Remote Control for Mobile Robots Using Gestures Captured by the RGB Camera and Recognized by Deep Learning Techniques

Mobile robotics tends to move forward with autonomous systems, but in some cases, it is not possible to use this strategy. Non-autonomous robots are controlled by some human-machine interface, mostly industrial equipment, which are often expensive. Controlling the robot using some fixed control attaches the operator to the equipment, causing it to lose mobility and agility, not to mention the lack of control accuracy. This paper proposes to solve this problem by using computer vision techniques to develop a gesture control interface. From the movement of the left index finger, the operator can send the robot linear speeds. By moving the finger of the right hand, angular movements are sent to the robot. This strategy has proven practical and easy for the operator to learn, as well as avoiding operator-attached cables and is ideal for situations that pose a risk.