具善意欺騙功能之視覺導航系統在老人居家照護之應用

根據內政部於今年所進行的人口年齡調查，指出臺灣地區65歲以上人口已占總人口的11.2%，而其中實際獨居比例為13.6%。世界衛生組織曾將獨居老人定為高危險群，原因在於當有緊急事故發生時無法即時的得到支援與幫助。因此，可及時後傳資訊並尋求協助的居家照護系統是有必要的，但居家照護系統一般須仰賴固定式攝影機及大量感應器以取得使用者的目前資訊，此種做法有冒犯使用者隱私權的問題。跌倒對老人的健康影響甚巨，隨著機器人技術的發展，本研究以Arduino機器人為平台，結合Asus Xtion的體感感測器來設計視覺導航系統。研究方法主要是以彩色及深度影像預測機器人的移動方向及距離，再藉由此移動資訊來規劃路徑。為了預防老人在居家環境不小心碰觸到傢俱等物品，造成跌倒的危險事件，本研究提出將善意欺騙方法融入視覺導航系統，當老人在居家環境行走時，若系統預測行進路徑將會碰撞到物體時，所提善意欺騙方法將定義一個危險程度函數，此函數是依據使用者與障礙物間的距離，及所使用之欺騙方法的有效性來設計其危險程度值。實驗結果顯示，在使用者接近障礙物時系統以發出聲響、螢幕閃爍等方法可以轉移注意力並使其遠離障礙。本研究所提之方法的確可以降低老人在居家環境碰觸到傢俱等物品之危險性。

關鍵字

善意欺騙方法；視覺里程計；機器人；居家照護

並列摘要

The elderly population in our nation significantly increased with the rise of average lifespan in the recent years. According to population aging statistics by Ministry of the Interior, 11.2% of total population in Taiwan is over 65-years old and 13.6% of them are solitary. World Health Organization (WHO) had defined solitary elderlies as high-risked, because help in time may be unavailable when emergency event occurs. To solve the problem, home care system is the most efficient way, but conventional home care systems highly rely on fixed surveillance equipment and sensors to acquire user information. The installed surveillance systems may invoke privacy issue. Fall prevention is a critical subject to elderly home care services. With the advent of new technologies in service robots, this study is based on Arduino platform that is integrated with Asus Xtion motion sensor to design a visual navigation system. Colors and depth of images are used to predict the walking direction and distance to the obstacles and to plan the route of service robot. To prevent the elderly from colliding with furniture in the house and cause any possible fall incidents, a graceful deception method is proposed in this paper. In case an incident is predicted to occur in the direction of walking route, the proposed system will calculate the dangerous degree from deceptive function. The dangerous degree is determined by the distance to the obstacle and the effectiveness of the deceptive methods. Simulation results verify that the proposed system is effective to reduce the danger of colliding with obstacles in the home care services for elderly.

並列關鍵字

graceful deception ； visual Odometry ； service robots ； home care services

參考文獻

[2] G. Verhenneman and A. Veys, “My camera, my buddy? legal and sociological assessment of the potential of video surveillance in eHomeCare,” in Proc. of ITAB Int. Conf. Information Technology and Applications in Biomedicine, Larnaca, Cyprus, pp.1-6, Nov. 2009.

[3] J. Campbell, R. Sukthankar, I. Nourbakhsh and A. Pahwa, “A robust visual odometry and precipice detection system using consumer-grade monocular vision,” in Proc. of IEEE Int. Conf. Robotics and Automation, Barcelona, Spain, pp.3421-3427, Apr. 2005.

[5] H. Bay, T. Turtelaars and L.V. Gol, “SURF: Speeded up robust features,” Computer Vision and Image Understanding (CVIU), vol. 110, no. 3, pp.346-359, June 2008.

[7] Joseph Khoury, Application to image compression. http://aix1.uottawa.ca/~jkhoury/haar.htm

[8] D.G. Lowe, “Object recognition from local scale-invariant features,” in Proc. of Int. Conf. on Computer Vision, Corfu, Greece, pp.1150-1157, Sep. 1999.

國際替代計量

具善意欺騙功能之視覺導航系統在老人居家照護之應用

全文下載

主題瀏覽