MIT develops a system 'Conduct-a-Bot' that controls the movements of robots with muscles
Computer Science & Artificial Intelligence Laboratory (CSAIL) of Massachusetts Institute of Technology (MIT) is a system that enables you to control robots using 'myoelectric potential signals' generated in the body when moving muscles `` Conduct-a -Bot 'was developed.
Controlling drones and other robots with gestures | MIT CSAIL
MIT muscle-control system for drones lets a pilot use gestures for accurate and specific navigation | TechCrunch
The MIT CSAIL-developed muscle control system 'Conduct-a-Bot' enables fine and complete control of the robot with only hand and arm gestures. In the following movie, you can check how to fly the drone using Conduct-a-Bot, and you can see that you can control the drone's movement quite finely with only gestures.
Controlling Drone with Gestures-YouTube
When a man wearing a red hat moves his right arm ...
The drone glows red ...
CSAIL has developed a Conduct-a-Bot that can control the drone with only hand and arm gestures.
Attached to the forearm.
In Conduct-a-Bot, the algorithm developed by CSAIL processes the myoelectric potential signal received from the sensor and detects the gesture.
The gestures for operating the drone are as follows. When you twist your hand, the drone changes direction.
Raise your fist and the drone will surface.
When you grasp your fist, the drone advances.
Conduct-a-Bot does not require any 'offline adjustment' or 'training data' for each user, and can easily introduce robot gesture operations.
The drone stops when you put effort into your biceps and triceps.
When testing Conduct-a-Bot using a drone, we succeeded in correctly identifying 81.6% of gestures. At the time of article creation, only eight gestures are set: forward, backward, move right, move left, ascend, descend, change direction, stop.
Conduct-a-Bot can also be extended to other robots and applications, 'to facilitate human-robot interaction,' said CSAIL.
According to CSAIL, Conduct-a-Bot is a system that can control the drone with only a simple wearable sensor and algorithm, 'to eliminate the barriers such as' difficult to operate 'felt by casual users who interact with robots. It is a 'developed system'.