We use essential cookies for the proper functioning of the website and additional ones to make interaction with the site as convenient as possible. It helps us personalize your user experience as well as obtain analytical information to improve the service.
If you agree to accept all cookies, click "Accept all"; if not, click "Only essential". To learn more, view the Cookie Policy.
Engineers taught AI to control drones (Video)
US engineers have taught AI to fly drones in unfamiliar or tight spaces. The devices will not touch each other even in a small room.
Researchers at the California Institute of Technology have unveiled a new way to fly drones in unfamiliar space. They have developed a machine learning algorithm that allows even multiple devices to autonomously navigate themselves in tight and unfamiliar spaces. The system gives each drone a certain degree of independence, which allows it to adapt to a changing environment.
Instead of relying on maps or routes from other drones, the new model allows each vehicle to independently navigate a given space, even if it is coordinating with others. This decentralized model helps drones to improvise and makes it easier to control drones, since the computation is distributed among many robots.
An optional tracking controller helps drones compensate for aerodynamic interactions. In preliminary tests, the controller turned out to be more efficient than analogues.
This technique can be used in search and rescue operations where drones can safely sweep areas in flocks, while autonomous units can minimize traffic jams and collisions. The researchers warn that the method needs to be tested in laboratories, but in a couple of years this model could be commercialized.
[video = "z9LjSfLfG6c"]
Researchers at the California Institute of Technology have unveiled a new way to fly drones in unfamiliar space. They have developed a machine learning algorithm that allows even multiple devices to autonomously navigate themselves in tight and unfamiliar spaces. The system gives each drone a certain degree of independence, which allows it to adapt to a changing environment.
Instead of relying on maps or routes from other drones, the new model allows each vehicle to independently navigate a given space, even if it is coordinating with others. This decentralized model helps drones to improvise and makes it easier to control drones, since the computation is distributed among many robots.
An optional tracking controller helps drones compensate for aerodynamic interactions. In preliminary tests, the controller turned out to be more efficient than analogues.
This technique can be used in search and rescue operations where drones can safely sweep areas in flocks, while autonomous units can minimize traffic jams and collisions. The researchers warn that the method needs to be tested in laboratories, but in a couple of years this model could be commercialized.
[video = "z9LjSfLfG6c"]