Gesture Controlled Robotic System using Monochromatic Camera and Array of IR-Sensors

Technology is a never ending process. To be able to design a product using the current technology that will be beneficial to the lives of others is a huge contribution to the community. The project area is in robotics. This abstract presents the design and implementation of a low cost but yet flexible and robust 3D gesture controlled Robotic Arm. Imagine things that are done in a computer just like in real life. Natural movements replaced all those clicks, taps, drags and drops. Hand gesture recognition has been a very active research topic in recent years with the goal of interpreting human gestures via mathematical algorithms and with motivating applications such as robot control, and sign language interpretation. Motivates for this work is a robot navigation problem, in which controlling a robot by hand pose signs given by a human. In this project designing a robotic arm that can be moved upward, downward, left and right depending on human hand gestures. The Robotic Arm is mounted over a movable platform. The human gestures will be captured via 3D sensor. After that images will be processed to recognize the gestures and take action accordingly. This processing is done in a PC/Laptop with a GUI. The information (in the form of Euler angles row, pitch and yaw) is then transmitted to the robotic arm section using a wireless module. The embedded system on the robotic arm computes the pulse width of the PWM signals for the servos that are required to move the arm to the position indicated by the Euler angles. The Pitch and Yaw angles to control 5 servos mounted perpendicularly. 3D SENSOR: Using two monochromatic IR cameras with three infrared LEDs, the device observes a roughly hemispherical area, to a distance of about 1 meter. Synthesizing 3D position data by comparing the 2D frames generated by the two cameras.