How cool will it be to control a flying drone by gestures? Well, MIT CSAIL researchers have developed a drone control method that uses muscle sensors and motion detection for robot control.
The team developed a drone control system with a feature called ‘Conduct-A-Bot’ or simply a ‘muscle control system’. The pilot could use gestures to navigate the drone instead of traditional methods like joysticks and touch screens.
A video showing the visual representation of how they integrated the muscle-control feature to control the device is shown below. Their newest drone can be fully controlled using hand-arm gestures to navigate through a series of rings.
“We envision a world in which machines help people with cognitive and physical work, and to do so, they adapt to people rather than the other way around,” says Daniela Rus, MIT professor and director of CSAIL, and co-author on a paper about the system.
For this to work, the pilot needs to wear electromyography (EMG) and motion sensors on the biceps, triceps, and forearms to produce signals that could result in drone movements.
Here is an example of few features that could be used – stiffening the upper arm to stop the robot, clenching fist to move forward, rotate clockwise, and counter-clockwise to rotate robot by using forearm gyroscope and waving a hand up or down to move the robot vertically or sideways.
The researchers used a Parrot Bebop 2 drone and it responded to 82 percent of 1,500+ gestures already. The new tech is not fully done, there are some refinements to be done, as per the researchers.
It will be definitely helpful in the future to better make use of drone and robotics technologies, and helpful for remote exploration