발간년도 : [2022]
| 논문정보 |
|
| 논문명(한글) |
[Vol.17, No.1] A Study on the Application of Reinforcement Learning Algorithm to Improve Walking in Dementia Patients |
|
| 논문투고자 |
Jin-Soo Kim |
|
| 논문내용 |
This study was conducted to effectively solve the abnormal walking problem caused by the increase in dementia patients, which is a representative problem in an aging society. Walking in dementia patients occurs in abnormal walking patterns and can cause fatal diseases in various human structures, including the brain. Therefore, detailed gait analysis is required for dementia patients, and measures are needed to identify the cause of the problem and solve the problem. Therefore, in this study, a method of applying a reinforcement learning algorithm within the existing walking analysis system was studied as a method to improve the walking condition of dementia patients. The reinforcement learning algorithm calculates the range of joints and the amount of change in walking variables by walking state for the state and behavior that occur while the patient is walking through the gap between the normal range of joints. Accordingly, the system may automatically perform gait motion feedback of the patient. This feedback is expected to be able to learn normal walking patterns and optimize walking. Future research plans to find ways to collect big data related to walking from dementia patients with various symptoms and use it for customized gait treatment for each patient. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
