| Common teaching processes like the teach pendant | | | | recognized specific movements, e.g., a swing to the |
| control most industrial robots. Accelerometer-based | | | | right. Slight jittery movements will not produce any |
| gesture recognition has become popular since the last | | | | meaningful specific patterns. Advantage of such |
| ten years. Advantages of the accelerometer are low | | | | control system is its programmable, repeatable robotic |
| to moderate in cost, and small in size. Wireless sensor | | | | arm movements. |
| system (using 'Wii' remote; accelerometer-based) can | | | | Two 'Wii Nunchucks' (3-axis accelerometer) can be |
| be used to control a robotic arm. This arm is designed | | | | attached to control a robotic arm. One is at the palm |
| to operate in similar motions to the human arm. | | | | of the user (held), another at his elbow. By this, at the |
| Calibration of these teaching methods require big | | | | two end points of the arm (palm and elbow), |
| amount of time. They can require extensive human | | | | differentiation between roll and pitch wrist and shoulder |
| intervention. Adjustments are done based on | | | | joint movements are allowed. Digital motion data is |
| accelerometer data. When teaching a walking robot | | | | then sent to (from both 'Wii Nunchucks') to 'Arduino' |
| (bipedal), feedback is useful to learn new walk | | | | micro controller. The 'Arduino' micro controller only |
| parameters. | | | | collects and processes acceleration data. A separate |
| 3-axis accelerometer is attached to various parts of a | | | | micro controller then controls the servo motors. |
| robotic arm. Sensor (accelerometer) is configured to | | | | 'Arduino', mentioned many times in this article is a |
| recognize gestures (changing arm positions). Only | | | | physical computing platform that is currently taking the |
| specific movements can activate the robotic arm to | | | | world by storm. |
| function. Accelerometer data is analyzed to fit in the | | | | |