This paper discussed about the Wearable computing which is the process of recognizing human activities from body-worn sensors and about the methods which enhances the robustness of activity-recognition systems. Wearable computers are miniature electronic devices which can be worn by the bearer under or on top of clothing. These computers are useful for applications which require complex computational support. The main feature of these computers is its consistency.
Overview:
As there is a rapid growth in mobile sensing and computing has developed few methods for recognizing human activities with the help of sensor data. Wearable computers have the ability to perform multi-tasking. Wearable computers can attach to the clothes of a human or can directly attach to the body by inserting it into chains or wristbands. The user can wear this computer and can use its power and functionality virtually in their environment. These computers are ready to use for the user at any time and there is no need to turn on or off these devices. These computers help to augment the senses. Wearable computers provide a ubiquitous sensing and computational capabilities. Wearable system helps for the applications which require activity based computing or interaction-based computing.
Annotation or Labeling and Recognition or Spotting are few of the terms which commonly used in wearable activity recognition process. Sensors help to acquires user’s activities or gestures through signals but these Sensors must be small devices which are invisible to outside and should not provide discomfort to the users. These sensors should select based on its wearability, power usage and computational needs. Sensor modalities are body-worn accelerometers, which are very small and have low power.
Conclusions:
The large collection of human behaviors by Wearable computing helps to bootstrap humanoid robot behaviors. The main object of wearable computing is Human activity recognition.
There are many challenges by using activity-recognition systems in robotics. The recognition of human activities for HRI must consider that the activities recognized by machines are not perfectly accurate.
Download BE CSE Seminar Topic with Report for Wearable computing .