本文爲英國謝菲爾德大學（作者：Ross Walker）的博士論文，共282頁。
本文提出了一種新的路徑規劃算法，用於機器人在擁擠行人環境中導航。這個機器人的設計是通過周密的動作在人羣中穿行。與許多其他的路徑規劃算法（假設與其他行人協作）不同，該算法是完全獨立的，只需要進行觀察。本文以自主移動機器人爲研究對象，提出了一種綜合導航策略。利用預測行人運動的簡單方法，以及機器人與行人之間的簡單相對距離和軌跡測量，機器人可以在人羣中導航，而不會對行人軌跡造成干擾。利用不確定橢圓預測動態行人位置，提出了一種新的Voronoi圖可見性混合路線圖，使得路徑規劃器能夠充分利用行人之間的任何可用間隙，規劃出合理的路徑。體貼的路徑規劃器的目標是讓機器人在人羣中以特定的方式移動。通過預測行人的軌跡，機器人可以避免干擾他們。在橫穿行人的軌跡時，跟隨行人的偏好移動；在可能的情況下朝人羣的同一方向移動；在擁擠的區域減速，將防止對個別行人的任何干擾，並防止人羣擁擠程度的增加。利用仿真行人、多個移動機器人以及真實的行人數據集對該策略的有效性進行了評估，該算法將突出其在考慮上述因素的情況下向每個動態智能體移動的能力。
This thesis presents a novel path planningalgorithm for robotic crowd navigation through a pedestrian environment. Therobot is designed to negotiate its way through the crowd using consideratemovements. Unlike many other path planning algorithms, which assume cooperationwith other pedestrians, this algorithm is completely independent and requiresonly observation. A considerate navigation strategy has been developed in thisthesis, which utilises consideration as an directs an autonomous mobile robot.Using simple methods of predicting pedestrian movements, as well as simplerelative distance and trajectory measurements between the robot and pedestrians,the robot can navigate through a crowd without causing disruption to pedestriantrajectories. Dynamic pedestrian positions are predicted using uncertaintyellipses. A novel Voronoi diagram-visibility graph hybrid roadmap isimplemented so that the path planner can exploit any available gaps in betweenpedestrians, and plan considerate paths. The aim of the considerate pathplanner is to have the robot behave in specific ways when moving through thecrowd. By predicting pedestrian trajectories, the robot can avoid interferingwith them. Following preferences to move behind pedestrians, when cuttingacross their trajectories; to move in the same direction of the crowd whenpossible; and to slow down in crowded areas, will prevent any interference to individualpedestrians, as well as preventing an increase in congestion to the crowd as awhole. The effectiveness of the considerate navigation strategy is evaluatedusing simulated pedestrians, multiple mobile robots loaded with the pathplanning algorithm, as well as a real-life pedestrian dataset. The algorithmwill highlight its ability to move with the aforementioned considerationtowards each individual dynamic agent.
- 引言
- 文獻回顧
- 開發考慮周到的路徑規劃器
- 一種新的Voronoi可見性圖混合路線圖
- 考慮周到的導航策略算法
- 測試導航策略
- 基於真實行人數據的導航策略評價
- 總結與未來工作展望
更多精彩文章請關注公衆號：