Visual remote control

Mechanical actuators are most often used to run predefined programs in industrial production systems. But some of the limitations of these actuators can be overcome by using artificial intelligence and more particularly computer vision and deep learning. The control of a robot by real-time image analysis and processing could be used to limit unnecessary movements of operators in the industry. But it can also be seen as a tool to assist people with reduced mobility for more autonomy.

​

There are 3 main steps to operate the robot:

​

1. The camera detects the sign proposed by the system operator.

2. The sign is decoded using a trained neural network and the linked program is transmitted to the robot.

3. The program is executed and allows the movement of the arm according to the pre-programmed scheme.

​