원문정보
초록
영어
Statistics have shown that falls are the leading cause of accidental deaths among people over the age of 75 and 55% of all falls take place in homes with most falls occurring in the bathroom. In this research, the aim of this project is to come up with a system that is able to detect falls in the washrooms, so that immediate medical aid can be provided so as to alleviate the aftereffects of the fall. Microsoft Visio was used to map the logic of the system which was then used in the creation of the rules for the system. Drools Rules is used to create a set of rules that gets input from the sensors detecting the motions of the patients, so as to infer the activities that are ongoing in order to check for errors. With input from the sensors, the set of rules should be able to determine the activity happening and possible errors that the input could imply. Simulations of real-li (Learn Not To Fall, 2009) scenarios were carried out twice. Errors and limitations of the system were noted during the simulation to further debug and improve the rules for a more robust and efficient system.
목차
1. INTRODUCTION
2. MATERIALS AND METHODS
2.1. Reasoning Engine
2.2. Environmental Set-up
2.3. Person
3. RESULTS
4. DISCUSSION
5. LIMITATIONS
5.1. Ultrasound Sensor and the inaccuracy of distances
5.2. The lack of a Bathing area
5.3. Sensitivity of the Passive Infrared Sensor
6. FURTHER IMPROVEMENTS/ PLANS FOR EXPERIMENTATION
7. CONCLUSION
8. MATERIALS AND METHODS
9. REFERENCES