.
The researchers used a model-based non-linear system for regular flying, but left the system without a model for more extreme acrobatics reinforced train based on simulations. The bird learned more on the basis of trial and error.
The research team also developed smaller insect-like robots with a weight of up to 1 gram that learned to move based on the algorithms. The wings of those models had to move at a higher frequency. The ultimate goal is to arrive at robots that can be used in various circumstances, such as during rescue operations.
The research is under the title Learning Extreme Hummingbird Maneuvers on Flapping Wing Robots published.