基於雷射雷達之外形偵測用於多機器人跟隨 以及輪椅機器人的乘坐舒適性評估

Abstract

最近護理人員超時工作的新聞層出不窮，如何減輕護理人員平時工作的辛勞，以及提升病患的醫療品質，即成為十分重要的課題，基於此因，我們使護士平時的工作夥伴「護理車」有了自主跟隨的功能，減少護理人員推行傳統護理車的負擔，可以有多餘的體力去照護病患。在本論文中，我們提出多機器人的跟隨，結合輪椅機器人與護理車，可以同時讓這兩台機器人跟著護理人員行走，可方便護理人員照護乘坐在輪椅上的病患，減少護理人員不足的問題。在機器人跟隨人的部分，本論文整合影像感測器與雷射雷達各自的優點，利用腳步追蹤去補足影像追蹤的缺點，讓機器人可以更穩定跟隨在不同的場景環境中。在多機器人跟隨的部分，在不用事先分析前方機器人外觀型態的情況下，本論文提出使用雷射雷達配合疊代最近點(Iterative Closest Point, ICP)演算法去偵測前方機器人。另外本論文依照ISO 2631-1舒適度標準，對於輪椅機器人乘坐的舒適度進行評估，實驗乘坐者在不同路面的舒適感受， 最後提出屬於輪椅機器人的舒適度指標，僅此，以此論文展示本實驗的成果。

Recently, the news of over-working of nurses emerge in an endless stream. Reducing the work load on nurses to enhance the medical qualities on patients become a very import issues. In that case, we let the medical cart which nurses uses everyday have the function of auto tracking to eliminate the efforts to push the medical cart so that the nurses have much more stamina to take care of patients. In the thesis, we propose multi-robots following structure which integrate wheelchair robot with intelligent medical cart so that they can follow the nurses simultaneously. Nurses can take care of patients easily and reducing the impact of nurses shortage. First, in the robot following to humans, we integrate the merits of image sensor and laser range finder. Leg tracking by laser range finder can compensate the shortage of image tracking so that the robots can follow in different scenario steadily. Second, in the multi-robots following, we propose Iterative Closet Point (ICP) by laser range finder to detect front robot without analyzing the shape in advance. Finally, we according to the ISO 2631-1 comfort standard to assess the comfort levels on the wheelchair robot. By experimenting the comfort levels on different roads, we propose a comfort standard belong to wheelchair. The experimental results verified and demonstrated the performance of the proposed system.