機器人運動軌跡規劃與人類追蹤導航控制演算法

近年來，人類追蹤導航之相關研究日趨增加，越來越多的應用著重於如何使機器人與人類之間的互動更加人性化。如果機器人能賦有跟隨人類之能力，它將能在不同的環境下適當地給予人類不同的協助。在企圖實現人類追蹤導航之演算法時，追蹤目標與其他障礙物之間的重疊狀況，是一個十分常見的問題。本篇論文中將探討如何規劃機器人追蹤人類的運動軌跡，使得追蹤目標可見性能夠最大化。在人類追蹤系統下，透過雷射測距儀的使用來感測周遭的行人並使用卡曼濾波器來追蹤每個行人。隨後，再運用一個以DWA*做為理論平台的追蹤導航演算法來達到行人跟隨的目的。本論文將提出兩個不同人類追蹤導航演算法。第一個演算法是個較直觀的方法，它在導航系統裡設置一個虛擬的目標來改進機器人的運動軌跡。第二個演算法利用啟發式搜尋的概念來找尋跟隨行人的最佳路徑，此最佳路徑不只能使機器人在最短時間內到達目標，並在過程中保持目標可見性的最大化。實驗部分，首先測試機器人在利用不同追蹤導航演算法下追隨行人的機動力和靈活度表現; 並確保機器人能在一個預先設置好的環境下順利達到跟隨行人的目的。最後，再挑戰一些較複雜的真實環境場景，如餐廳和辦公室。結果顯示，當運用所提出的演算法時，即使在複雜性高的未設置環境下，機器人仍具有成功追隨目標行人的能力。

關鍵字

人類追蹤與追隨；動態窗格演算法；路徑規劃；動態物體偵測

並列摘要

In recent years, people tracking and following has become an increasingly popular research topic. More and more applications have focused on improving robots’ ability to interact with humans. By providing the robot the capability of following a target pedestrian in an appropriate manner, the robot can assist people in various ways under different environments. One of the main difficulties when performing people tracking and following is the occlusion problem caused by static and dynamic obstacles. In this thesis, the aim is to solve the occlusion problem by planning a robot trajectory which can maximize target visibility when following a moving target. Initially, a laser range finder is used to detect the human target and then track the target using Kalman Filter. Afterward, a pedestrian following algorithm is based on a look-ahead algorithm, DWA*, is implemented to pursue the target while avoiding any static or dynamic obstacles. In this thesis, two target following methods are proposed and compared. The first method is an intuitive approach, it sets a pseudo goal further away than the actual target location and use obstacle avoidance algorithm to reach the pseudo goal. The second method uses heuristic search to find the most optimal path: a trajectory that not only has the shortest distance to reach the goal but also able to maintain the maximum target visibility. The experiments were performed to evaluate robot maneuvers using different methods when following a target. The results compare and evaluate each following algorithm. Furthermore, the experiments were also performed in more complex environments such as cafeteria and office. The robot has proven that it is still capable of following the target when noises and other dynamic objects are present.

並列關鍵字

Human following ； DWA* ； path planning ； DATMO

參考文獻

[4] H. Koyasu, J. Miura, and Y. Shirai, “Integrating Multiple Scan Matching Results for Ego-Motion Estimation with Uncertainty,” Proceedings of the IEEE International Conference on Intelligent Robots and Systems, pp. 3104-3109, Sendai, Japan, Sept. 2004.

[7] D. Schulz, W. Burgard, D. Fox, and A. Cremers, “Tracking Multiple Moving Targets with a Mobile Robot using Particle Filters and Statistical Data Association,” Proceedings of IEEE International Conference on Robotics and Automation, pp.1665-1670, Seoul, Korea, Jul. 2001.

[12] L. Spinello, R. Triebel, and R. Siegwart, “Multiclass Multimodal Detection and Tracking in Urban Environments,” The International Journal of Robotics Research, Vol. 29, No. 12, pp. 1498-1515, Oct. 2010.

[14] C.-C Wang, C. Thorpe, and S. Thrun, “Online Simultaneous Localization and Mapping with Detection and Tracking of Moving Objects: Theory and Results from a Ground Vehicle in Crowded Urban Areas,” Proceedings of the IEEE International Conference on Robotics and Automation, pp. 842-849, Taipei, Taiwan, Sep. 2003.

[15] L. Montesano, J. Minguez, and L. Montano. “Modeling the Static and the Dynamic Parts of the Environment to Improve Sensor-based Navigation,” Proceedings of the IEEE International Conference on Robotics and Automation, pp. 4556-4562, Barcelona, Spain, Apr. 2005.

國際替代計量

機器人運動軌跡規劃與人類追蹤導航控制演算法

全文下載

主題瀏覽